Datasets:

ChicagoHAI

/

CaseSumm

like 11

Follow

ChicagoHAI 19

578.US.2015_14-8349

Petitioner Timothy Foster was convicted of capital murder and sentenced to death in a Georgia court. During jury selection at his trial, the State used peremptory challenges to strike all four black prospective jurors qualified to serve on the jury. Foster argued that the State’s use of those strikes was racially motivated, in violation of Batson v. Kentucky, 476 U. S. 79 . The trial court rejected that claim, and the Georgia Supreme Court affirmed. Foster then renewed his Batson claim in a state habeas proceeding. While that proceeding was pending, Foster, through the Georgia Open Records Act, obtained from the State copies of the file used by the prosecution during his trial. Among other documents, the file contained (1) copies of the jury venire list on which the names of each black prospective juror were highlighted in bright green, with a legend indicating that the highlighting “represents Blacks”; (2) a draft affidavit from an investigator comparing black prospective jurors and concluding, “If it comes down to having to pick one of the black jurors, [this one] might be okay”; (3) notes identifying black prospective jurors as “B#1,” “B#2,” and “B#3”; (4) notes with “N” (for “no”) appearing next to the names of all black prospective jurors; (5) a list titled “[D]efinite NO’s” containing six names, including the names of all of the qualified black prospective jurors; (6) a document with notes on the Church of Christ that was annotated “NO. No Black Church”; and (7) the questionnaires filled out by five prospective black jurors, on which each juror’s response indicating his or her race had been circled. The state habeas court denied relief. It noted that Foster’s Batson claim had been adjudicated on direct appeal. Because Foster’s renewed Batson claim “fail[ed] to demonstrate purposeful discrimination,” the court concluded that he had failed to show “any change in the facts sufficient to overcome” the state law doctrine of res judicata. The Georgia Supreme Court denied Foster the Certificate of Probable Cause necessary to file an appeal. Held: 1. This Court has jurisdiction to review the judgment of the Georgia Supreme Court denying Foster a Certificate of Probable Cause on his Batson claim. Although this Court cannot ascertain the grounds for that unelaborated judgment, there is no indication that it rested on a state law ground that is both “independent of the merits” of Foster’s Batson claim and an “adequate basis” for that decision, so as to preclude jurisdiction. Harris v. Reed, 489 U. S. 255 . The state habeas court held that the state law doctrine of res judicata barred Foster’s claim only by examining the entire record and determining that Foster had not alleged a change in facts sufficient to overcome the bar. Based on this lengthy “Batson analysis,” the state habeas court concluded that Foster’s renewed Batson claim was “without merit.” Because the state court’s application of res judicata thus “depend[ed] on a federal constitutional ruling, [that] prong of the court’s holding is not independent of federal law, and [this Court’s] jurisdiction is not precluded.” Ake v. Oklahoma, 470 U. S. 68 ; see also Three Affiliated Tribes of Fort Berthold Reservation v. Wold Engineering, P. C., 467 U. S. 138 . Pp. 6–9. 2. The decision that Foster failed to show purposeful discrimination was clearly erroneous. Pp. 9–25. (a) Batson provides a three-step process for adjudicating claims such as Foster’s. “First, a defendant must make a prima facie showing that a preemptory challenge has been exercised on the basis of race; second, if that showing has been made, the prosecution must offer a race-neutral basis for striking the juror in question; and third, in light of the parties’ submissions, the trial court must determine whether the defendant has shown purposeful discrimination.” Snyder v. Louisiana, 552 U. S. 472 (internal quotation marks and brackets omitted). Only Batson’s third step is at issue here. That step turns on factual findings made by the lower courts, and this Court will defer to those findings unless they are clearly erroneous. See ibid. Pp. 9–10. (b) Foster established purposeful discrimination in the State’s strikes of two black prospective jurors: Marilyn Garrett and Eddie Hood. Though the trial court accepted the prosecution’s justifications for both strikes, the record belies much of the prosecution’s reasoning. Pp. 10–22. (i) The prosecution explained to the trial court that it made a last-minute decision to strike Garrett only after another juror, Shirley Powell, was excused for cause on the morning that the strikes were exercised. That explanation is flatly contradicted by evidence showing that Garrett’s name appeared on the prosecution’s list of “[D]efinite NO’s”—the six prospective jurors whom the prosecution was intent on striking from the outset. The record also refutes several of the reasons the prosecution gave for striking Garrett instead of Arlene Blackmon, a white prospective juror. For example, while the State told the trial court that it struck Garrett because the defense did not ask her for her thoughts about such pertinent trial issues as insanity, alcohol, or pre-trial publicity, the record reveals that the defense asked Garrett multiple questions on each topic. And though the State gave other facially reasonable justifications for striking Garrett, those are difficult to credit because of the State’s willingness to accept white jurors with the same characteristics. For example, the prosecution claims that it struck Garrett because she was divorced and, at age 34, too young, but three out of four divorced white prospective jurors and eight white prospective jurors under age 36 were allowed to serve. Pp. 11–17. (ii) With regard to prospective juror Hood, the record similarly undermines the justifications proffered by the State to the trial court for the strike. For example, the prosecution alleged in response to Foster’s pretrial Batson challenge that its only concern with Hood was the fact that his son was the same age as the defendant. But then, at a subsequent hearing, the State told the court that its chief concern was with Hood’s membership in the Church of Christ. In the end, neither of those reasons for striking Hood withstands scrutiny. As to the age of Hood’s son, the prosecution allowed white prospective jurors with sons of similar age to serve, including one who, in contrast to Hood, equivocated when asked whether Foster’s age would be a factor at sentencing. And as to Hood’s religion, the prosecution erroneously claimed that three white Church of Christ members were excused for cause because of their opposition to the death penalty, when in fact the record shows that those jurors were excused for reasons unrelated to their views on the death penalty. Moreover, a document acquired from the State’s file contains a handwritten note stating, “NO. NO Black Church,” while asserting that the Church of Christ does not take a stand on the death penalty. Other justifications for striking Hood fail to withstand scrutiny because no concerns were expressed with regard to similar white prospective jurors. Pp. 17–23. (c) Evidence that a prosecutor’s reasons for striking a black prospective juror apply equally to an otherwise similar nonblack prospective juror who is allowed to serve tends to suggest purposeful discrimination. Miller-El v. Dretke, 545 U. S. 231 . Such evidence is compelling with respect to Garrett and Hood and, along with the prosecution’s shifting explanations, misrepresentations of the record, and persistent focus on race, leads to the conclusion that the striking of those prospective jurors was “motivated in substantial part by discriminatory intent.” Snyder, 552 U. S., at 485. P. 23. (d) Because Batson was decided only months before Foster’s trial, the State asserts that the focus on black prospective jurors in the prosecution’s file was an effort to develop and maintain a detailed account should the prosecution need a defense against any suggestion that its reasons were pretextual. That argument, having never before been raised in the 30 years since Foster’s trial, “reeks of afterthought.” Miller-El, 545 U. S., at 246. And the focus on race in the prosecution’s file plainly demonstrates a concerted effort to keep black prospective jurors off the jury. Pp. 23–25. Reversed and remanded. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in the judgment. Thomas, J., filed a dissenting opinion.

Petitioner Timothy Foster was convicted of capital murder and sentenced to death in a Georgia court. During jury selection at his trial, the State exercised peremptory strikes against all four black prospective jurors qualified to serve. Foster argued that the State’s use of those strikes was racially motivated, in violation of our decision in Batson v. Kentucky, 476 U. S. 79 (1986) . The trial court and the Georgia Supreme Court rejected Foster’s Batson claim. Foster then sought a writ of habeas corpus from the Superior Court of Butts County, Georgia, renewing his Batson objection. That court denied relief, and the Georgia Supreme Court declined to issue the Certificate of Probable Cause necessary under Georgia law for Foster to pursue an appeal. We granted certiorari and now reverse. I On the morning of August 28, 1986, police found Queen Madge White dead on the floor of her home in Rome, Georgia. White, a 79-year-old widow, had been beaten, sexually assaulted, and strangled to death. Her home had been burglarized. Timothy Foster subsequently confessed to killing White, and White’s possessions were recovered from Foster’s home and from Foster’s two sisters. The State indicted Foster on charges of malice murder and burglary. He faced the death penalty. Foster v. State, 258 Ga. 736, 374 S. E. 2d 188 (1988). District Attorney Stephen Lanier and Assistant District Attorney Douglas Pullen represented the State at trial. Jury selection proceeded in two phases: removals for cause and peremptory strikes. In the first phase, each prospective juror completed a detailed questionnaire, which the prosecution and defense reviewed. The trial court then conducted a juror-by-juror voir dire of approximately 90 prospective jurors. Throughout this process, both parties had the opportunity to question the prospective jurors and lodge challenges for cause. This first phase whittled the list down to 42 “qualified” prospective jurors. Five were black. In the second phase, known as the “striking of the jury,” both parties had the opportunity to exercise peremptory strikes against the array of qualified jurors. Pursuant to state law, the prosecution had ten such strikes; Foster twenty. See Ga. Code Ann. §15–12–165 (1985). The process worked as follows: The clerk of the court called the qualified prospective jurors one by one, and the State had the option to exercise one of its peremptory strikes. If the State declined to strike a particular prospective juror, Foster then had the opportunity to do so. If neither party exercised a peremptory strike, the prospective juror was selected for service. This second phase continued until 12 jurors had been accepted. The morning the second phase began, Shirley Powell, one of the five qualified black prospective jurors, notified the court that she had just learned that one of her close friends was related to Foster. The court removed Powell for cause. That left four black prospective jurors: Eddie Hood, Evelyn Hardge, Mary Turner, and Marilyn Garrett. The striking of the jury then commenced. The State exercised nine of its ten allotted peremptory strikes, removing all four of the remaining black prospective jurors. Foster immediately lodged a Batson challenge. The trial court rejected the objection and empaneled the jury. The jury convicted Foster and sentenced him to death. Following sentencing, Foster renewed his Batson claim in a motion for a new trial. After an evidentiary hearing, the trial court denied the motion. The Georgia Supreme Court affirmed, 258 Ga., at 747, 374 S. E. 2d, at 197, and we denied certiorari, Foster v. Georgia, 490 U. S. 1085 (1989) . Foster subsequently sought a writ of habeas corpus from the Superior Court of Butts County, Georgia, again pressing his Batson claim. While the state habeas proceeding was pending, Foster filed a series of requests under the Georgia Open Records Act, see Ga. Code Ann. §§50–18–70 to 50–18–77 (2002), seeking access to the State’s file from his 1987 trial. In response, the State disclosed documents related to the jury selection at that trial. Over the State’s objections, the state habeas court admitted those documents into evidence. They included the following: (1) Four copies of the jury venire list. On each copy, the names of the black prospective jurors were highlighted in bright green. A legend in the upper right corner of the lists indicated that the green highlighting “represents Blacks.” See, e.g., App. 253. The letter “B” also appeared next to each black prospective juror’s name. See, e.g., ibid. According to the testimony of Clayton Lundy, an investigator who assisted the prosecution during jury selection, these highlighted venire lists were circulated in the district attorney’s office during jury selection. That allowed “everybody in the office”—approximately “10 to 12 people,” including “[s]ecretaries, investigators, [and] district attorneys”—to look at them, share information, and contribute thoughts on whether the prosecution should strike a particular juror. Pl. Exh. 1, 2 Record 190, 219 (Lundy deposition) (hereinafter Tr.). The documents, Lundy testified, were returned to Lanier before jury selection. Id., at 220. (2) A draft of an affidavit that had been prepared by Lundy “at Lanier’s request” for submission to the state trial court in response to Foster’s motion for a new trial. Id., at 203. The typed draft detailed Lundy’s views on ten black prospective jurors, stating “[m]y evaluation of the jurors are a[s] follows.” App. 343. Under the name of one of those jurors, Lundy had written: “If it comes down to having to pick one of the black jurors, [this one] might be okay. This is solely my opinion. . . . Upon picking of the jury after listening to all of the jurors we had to pick, if we had to pick a black juror I recommend that [this juror] be one of the jurors.” Id., at 345 (paragraph break omitted). That text had been crossed out by hand; the version of the affidavit filed with the trial court did not contain the crossed-out language. See id., at 127–129. Lundy testified that he “guess[ed]” the redactions had been done by Lanier. Tr. 203. (3) Three handwritten notes on black prospective jurors Eddie Hood, Louise Wilson, and Corrie Hinds. Annotations denoted those individuals as “B#1,” “B#2,” and “B#3,” respectively. App. 295–297. Lundy testified that these were examples of the type of “notes that the team—the State would take down during voir dire to help select the jury in Mr. Foster’s case.” Tr. 208–210. (4) A typed list of the qualified jurors remaining after voir dire. App. 287–290. It included “Ns” next to ten jurors’ names, which Lundy told the state habeas court “signif[ied] the ten jurors that the State had strikes for during jury selection.” Tr. 211. Such an “N” appeared alongside the names of all five qualified black prospective jurors. See App. 287–290. The file also included a handwritten version of the same list, with the same markings. Id., at 299–300; see Tr. 212. Lundy testified that he was unsure who had prepared or marked the two lists. (5) A handwritten document titled “definite NO’s,” listing six names. The first five were those of the five qualified black prospective jurors. App. 301. The State concedes that either Lanier or Pullen compiled the list, which Lundy testified was “used for preparation in jury selection.” Tr. 215; Tr. of Oral Arg. 45. (6) A handwritten document titled “Church of Christ.” A notation on the document read: “NO. No Black Church.” App. 302. (7) The questionnaires that had been completed by several of the black prospective jurors. On each one, the juror’s response indicating his or her race had been circled. Id., at 311, 317, 323, 329, 334. In response to the admission of this evidence, the State introduced short affidavits from Lanier and Pullen. Lanier’s affidavit stated: “I did not make any of the highlighted marks on the jury venire list. It was common practice in the office to highlight in yellow those jurors who had prior case experience. I did not instruct anyone to make the green highlighted marks. I reaffirm my testimony made during the motion for new trial hearing as to how I used my peremptory jury strikes and the basis and reasons for those strikes.” Id., at 169 (paragraph numeral omitted). Pullen’s affidavit averred: “I did not make any of the highlighted marks on the jury venire list, and I did not instruct anyone else to make the highlighted marks. I did not rely on the highlighted jury venire list in making my decision on how to use my peremptory strikes.” Id., at 170–171 (paragraph numeral omitted). Neither affidavit provided further explanation of the documents, and neither Lanier nor Pullen testified in the habeas proceeding. After considering the evidence, the state habeas court denied relief. The court first stated that, “[a]s a preliminary matter,” Foster’s Batson claim was “not reviewable based on the doctrine of res judicata” because it had been “raised and litigated adversely to [Foster] on his direct appeal to the Georgia Supreme Court.” App. 175. The court nonetheless announced that it would “mak[e] findings of fact and conclusions of law” on that claim. Id., at 191. Based on what it referred to as a “Batson . . . analysis,” the court concluded that Foster’s “renewed Batson claim is without merit,” because he had “fail[ed] to demonstrate purposeful discrimination.” Id., at 192, 195, 196. The Georgia Supreme Court denied Foster the “Certificate of Probable Cause” necessary under state law for him to pursue an appeal, determining that his claim had no “arguable merit.” Id., at 246; see Ga. Code Ann. §9–14–52 (2014); Ga. Sup. Ct. Rule 36 (2014). We granted certiorari. 575 U. S. ___ (2015). II Before turning to the merits of Foster’s Batson claim, we address a threshold issue. Neither party contests our jurisdiction to review Foster’s claims, but we “have an independent obligation to determine whether subject-matter jurisdiction exists, even in the absence of a challenge from any party.” Arbaugh v. Y & H Corp., 546 U. S. 500, 514 (2006) . This Court lacks jurisdiction to entertain a federal claim on review of a state court judgment “if that judgment rests on a state law ground that is both ‘independent’ of the merits of the federal claim and an ‘adequate’ basis for the court’s decision.” Harris v. Reed, 489 U. S. 255, 260 (1989) . The state habeas court noted that Foster’s Batson claim was “not reviewable based on the doctrine of res judicata” under Georgia law. App. 175. The Georgia Supreme Court’s unelaborated order on review provides no reasoning for its decision.[1] That raises the question whether the Georgia Supreme Court’s order—the judgment from which Foster sought certiorari[2]—rests on an adequate and independent state law ground so as to preclude our jurisdiction over Foster’s federal claim. We conclude that it does not. When application of a state law bar “depends on a federal constitutional ruling, the state-law prong of the court’s holding is not independent of federal law, and our jurisdiction is not precluded.” Ake v. Oklahoma, 470 U. S. 68, 75 (1985) ; see also Three Affiliated Tribes of Fort Berthold Reservation v. Wold Engineering, P. C., 467 U. S. 138, 152 (1984) . In this case, the Georgia habeas court’s analysis in the section of its opinion labeled “Batson claim” proceeded as follows: “The [State] argues that this claim is not reviewable due to the doctrine of res judicata. However, because [Foster] claims that additional evidence allegedly supporting this ground was discovered subsequent to the Georgia Supreme Court’s ruling [on direct appeal], this court will review the Batson claim as to whether [Foster] has shown any change in the facts sufficient to overcome the res judicata bar.” App. 192. To determine whether Foster had alleged a sufficient “change in the facts,” the habeas court engaged in four pages of what it termed a “Batson . . . analysis,” in which it evaluated the original trial record and habeas record, including the newly uncovered prosecution file. Id., at 192–196. Ultimately, that court concluded that Foster’s “renewed Batson claim is without merit.” Id., at 196 (emphasis added). In light of the foregoing, it is apparent that the state habeas court’s application of res judicata to Foster’s Batson claim was not independent of the merits of his federal constitutional challenge.[3] That court’s invocation of res judicata therefore poses no impediment to our review of Foster’s Batson claim. See Ake, 470 U. S., at 75.[4] III A The “Constitution forbids striking even a single prospective juror for a discriminatory purpose.” Snyder v. Louisiana, 552 U. S. 472, 478 (2008) (internal quotation marks omitted). Our decision in Batson v. Kentucky, 476 U. S. 79 , provides a three-step process for determining when a strike is discriminatory: “First, a defendant must make a prima facie showing that a peremptory challenge has been exercised on the basis of race; second, if that showing has been made, the prosecution must offer a race-neutral basis for striking the juror in question; and third, in light of the parties’ submissions, the trial court must determine whether the defendant has shown purposeful discrimination.” Snyder, 552 U. S., at 476–477 (internal quotation marks and brackets omitted). Both parties agree that Foster has demonstrated a prima facie case, and that the prosecutors have offered race-neutral reasons for their strikes. We therefore address only Batson’s third step. That step turns on factual determinations, and, “in the absence of exceptional circumstances,” we defer to state court factual findings unless we conclude that they are clearly erroneous. Synder, 552 U. S., at 477. Before reviewing the factual record in this case, a brief word is in order regarding the contents of the prosecution’s file that Foster obtained through his Georgia Open Records Act requests. Pursuant to those requests, Foster received a “certif[ied] . . . true and correct copy of 103 pages of the State’s case file” from his 1987 trial. App. 247. The State argues that “because [Foster] did not call either of the prosecutors to the stand” to testify in his state habeas proceedings, “he can only speculate as to the meaning of various markings and writings” on thosepages, “the author of many of them, and whether the twoprosecutors at trial (District Attorney Lanier and Assistant District Attorney Pullen) even saw many of them.” Brief for Respondent 20. For these reasons, the State argues, “none of the specific pieces of new evidence [found in the file] shows an intent to discriminate.” Ibid. (capitalization omitted). For his part, Foster argues that “[t]here is no question that the prosecutors used the lists and notes, which came from the prosecution’s file and were certified as such,” and therefore the “source of the lists and notes, their timing, and their purpose is hardly ‘unknown’ or based on ‘conjecture.’ ” Reply Brief 4–5 (quoting Brief for Respondent 27–28). The State concedes that the prosecutors themselves authored some documents, see, e.g., Tr. of Oral Arg. 45 (admitting that one of the two prosecutors must have written the list titled “definite NO’s”), and Lundy’s testimony strongly suggests that the prosecutors viewed others, see, e.g., Tr. 220 (noting that the highlighted jury venire lists were returned to Lanier prior to jury selection). There are, however, genuine questions that remain about the provenance of other documents. Nothing in the record, for example, identifies the author of the notes that listed three black prospective jurors as “B#1,” “B#2,” and “B#3.” Such notes, then, are not necessarily attributable directly to the prosecutors themselves. The state habeas court was cognizant of those limitations, but nevertheless admitted the file into evidence, reserving “a determination as to what weight the Court is going to put on any of [them]” in light of the objections urged by the State. 1 Record 20. We agree with that approach. Despite questions about the background of particular notes, we cannot accept the State’s invitation to blind ourselves to their existence. We have “made it clear that in considering a Batson objection, or in reviewing a ruling claimed to be Batson error, all of the circumstances that bear upon the issue of racial animosity must be consulted.” Snyder, 552 U. S., at 478. As we have said in a related context, “[d]etermining whether invidious discriminatory purpose was a motivating factor demands a sensitive inquiry into such circumstantial . . . evidence of intent as may be available.” Arlington Heights v. Metropolitan Housing Development Corp., 429 U. S. 252, 266 (1977) . At a minimum, we are comfortable that all documents in the file were authored by someone in the district attorney’s office. Any uncertainties concerning the documents are pertinent only as potential limits on their probative value. B Foster centers his Batson claim on the strikes of two black prospective jurors, Marilyn Garrett and Eddie Hood. We turn first to Marilyn Garrett. According to Lanier, on the morning that the State was to use its strikes he had not yet made up his mind to remove Garrett. Rather, he decided to strike her only after learning that he would not need to use a strike on another black prospective juror, Shirley Powell, who was excused for cause that morning. Ultimately, Lanier did strike Garrett. In justifying that strike to the trial court, he articulated a laundry list of reasons. Specifically, Lanier objected to Garrett because she: (1) worked with disadvantaged youth in her job as a teacher’s aide; (2) kept looking at the ground during voir dire; (3) gave short and curt answers during voir dire; (4) appeared nervous; (5) was too young; (6) misrepresented her familiarity with the location of the crime; (7) failedto disclose that her cousin had been arrested on a drug charge; (8) was divorced; (9) had two children and two jobs; (10) was asked few questions by the defense; and (11) did not ask to be excused from jury service. See App. 55–57 (pretrial hearing); id., at 93–98, 105, 108, 110–112 (new trial hearing); Record in No. 45609 (Ga. 1988), pp. 439–440 (hereinafter Trial Record) (brief in opposition to new trial). The trial court accepted Lanier’s justifications, concluding that “[i]n the totality of circumstances,” there was “no discriminatory intent, and that there existed reasonably clear, specific, and legitimate reasons” for the strike. App. 143. On their face, Lanier’s justifications for the strike seem reasonable enough. Our independent examination of the record, however, reveals that much of the reasoning provided by Lanier has no grounding in fact. Lanier’s misrepresentations to the trial court began with an elaborate explanation of how he ultimately came to strike Garrett: “[T]he prosecution considered this juror [to have] the most potential to choose from out of the four remaining blacks in the 42 [member] panel venire. However, a system of events took place on the morning of jury selection that caused the excusal of this juror. The [S]tate had, in his jury notes, listed this juror as questionable. The four negative challenges were allocated for Hardge, Hood, Turner and Powell. . . . But on the morning of jury selection, Juror Powell was excused for cause with no objections by [d]efense counsel. She was replaced by Juror Cadle [who] was acceptable to the State. This left the State with an additional strike it had not anticipated or allocated. Consequently, the State had to choose between [white] Juror Blackmon or Juror Garrett, the only two questionable jurors the State had left on the list.” Trial Record 438–440 (brief in opposition to new trial) (emphasis added and citations omitted). Lanier then offered an extensive list of reasons for striking Garrett and explained that “[t]hese factors, with no reference to race, were considered by the prosecutor in this particular case to result in a juror less desirable from the prosecutor’s viewpoint than Juror Blackmon.” Id., at 441 (emphasis deleted). Lanier then compared Blackmon to Garrett. In contrast to Garrett, Juror Blackmon “was 46 years old, married 13 years to her husband who works at GE, buying her own home and [was recommended by a third party to] this prosecutor. She was no longer employed at Northwest Georgia Regional Hospital and she attended Catholic church on an irregular basis. She did not hesitate when answering the questions concerning the death penalty, had good eye contact with the prosecutor and gave good answers on the insanity issue. She was perceived by the prosecutor as having a stable home environment, of the right age and no association with any disadvantaged youth organizations.” Ibid. Lanier concluded that “the chances of [Blackmon] returning a death sentence were greater when all these factors were considered than Juror Garrett. Consequently, Juror Garrett was excused.” Ibid. The trial court accepted this explanation in denying Foster’s motion for a new trial. See App. 142–143. But the predicate for the State’s account—that Garrett was “listed” by the prosecution as “questionable,” making that strike a last-minute race-neutral decision—was false. During jury selection, the State went first. As a consequence, the defense could accept any prospective juror not struck by the State without any further opportunity for the State to use a strike against that prospective juror. Accordingly, the State had to “pretty well select the ten specific people [it] intend[ed] to strike” in advance. Id., at 83 (pretrial hearing); accord, ibid. (“[T]he ten people that we felt very uncomfortable with, we have to know up front.” (Lanier testimony)). The record evidence shows that Garrett was one of those “ten specific people.” That much is evident from the “definite NO’s” list in the prosecution’s file. Garrett’s name appeared on that list, which the State concedes was written by one of the prosecutors. Tr. of Oral Arg. 45. That list belies Lanier’s assertion that the State considered allowing Garrett to serve. The title of the list meant what it said: Garrett was a “definite NO.” App. 301 (emphasis added). The State from the outset was intent on ensuring that none of the jurors on that list would serve. The first five names on the “definite NO’s” list were Eddie Hood, Evelyn Hardge, Shirley Powell, Marilyn Garrett, and Mary Turner. All were black. The State struck each one except Powell (who, as discussed, was excused for cause at the last minute—though the prosecution informed the trial court that the “State was not, under any circumstances, going to take [Powell],” Trial Record 439 (brief in opposition to new trial)). Only in the number six position did a white prospective juror appear, and she had informed the court during voir dire that she could not “say positively” that she could impose the death penalty even if the evidence warranted it. 6 Tr. in No. 86–2218–2 (Super. Ct. Floyd Cty., Ga., 1987), p. 1152 (hereinafter Trial Transcript); see also id., at 1153–1158. In short, contrary to the prosecution’s submissions, the State’s resolve to strike Garrett was never in doubt. See also App. 290 (“N” appears next to Garrett’s name on juror list); id., at 300 (same). The State attempts to explain away the contradiction between the “definite NO’s” list and Lanier’s statements to the trial court as an example of a prosecutor merely “misspeak[ing].” Brief for Respondent 51. But this was not some off-the-cuff remark; it was an intricate story expounded by the prosecution in writing, laid out over three single-spaced pages in a brief filed with the trial court. Moreover, several of Lanier’s reasons for why he chose Garrett over Blackmon are similarly contradicted by the record. Lanier told the court, for example, that he struck Garrett because “the defense did not ask her questions about” pertinent trial issues such as her thoughts on “insanity” or “alcohol,” or “much questions on publicity.” App. 56 (pretrial hearing). But the trial transcripts reveal that the defense asked her several questions on all three topics. See 5 Trial Transcript 955–956 (two questions on insanity and one on mental illness); ibid. (four questions on alcohol); id., at 956–957 (five questions on publicity). Still other explanations given by the prosecution, while not explicitly contradicted by the record, are difficult to credit because the State willingly accepted white jurors with the same traits that supposedly rendered Garrett an unattractive juror. Lanier told the trial court that he struck Garrett because she was divorced. App. 56 (pre-trial hearing). But he declined to strike three out of the four prospective white jurors who were also divorced. SeeJuror Questionnaire in No. 86–2218–2 (Super. Ct. Floyd Cty., Ga., 1987) (hereinafter Juror Questionnaire), for Juror No. 23, p. 2 (juror Coultas, divorced); id., No. 33, p. 2 (juror Cochran, divorced); id., No. 107, p. 2 (juror Hatch, divorced); App. 23–24, 31 (State accepting jurors Coultas, Cochran, and Hatch). Additionally, Lanier claimed that he struck Garrett because she was too young, and the “State was looking for older jurors that would not easily identify with the defendant.” Trial Record 439; see App. 55 (pretrial hearing). Yet Garrett was 34, and the State declined to strike eight white prospective jurors under the age of 36. See Trial Record 439; Juror Questionnaire No. 4, p. 1; id., No. 10, p. 1; id., No. 23, p. 1; id., No. 48, p. 1; id., No. 70, p. 1; id., No. 71, p. 1; id., No. 92, p. 1; id., No. 106, p. 1; see App. 22–31. Two of those white jurors served on the jury; one of those two was only 21 years old. See id., at 35. Lanier also explained to the trial court that he struck Garrett because he “felt that she was less than truthful” in her answers in voir dire. Id., at 108 (new trial hearing). Specifically, the State pointed the trial court to the following exchange: “[Court]: Are you familiar with the neighborhood where [the victim] lived, North Rome? “[Garrett]: No.” 5 Trial Transcript 950–951. Lanier, in explaining the strike, told the trial court that in apparent contradiction to that exchange (which represented the only time that Garrett was asked about the topic during voir dire), he had “noted that [Garrett] attended Main High School, which is only two blocks from where [the victim] lived and certainly in the neighborhood. She denied any knowledge of the area.” Trial Record 439 (brief in opposition to new trial). We have no quarrel with the State’s general assertion that it “could not trust someone who gave materially untruthful answers on voir dire.” Foster, 258 Ga., at 739, 374 S. E. 2d, at 192. But even this otherwise legitimate reason is difficult to credit in light of the State’s acceptance of (white) juror Duncan. Duncan gave practically the same answer as Garrett did during voir dire: “[Court]: Are you familiar with the neighborhood in which [the victim] live[d]? “[Duncan]: No. I live in Atteiram Heights, but it’s not—I’m not familiar with up there, you know.” 5 Trial Transcript 959. But, as Lanier was aware, Duncan’s “residence [was] less than a half a mile from the murder scene” and her workplace was “located less than 250 yards” away. Trial Record 430 (brief in opposition to new trial). In sum, in evaluating the strike of Garrett, we are not faced with a single isolated misrepresentation. C We turn next to the strike of Hood. According to Lanier, Hood “was exactly what [the State] was looking for in terms of age, between forty and fifty, good employment and married.” App. 44 (pretrial hearing). The prosecution nonetheless struck Hood, giving eight reasons for doing so. Hood: (1) had a son who was the same age as the defendant and who had previously been convicted of a crime; (2) had a wife who worked in food service at the local mental health institution; (3) had experienced food poisoning during voir dire; (4) was slow in responding to death penalty questions; (5) was a member of the Church of Christ; (6) had a brother who counseled drug offenders; (7) was not asked enough questions by the defense during voir dire; and (8) asked to be excused from jury service. See id., at 44–47; id., at 86, 105, 110–111 (new trial hearing); Trial Record 433–435 (brief in opposition to new trial). An examination of the record, however, convinces us that many of these justifications cannot be credited. As an initial matter, the prosecution’s principal reasons for the strike shifted over time, suggesting that those reasons may be pretextual. In response to Foster’s pre-trial Batson challenge, District Attorney Lanier noted all eight reasons, but explained: “The only thing I was concerned about, and I will state it for the record. He has an eighteen year old son which is about the same age as the defendant. “In my experience prosecuting over twenty-five murder cases . . . individuals having the same son as [a] defendant who is charged with murder [have] serious reservations and are more sympathetic and lean toward that particular person. “It is ironic that his son, . . . Darrell Hood[,] has been sentenced . . . by the Court here, to theft by taking on April 4th, 1982. . . . [T]heft by taking is basi-cally the same thing that this defendant is chargedwith.” App. 44–45 (pretrial hearing; emphasis added). But by the time of Foster’s subsequent motion for a new trial, Lanier’s focus had shifted. He still noted the similarities between Hood’s son and Foster, see id., at 105 (new trial hearing), but that was no longer the key reason behind the strike. Lanier instead told the court that his paramount concern was Hood’s membership in the Church of Christ: “The Church of Christ people, while they may not take a formal stand against the death penalty, they are very, very reluctant to vote for the death penalty.” Id., at 84 (new trial hearing); accord, Trial Record 434–435 (“It is the opinion of this prosecutor that in a death penalty case, Church of Christ affiliates are reluctant to return a verdict of death.” (brief in opposition to new trial)). Hood’s religion, Lanier now explained, was the most important factor behind the strike: “I evaluated the whole Eddie Hood. . . . And the bottom line on Eddie Hood is the Church of Christ affiliation.” App. 110–111 (new trial hearing; emphasis added). Of course it is possible that Lanier simply misspoke in one of the two proceedings. But even if that were so, we would expect at least one of the two purportedly principal justifications for the strike to withstand closer scrutiny. Neither does. Take Hood’s son. If Darrell Hood’s age was the issue, why did the State accept (white) juror Billy Graves, who had a 17-year-old son? Juror Questionnaire No. 31, p. 3; see App. 24. And why did the State accept (white) juror Martha Duncan, even though she had a 20-year-old son? Juror Questionnaire No. 88, p. 3; see App. 30. The comparison between Hood and Graves is particu-larly salient. When the prosecution asked Hood if Foster’sage would be a factor for him in sentencing, he answered “None whatsoever.” Trial Transcript 280. Graves, on the other hand, answered the same question “probably so.” Id., at 446. Yet the State struck Hood and accepted Graves. The State responds that Duncan and Graves were not similar to Hood because Hood’s son had been convicted of theft, while Graves’s and Duncan’s sons had not. See Brief for Respondent 34–35; see also App. 135–136 (“While the defense asserts that the state used different standards for white jurors, insofar as many of them had children near the age of the Defendant, the Court believes that [Darrell Hood’s] conviction is a distinction that makes the difference.” (trial court opinion denying new trial)). Lanier had described Darrell Hood’s conviction to the trial court as being for “basically the same thing that this defendant is charged with.” Id., at 45 (pretrial hearing). Nonsense. Hood’s son had received a 12-month suspended sentence for stealing hubcaps from a car in a mall parking lot five years earlier. Trial Record 446. Foster was charged with capital murder of a 79-year-old widow after a brutal sexual assault. The “implausible” and “fantastic” assertion that the two had been charged with “basically the same thing” supports our conclusion that the focus on Hood’s son can only be regarded as pretextual. Miller-El v. Cockrell, 537 U. S. 322, 339 (2003) ; see also ibid. (“Credibility can be measured by, among other factors, . . . how reasonable, or how improbable, the [State’s] explanations are.”). The prosecution’s second principal justification for striking Hood—his affiliation with the Church of Christ, and that church’s alleged teachings on the death penalty—fares no better. Hood asserted no fewer than four times during voir dire that he could impose the death penalty.[5] A prosecutor is entitled to disbelieve a juror’s voir dire answers, of course. But the record persuades us that Hood’s race, and not his religious affiliation, was Lanier’s true motivation. The first indication to that effect is Lanier’s mischaracterization of the record. On multiple occasions, Lanier asserted to the trial court that three white prospective jurors who were members of the Church of Christ had been struck for cause due to their opposition to the death penalty. See App. 46 (“[Hood’s] religious preference is Church of Christ. There have been [three] other jurors that have been excused for cause by agreement that belong to the Church of Christ, Juror No. 35, 53, and 78.” (pretrial hearing)); id., at 114 (“Three out of four jurors who professed to be members of the Church of Christ, went off for [cause related to opposition to the death penalty].” (new trial hearing)); Trial Record 435 (“Church of Christ jurors Terry (#35), Green (#53), and Waters (#78) [were] excused for cause due to feeling[s] against the death penalty.” (brief in opposition to new trial)). That was not true. One of those prospective jurors was excused before even being questioned during voir dire because she was five-and-a-half months pregnant. 5 Trial Transcript 893. Another was excused by the agreement of both parties because her answers on the death penalty made it difficult to ascertain her precise views on capital punishment. See Brief for Respondent 39 (“[I]t was entirely unclear if [this juror] understood any of the trial court’squestions and her answers are equivocal at best.”). And the judge found cause to dismiss the third because she had already formed an opinion about Foster’s guilt. See 3 Trial Transcript 558 (“[Court]: And you have made up your mind already as to the guilt of the accused? A: Yes, sir. [Court]: I think that’s cause.”). The prosecution’s file fortifies our conclusion that any reliance on Hood’s religion was pretextual. The file contains a handwritten document titled “Church of Christ.” The document notes that the church “doesn’t take a stand on [the] Death Penalty,” and that the issue is “left for each individual member.” App. 302. The document then states: “NO. NO Black Church.” Ibid. The State tries to downplay the significance of this document by emphasizing that the document’s author is unknown. That uncertainty is pertinent. But we think the document is nonetheless entitled to significant weight, especially given that it is consistent with our serious doubts about the prosecution’s account of the strike. Many of the State’s secondary justifications similarly come undone when subjected to scrutiny. Lanier told the trial court that Hood “appeared to be confused and slow in responding to questions concerning his views on the death penalty.” Trial Record 434 (brief in opposition to new trial). As previously noted, however, Hood unequivocally voiced his willingness to impose the death penalty, and a white juror who showed similar confusion served on the jury. Compare 5 Trial Transcript 1100–1101 (white juror Huffman’s answers) with 2 id., at 269–278 (Hood’s answers); see App. 35. According to the record, such confusion was not uncommon. See id., at 138 (“The Court notes that [Hood’s] particular confusion about the death penalty questions was not unusual.”); accord, 5 Trial Transcript 994 (“[Court]: I think these questions should be reworded. I haven’t had a juror yet that understood what that meant.”); id., at 1101–1102 (“[Court]: I still say that these questions need changing overnight, because one out of a hundred jurors, I think is about all that’s gone along with knowing what [you’re asking].”). Lanier also stated that he struck Hood because Hood’s wife worked at Northwest Regional Hospital as a food services supervisor. App. 45 (pretrial hearing). That hospital, Lanier explained, “deals a lot with mentally disturbed, mentally ill people,” and so people associated with it tend “to be more sympathetic to the underdog.” Ibid. But Lanier expressed no such concerns about white juror Blackmon, who had worked at the same hospital. Blackmon, as noted, served on the jury. Lanier additionally stated that he struck Hood because the defense “didn’t ask [Hood] any question[s] about the age of the defendant,” “his feelings about criminal responsibility involved in insanity,” or “publicity.” Id., at 47. Yet again, the trial transcripts clearly indicate the contrary. See 2 Trial Transcript 280 (“Q: Is age a factor to you in trying to determine whether or not a defendant should receive a life sentence or a death sentence? A: None whatsoever.”); ibid. (“Q: Do you have any feeling about the insanity defense? A: Do I have any opinion about that? I have not formed any opinion about that.”); id., at 281 (“Q: Okay. The publicity that you have heard, has that pub-licity affected your ability to sit as a juror in this case and be fair and impartial to the defendant? A: No, it has no effect on me.”). D As we explained in Miller-El v. Dretke, “[i]f a prosecutor’s proffered reason for striking a black panelist applies just as well to an otherwise-similar nonblack [panelist] who is permitted to serve, that is evidence tending to prove purposeful discrimination.” 545 U. S. 231, 241 (2005) . With respect to both Garrett and Hood, such evidence is compelling. But that is not all. There are also the shifting explanations, the misrepresentations of the record, and the persistent focus on race in the prosecution’s file. Considering all of the circumstantial evidence that “bear[s] upon the issue of racial animosity,” we are left with the firm conviction that the strikes of Garrett and Hood were “motivated in substantial part by discriminatory intent.” Snyder, 552 U. S., at 478, 485.[6] IV Throughout all stages of this litigation, the State has strenuously objected that “race [was] not a factor” in its jury selection strategy. App. 41 (pretrial hearing); but see id., at 120 (Lanier testifying that the strikes were “based on many factors and not purely on race.” (emphasis added) (new trial hearing)). Indeed, at times the State has been downright indignant. See Trial Record 444 (“The Defenses’s [sic] misapplication of the law and erroneous distortion of the facts are an attempt to discredit the pro-secutor. . . . The State and this community demand an apology.” (brief in opposition to new trial)). The contents of the prosecution’s file, however, plainly belie the State’s claim that it exercised its strikes in a “color-blind” manner. App. 41, 60 (pretrial hearing). The sheer number of references to race in that file is arresting. The State, however, claims that things are not quite as bad as they seem. The focus on black prospective jurors, it contends, does not indicate any attempt to exclude them from the jury. It instead reflects an effort to ensure that the State was “thoughtful and non-discriminatory in [its] consideration of black prospective jurors [and] to develop and maintain detailed information on those prospective jurors in order to properly defend against any suggestion that decisions regarding [its] selections were pretextual.” Brief for Respondent 6. Batson, after all, had come down only months before Foster’s trial. The prosecutors, according to the State, were uncertain what sort of showing might be demanded of them and wanted to be prepared. This argument falls flat. To begin, it “reeks of afterthought,” Miller-El, 545 U. S., at 246, having never before been made in the nearly 30-year history of this litigation: not in the trial court, not in the state habeas court, and not even in the State’s brief in opposition to Foster’s petition for certiorari. In addition, the focus on race in the prosecution’s file plainly demonstrates a concerted effort to keep black prospective jurors off the jury. The State argues that it “was actively seeking a black juror.” Brief for Respondent 12; see also App. 99 (new trial hearing). But this claim is not credible. An “N” appeared next to each of the black prospective jurors’ names on the jury venire list. See, e.g., id., at 253. An “N” was also noted next to the name of each black prospective juror on the list of the 42 qualified prospective jurors; each of those names also appeared on the “definite NO’s” list. See id., 299–301. And a draft affidavit from the prosecution’s investigator stated his view that “[i]f it comes down to having to pick one of the black jurors, [Marilyn] Garrett, might be okay.” Id., at 345 (emphasis added); see also ibid. (recommending Garrett “if we had to pick a black juror” (emphasis added)). Such references are inconsistent with attempts to “actively see[k]” a black juror. The State’s new argument today does not dissuade us from the conclusion that its prosecutors were motivated in substantial part by race when they struck Garrett and Hood from the jury 30 years ago. Two peremptory strikes on the basis of race are two more than the Constitution allows. The order of the Georgia Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 The order stated, in its entirety: “Upon consideration of the Application for Certificate of Probable Cause to appeal the denial of habeas corpus, it is ordered that it be hereby denied. All the Justices concur, except Benham, J., who dissents.” App. 246. 2 We construe Foster’s petition for writ of certiorari as seeking review of the Georgia Supreme Court’s order denying him a “Certificate of Probable Cause.” App. 246. The Georgia Supreme Court Rules provide that such a certificate “will be issued where there is arguable merit.” Rule 36 (emphasis added); see also Hittson v. GDCP Warden, 759 F. 3d 1210, 1231–1232 (CA11 2014). A decision by the Georgia Supreme Court that Foster’s appeal had no “arguable merit” would seem to be a decision on the merits of his claim. In such circumstances the Georgia Supreme Court’s order is subject to review in this Court pursuant to a writ of certiorari under 28 U. S. C. §1257(a). R. J. Reynolds Tobacco Co. v. Durham County, 479 U. S. 130 –139 (1986); see Sears v. Upton, 561 U. S. 945 (2010) (per curiam) (exercising jurisdiction over order from Georgia Supreme Court denying a Certificate of Probable Cause). We reach the conclusion that such an order is a decision on the merits “in the absence of positive assurance to the contrary” from the Georgia Supreme Court. R. J. Reynolds, 479 U. S., at 138. 3 Contrary to the dissent’s assertion, see post, at 6–8, it is perfectly consistent with this Court’s past practices to review a lower court decision—in this case, that of the Georgia habeas court—in order to ascertain whether a federal question may be implicated in an unreasoned summary order from a higher court. See, e.g., R. J. Reynolds, 479 U. S., at 136–139 (exercising §1257 jurisdiction over unreasoned judgment by the North Carolina Supreme Court after examining grounds of decision posited by North Carolina Court of Appeal); see also Stephen M. Shapiro, Kenneth S. Geller, Timothy S. Bishop, Edward A. Hartnett, Dan Himmelfarb, Supreme Court Practice 211 (10th ed. 2013) (“[W]here the state court opinion fails to yield precise answers as to the grounds of decision, the Court may be forced to turn to other parts of the record, such as pleadings, motions, and trial court rulings, to determine if a federal claim is so central to the controversy as to preclude resting the judgment on independent and adequate state grounds.”). And even the dissent does not follow its own rule. It too goes beyond the unreasoned order of the Georgia Supreme Court in determining that the “likely explanation for the court’s denial of habeas relief is that Foster’s claim is procedurally barred.” Post, at 2. There would be no way to know this, of course, from the face of the Georgia Supreme Court’s summary order. 4 The concurrence notes that the “res judicata rule applied by the Superior Court in this case is quite different” from the state procedural bar at issue in Ake, which was “entirely dependent on federal law.” Post, at 8. But whether a state law determination is characterized as “entirely dependent on,” ibid., “resting primarily on,” Stewart v. Smith, 536 U. S. 856, 860 (2002) (per curiam), or “influenced by” a question of federal law, Three Affiliated Tribes of Fort Berthold Reservation v. Wold Engineering, P. C., 467 U. S. 138, 152 (1984) , the result is the same: the state law determination is not independent of federal law and thus poses no bar to our jurisdiction. 5 See 2 Trial Transcript 269 (“[Court]: Are you opposed to or against the death penalty? A: I am not opposed to it. Q: If the facts and circumstances warrant the death penalty, are you prepared to vote for the death penalty? A: Yes.”); id., at 270 (“[Court]: [A]re you prepared to vote for the death penalty? Now you said yes to that. A: All right. Q: Are you still saying yes? A: Uh-huh.”); id., at 274 (“[Court]: If the evidence warrants the death penalty, could you vote for the death penalty? A: Yes. I could vote for the death penalty.”); id., at 278 (“[Pullen]: And if the facts and circumstances warranted, you could vote to impose the death penalty? Yes.”). 6 In Snyder, we noted that we had not previously allowed the prosecution to show that “a discriminatory intent [that] was a substantial or motivating factor” behind a strike was nevertheless not “determinative” to the prosecution’s decision to exercise the strike. 552 U. S., at 485. The State does not raise such an argument here and so, as in Snyder, we need not decide the availability of such a defense.

578.US.2015_14-1175

Respondent Hyatt claims that he moved from California to Nevada in 1991, but petitioner Franchise Tax Board of California, a state agency, claims that he actually moved in 1992 and thus owes California millions in taxes, penalties, and interest. Hyatt filed suit in Nevada state court, which had jurisdiction over California under Nevada v. Hall, 440 U. S. 410 , seeking damages for California’s alleged abusive audit and investigation practices. After this Court affirmed the Nevada Supreme Court’s ruling that Nevada courts, as a matter of comity, would immunize California to the same extent that Nevada law would immunize its own agencies and officials, see Franchise Tax Bd. of Cal. v. Hyatt, 538 U. S. 488 , the case went to trial, where Hyatt was awarded almost $500 million in damages and fees. On appeal, California argued that the Constitution’s Full Faith and Credit Clause, Art. IV, §1, required Nevada to limit damages to $50,000, the maximum that Nevada law would permit in a similar suit against its own officials. The Nevada Supreme Court, however, affirmed $1 million of the award and ordered a retrial on another damages issue, stating that the $50,000 maximum would not apply on remand. Held: 1. The Court is equally divided on the question whether Nevada v. Hall should be overruled and thus affirms the Nevada courts’ exercise of jurisdiction over California’s state agency. P. 4. 2. The Constitution does not permit Nevada to apply a rule of Nevada law that awards damages against California that are greater than it could award against Nevada in similar circumstances. This conclusion is consistent with this Court’s precedents. A statute is a “public Act” within the meaning of the Full Faith and Credit Clause. While a State is not required “to substitute for its own statute . . . the statute of another State reflecting a conflicting and opposed policy,” Carroll v. Lanza, 349 U. S. 408 , a State’s decision to decline to apply another State’s statute on this ground must not embody a “policy of hostility to the public Acts” of that other State, id., at 413. Using this approach, the Court found no violation of the Clause in Carroll v. Lanza or in Franchise Tax Bd. the first time this litigation was considered. By contrast, the rule of unlimited damages applied here is not only “opposed” to California’s law of complete immunity; it is also inconsistent with the general principles of Nevada immunity law, which limit damages awards to $50,000. Nevada explained its departure from those general principles by describing California’s own system of controlling its agencies as an inadequate remedy for Nevada’s citizens. A State that disregards its own ordinary legal principles on this ground employs a constitutionally impermissible “ ‘policy of hostility to the public Acts’ of a sister State.” 538 U. S., at 499. The Nevada Supreme Court’s decision thereby lacks the “healthy regard for California’s sovereign status” that was the hallmark of its earlier decision. Ibid. This holding does not indicate a return to a complex “balancing-of-interests approach to conflicts of law under the Full Faith and Credit Clause.” Id., at 496. Rather, Nevada’s hostility toward California is clearly evident in its decision to devise a special, discriminatory damages rule that applies only to a sister State. Pp. 4–9. 130 Nev. ___, 335 P. 3d 125, vacated and remanded. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Alito, J., concurred in the judgment. Roberts, C. J., filed a dissenting opinion, in which Thomas, J., joined.

In Nevada v. Hall, 440 U. S. 410 (1979) , this Court held that one State (here, Nevada) can open the doors of its courts to a private citizen’s lawsuit against another State (here, California) without the other State’s consent. In this case, a private citizen, a resident of Nevada, has brought a suit in Nevada’s courts against the Franchise Tax Board of California, an agency of the State of California. The board has asked us to overrule Hall and hold that the Nevada courts lack jurisdiction to hear this lawsuit. The Court is equally divided on this question, and we consequently affirm the Nevada courts’ exercise of jurisdiction over California. See, e.g., Exxon Shipping Co. v. Baker, 554 U. S. 471, 484 (2008) (citing Durant v. Essex Co., 7 Wall. 107, 112 (1869)). California also asks us to reverse the Nevada court’s decision insofar as it awards the private citizen greater damages than Nevada law would permit a private citizen to obtain in a similar suit against Nevada’s own agencies. We agree that Nevada’s application of its damages law in this case reflects a special, and constitutionally forbidden, “ ‘policy of hostility to the public Acts’ of a sister State,” namely, California. U. S. Const., Art. IV, §1 (Full Faith and Credit Clause); Franchise Tax Bd. of Cal. v. Hyatt, 538 U. S. 488, 499 (2003) (quoting Carroll v. Lanza, 349 U. S. 408, 413 (1955) ). We set aside the Nevada Supreme Court’s decision accordingly. I Gilbert P. Hyatt, the respondent here, moved from California to Nevada in the early 1990’s. He says that he moved to Nevada in September 1991. California’s Franchise Tax Board, however, after an investigation and tax audit, claimed that Hyatt moved to Nevada later, in April 1992, and that he consequently owed California more than $10 million in taxes, associated penalties, and interest. Hyatt filed this lawsuit in Nevada state court against California’s Franchise Tax Board, a California state agency. Hyatt sought damages for what he considered the board’s abusive audit and investigation practices, including rifling through his private mail, combing through his garbage, and examining private activities at his place of worship. See App. 213–245, 267–268. California recognized that, under Hall, the Constitution permits Nevada’s courts to assert jurisdiction over California despite California’s lack of consent. California nonetheless asked the Nevada courts to dismiss the case on other constitutional grounds. California law, it pointed out, provided state agencies with immunity from lawsuits based upon actions taken during the course of collecting taxes. Cal. Govt. Code Ann. §860.2 (West 1995); see also §860.2 (West 2012). It argued that the Constitution’s Full Faith and Credit Clause required Nevada to apply California’s sovereign immunity law to Hyatt’s case. Nevada’s Supreme Court, however, rejected California’s claim. It held that Nevada’s courts, as a matter of comity, would immunize California where Nevada law would similarly immunize its own agencies and officials (e.g., for actions taken in the performance of a “discretionary” function), but they would not immunize California where Nevada law permitted actions against Nevada agencies, say, for acts taken in bad faith or for intentional torts. App. to Pet. for Cert. in Franchise Tax Bd. of Cal. v. Hyatt, O. T. 2002, No. 42, p. 12. We reviewed that decision, and we affirmed. Franchise Tax Bd., supra, at 499. On remand, the case went to trial. A jury found in Hyatt’s favor and awarded him close to $500 million in damages (both compensatory and punitive) and fees (including attorney’s fees). California appealed. It argued that the trial court had not properly followed the Nevada Supreme Court’s earlier decision. California explained that in a similar suit against similar Nevada officials, Nevada statutory law would limit damages to $50,000, and it argued that the Constitution’s Full Faith and Credit Clause required Nevada to limit damages similarly here. The Nevada Supreme Court accepted the premise that Nevada statutes would impose a $50,000 limit in a similar suit against its own officials. See 130 Nev. ___, ___, 335 P. 3d 125, 145–146 (2014); see also Nev. Rev. Stat. §41.035(1) (1995). But the court rejected California’s conclusion. Instead, while setting aside much of the damages award, it nonetheless affirmed $1 million of the award (earmarked as compensation for fraud), and it remanded for a retrial on the question of damages for intentional infliction of emotional distress. In doing so, it stated that “damages awarded on remand . . . are not subject to any statutory cap.” 130 Nev., at ___, 335 P. 3d, at 153. The Nevada Supreme Court explained its holding by stating that California’s efforts to control the actions of its own agencies were inadequate as applied to Nevada’s own citizens. Hence, Nevada’s “policy interest in providing adequate redress to Nevada’s citizens [wa]s paramount to providing [California] a statutory cap on damages under comity.” Id., at ___, 335 P. 3d, at 147. California petitioned for certiorari. We agreed to decide two questions. First, whether to overrule Hall. And, second, if we did not do so, whether the Constitution permits Nevada to award Hyatt damages against a California state agency that are greater than those that Nevada would award in a similar suit against its own stateagencies. II In light of our 4-to-4 affirmance of Nevada’s exercise of jurisdiction over California’s state agency, we must consider the second question: Whether the Constitution permits Nevada to award damages against California agencies under Nevada law that are greater than it could award against Nevada agencies in similar circumstances. We conclude that it does not. The Nevada Supreme Court has ignored both Nevada’s typical rules of immunity and California’s immunity-related statutes (insofar as California’s statutes would prohibit a monetary recovery that is greater in amount than the maximum recovery that Nevada law would permit in similar circumstances). Instead, it has applied a special rule of law that evinces a “ ‘policy of hostility’ ” toward California. Franchise Tax Bd., supra, at 499 (quoting Carroll v. Lanza, supra, at 413). Doing so violates the Constitution’s requirement that “Full Faith and Credit shall be given in each State to the public Acts, Records and judicial Proceedings of every other State.” Art. IV, §1. The Court’s precedents strongly support this conclusion. A statute is a “public Act” within the meaning of the Full Faith and Credit Clause. See, e.g., Carroll v. Lanza, supra, at 411; see also 28 U. S. C. §1738 (referring to “[t]he Acts of the legislature” in the full faith and credit context). We have said that the Clause “does not require a State to substitute for its own statute, applicable to persons and events within it, the statute of another State reflecting a conflicting and opposed policy.” Carroll v. Lanza, 349 U. S., at 412. But when affirming a State’s decision to decline to apply another State’s statute on this ground, we have consistently emphasized that the State had “not adopt[ed] any policy of hostility to the public Acts” of that other State. Id., at 413. In Carroll v. Lanza, the Court considered a negligence action brought by a Missouri worker in Arkansas’ courts. We held that the Arkansas courts need not apply a time limitation contained in Missouri’s (but not in Arkansas’) workman’s compensation law. Id., at 413–414. In doing so, we emphasized both that (1) Missouri law (compared with Arkansas law) embodied “a conflicting and opposed policy,” and (2) Arkansas law did not embody “any policy of hostility to the public Acts of Missouri.” Id., at 412–413. This second requirement was well established in earlier law. See, e.g., Broderick v. Rosner, 294 U. S. 629 –643 (1935) (New Jersey may not enforce a jurisdictional statute that would permit enforcement of certain claims under New Jersey law but “deny the enforcement” of similar, valid claims under New York law); Hughes v. Fetter, 341 U. S. 609 –612 (1951) (invalidating a Wisconsin statute that “close[d] the doors of its courts” to an Illinois cause of action while permitting adjudication of similar Wisconsin claims). We followed this same approach when we considered the litigation now before us for the first time. See Franchise Tax Bd., 538 U. S., at 498–499. Nevada had permitted Hyatt to sue California in Nevada courts. See id., at 497 (citing Hall, 440 U. S., at 414–421). Nevada’s courts recognized that California’s law of complete immunity would prevent any recovery in this case. The Nevada Supreme Court consequently did not apply California law. It applied Nevada law instead. We upheld that decision as consistent with the Full Faith and Credit Clause. But in doing so, we emphasized both that (1) the Clause does not require one State to apply another State’s law that violates its “own legitimate public policy,” Franchise Tax Bd., supra, at 497–498 (citing Hall, supra, at 424), and (2) Nevada’s choice of law did not “exhibi[t] a ‘policy of hostility to the public Acts’ of a sister State.” Franchise Tax Bd., supra, at 499 (quoting Carroll v. Lanza, supra, at 413). Rather, Nevada had evinced “a healthy regard for California’s sovereign status,” we said, by “relying on the contours of Nevada’s own sovereign immunity from suit as a benchmark for its analysis.” Franchise Tax Bd., supra, at 499. The Nevada decision before us embodies a critical departure from its earlier approach. Nevada has not applied the principles of Nevada law ordinarily applicable to suits against Nevada’s own agencies. Rather, it has applied a special rule of law applicable only in lawsuits against its sister States, such as California. With respect to damages awards greater than $50,000, the ordinary principles of Nevada law do not “conflic[t]” with California law, for both laws would grant immunity. Carroll v. Lanza, 349 U. S., at 412. Similarly, in respect to such amounts, the “polic[ies]” underlying California law and Nevada’s usual approach are not “opposed”; they are consistent. Id., at 412–413. But that is not so in respect to Nevada’s special rule. That rule, allowing damages awards greater than $50,000, is not only “opposed” to California law, ibid.; it is also inconsistent with the general principles of Nevada immunity law, see Franchise Tax Bd., supra, at 499. The Nevada Supreme Court explained its departure from those general principles by describing California’s system of controlling its own agencies as failing to provide “adequate” recourse to Nevada’s citizens. 130 Nev., at ___, 335 P. 3d, at 147. It expressed concerns about the fact that California’s agencies “ ‘operat[e] outside’ ” the systems of “ ‘legislative control, administrative oversight, and public accountability’ ” that Nevada applies to its own agencies. Ibid. (quoting Faulkner v. University of Tenn., 627 So. 2d 362 (Ala. 1992)). Such an explanation, which amounts to little more than a conclusory statement disparaging California’s own legislative, judicial, and administrative controls, cannot justify the application of a special and discriminatory rule. Rather, viewed through a full faith and credit lens, a State that disregards its own ordinary legal principles on this ground is hostile to another State. A constitutional rule that would permit this kind of discriminatory hostility is likely to cause chaotic interference by some States into the internal, legislative affairs of others. Imagine, for example, that many or all States enacted such discriminatory, special laws, and justified them on the sole basis that (in their view) a sister State’s law provided inadequate protection to their citizens. Would each affected sister State have to change its own laws? Entirely? Piece-by-piece, in order to respond to the new special laws enacted by every other State? It is difficult to reconcile such a system of special and discriminatory rules with the Constitution’s vision of 50 individual and equally dignified States. In light of the “constitutional equality” among the States, Coyle v. Smith, 221 U. S. 559, 580 (1911) , Nevada has not offered “sufficient policy considerations” to justify the application of a special rule of Nevada law that discriminates against its sister States, Carroll v. Lanza, supra, at 413. In our view, Nevada’s rule lacks the “healthy regard for California’s sovereign status” that was the hallmark of its earlier decision, and it reflects a constitutionally impermissible “ ‘policy of hostility to the public Acts’ of a sister State.” Franchise Tax Bd., supra, at 499 (quoting Carroll v. Lanza, supra, at 413). In so holding we need not, and do not, intend to return to a complex “balancing-of-interests approach to conflicts of law under the Full Faith and Credit Clause.” Franchise Tax Bd., 538 U. S., at 496. Long ago this Court’s efforts to apply that kind of analysis led to results that seemed to differ depending, for example, upon whether the case involved commercial law, a shareholders’ action, insurance claims, or workman’s compensation statutes. See, e.g., Bradford Elec. Light Co. v. Clapper, 286 U. S. 145 –159 (1932); Carroll v. Lanza, supra, at 414–420 (Frankfurter, J., dissenting) (listing, and trying to classify, nearly 50 cases). We have since abandoned that approach, and we continue to recognize that a State need not “ ‘substitute the statutes of other states for its own statutes dealing with a subject matter concerning which it is competent to legislate.’ ” Franchise Tax Bd., supra, at 496 (quoting Pacific Employers Ins. Co. v. Industrial Accident Comm’n, 306 U. S. 493, 501 (1939) ). But here, we can safely conclude that, in devising a special—and hostile—rule for California, Nevada has not “sensitively applied principles of comity with a healthy regard for California’s sovereign status.” Franchise Tax Bd., supra, at 499; see Thomas v. Washington Gas Light Co., 448 U. S. 261, 272 (1980) (plurality opinion) (Clause seeks to prevent “parochial entrenchment on the interests of other States”); Allstate Ins. Co. v. Hague, 449 U. S. 302 , and n. 10 (1981) (Stevens, J., concurring in judgment) (Clause is properly brought to bear when a State’s choice of law “threatens the federal interest in national unity by unjustifiably infringing upon the legitimate interests of another State”); cf. Supreme Court of N. H. v. Piper, 470 U. S. 274, 288 (1985) (Privileges and Immunities Clause prevents the New Hampshire Supreme Court from promulgating a rule that limits bar admission to state residents, discriminating against out-of-state lawyers); Bendix Autolite Corp. v. Midwesco Enterprises, Inc., 486 U. S. 888, 894 (1988) (Commerce Clause invalidates a statute of limitations that “imposes a greater burden on out-of-state companies than it does on [in-state] companies”). For these reasons, insofar as the Nevada Supreme Court has declined to apply California law in favor of a special rule of Nevada law that is hostile to its sister States, we find its decision unconstitutional. We vacate its judgment and remand the case for further proceedings not inconsistent with this opinion. It is so ordered. Justice Alito concurs in the judgment.

577.US.2015_14-181

Vermont law requires certain entities, including health insurers, to report payments relating to health care claims and other information relating to health care services to a state agency for compilation in an all-inclusive health care database. Respondent Liberty Mutual Insurance Company’s health plan (Plan), which provides benefits in all 50 States, is an “employee welfare benefit plan” under the Employee Retirement Income Security Act of 1974 (ERISA). The Plan’s third-party administrator, Blue Cross Blue Shield of Massachusetts, Inc. (Blue Cross), which is subject to Vermont’s disclosure statute, was ordered to transmit its files on eligibility, medical claims, and pharmacy claims for the Plan’s Vermont members. Respondent, concerned that the disclosure of such confidential information might violate its fiduciary duties, instructed Blue Cross not to comply and filed suit, seeking a declaration that ERISA pre-empts application of Vermont’s statute and regulation to the Plan and an injunction prohibiting Vermont from trying to acquire data about the Plan or its members. The District Court granted summary judgment to Vermont, but the Second Circuit reversed, concluding that Vermont’s reporting scheme is pre-empted by ERISA. Held: ERISA pre-empts Vermont’s statute as applied to ERISA plans. Pp. 5–13. (a) ERISA expressly pre-empts “any and all State laws insofar as they may now or hereafter relate to any employee benefit plan.” 29 U. S. C. §1144(a). As relevant here, the clause pre-empts a state law that has an impermissible “connection with” ERISA plans, i.e., a law that governs, or interferes with the uniformity of, plan administration. Egelhoff v. Egelhoff, 532 U. S. 141 . Pp. 5–6. (b) The considerations relevant to the determination whether an impermissible connection exists—ERISA’s objectives “as a guide to the scope of the state law that Congress understood would survive,” New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645 , and “the nature of” the state law’s “effect . . . on ERISA plans,” California Div. of Labor Standards Enforcement v. Dillingham Constr., N. A., Inc., 519 U. S. 316 —lead to the conclusion that Vermont’s regime, as applied to ERISA plans, is pre-empted. Pp. 6–12. (1) ERISA seeks to make the benefits promised by an employer more secure by mandating certain oversight systems and other standard procedures, Travelers, 514 U. S., at 651, and those systems and procedures are intended to be uniform, id., at 656. ERISA’s extensive reporting, disclosure, and recordkeeping requirements are central to, and an essential part of, this uniform plan administration system. Vermont’s law and regulation, however, also govern plan reporting, disclosure, and recordkeeping. Pre-emption is necessary in order to prevent multiple jurisdictions from imposing differing, or even parallel, regulations, creating wasteful administrative costs and threatening to subject plans to wide-ranging liability. ERISA’s uniform rule design also makes clear that it is the Secretary of Labor, not the separate States, that is authorized to decide whether to exempt plans from ERISA reporting requirements or to require ERISA plans to report data such as that sought by Vermont. Pp. 7–10. (2) Vermont’s counterarguments are unpersuasive. Vermont argues that respondent has not shown that the State scheme has caused it to suffer economic costs, but respondent need not wait to bring its pre-emption claim until confronted with numerous inconsistent obligations and encumbered with any ensuing costs. In addition, the fact that ERISA and the state reporting scheme have different objectives does not transform Vermont’s direct regulation of a fundamental ERISA function into an innocuous and peripheral set of additional rules. Vermont’s regime also cannot be saved by invoking the State’s traditional power to regulate in the area of public health. Pp. 10–12. (c) ERISA’s pre-existing reporting, disclosure, and recordkeeping provisions maintain their pre-emptive force regardless of whether the new Patient Protection and Affordable Care Act’s reporting obligations also pre-empt state law. Pp. 12–13. 746 F. 3d 497, affirmed. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Thomas, Breyer, Alito, and Kagan, JJ., joined. Thomas, J., and Breyer, J., filed concurring opinions. Ginsburg, J., filed a dissenting opinion, in which Sotomayor, J., joined.

This case presents a challenge to the applicability of a state law requiring disclosure of payments relating to health care claims and other information relating to health care services. Vermont enacted the statute so it could maintain an all-inclusive health care database.Vt. Stat. Ann., Tit. 18, §9410(a)(1) (2015 Cum. Supp.) ( V. S. A.). The state law, by its terms, applies to health plans established by employers and regulated by the Employee Retirement Income Security Act of 1974 (ERISA), 88Stat. 829, as amended, 29 U. S. C. §1001 et seq. The question before the Court is whether ERISA pre-empts the Vermont statute as it applies to ERISA plans. I A Vermont requires certain public and private entities that provide and pay for health care services to report information to a state agency. The reported information is compiled into a database reflecting “all health care utilization, costs, and resources in [ Vermont], and health care utilization and costs for services provided to Vermont residents in another state.” 18 V. S. A. §9410(b). A database of this kind is sometimes called an all-payer claims database, for it requires submission of data from all health insurers and other entities that pay for health care services. Almost 20 States have or are implementing similar databases. See Brief for State of New York et al. as Amici Curiae 1, and n. 1. Vermont’s law requires health insurers, health care providers, health care facilities, and governmental agencies to report any “information relating to health care costs, prices, quality, utilization, or resources required” by the state agency, including data relating to health insurance claims and enrollment. §9410(c)(3). Health insurers must submit claims data on members, subscribers, and policyholders. §9410(h). The Vermont law defines health insurer to include a “self-insured . . . health care benefit plan,” §9402(8), as well as “any third party administrator” and any “similar entity with claims data, eligibility data, provider files, and other information relating to health care provided to a Vermont resident.” §9410( j)(1)(B). The database must be made “available as a resource for insurers, employers, providers, purchasers of health care, and State agencies to continuously review health care utili-zation, expenditures, and performance in Vermont.” §9410(h)(3)(B). Vermont law leaves to a state agency the responsibility to “establish the types of information to be filed under this section, and the time and place and the manner in which such information shall be filed.” §9410(d). The law has been implemented by a regulation creating the Vermont Healthcare Claims Uniform Reporting and Evaluation System. The regulation requires the submission of “medical claims data, pharmacy claims data, member eligibility data, provider data, and other information,” Reg. H–2008–01, Code of Vt. Rules 21–040–021, §4(D) (2016) (CVR),in accordance with specific formatting, coding, and other requirements, §5. Under the regulation, health insurers must report data about the health care services provided to Vermonters regardless of whether they are treated in Vermont or out-of-state and about non-Vermonters who are treated in Vermont. §4(D); see also §1. The agency at present does not collect data on denied claims, §5(A)(8), but the statute would allow it to do so. Covered entities (reporters) must register with the State and must submit data monthly, quarterly, or annually, depending on the number of individuals that an entity serves. The more people served, the more frequently the reports must be filed. §§4, 6(I). Entities with fewer than 200 members need not report at all, ibid., and are termed “voluntary” reporters as distinct from “mandated” reporters, §3. Reporters can be fined for not complying with the statute or the regulation. §10; 18 V. S. A. §9410(g). B Respondent Liberty Mutual Insurance Company maintains a health plan (Plan) that provides benefits in all 50 States to over 80,000 individuals, comprising respondent’s employees, their families, and former employees. The Plan is self-insured and self-funded, which means that Plan benefits are paid by respondent. The Plan, which qualifies as an “employee welfare benefit plan” under ERISA, 29 U. S. C. §1002(1), is subject to “ERISA’s comprehensive regulation,” New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645, 650 (1995) . Respondent, as the Plan sponsor, is both a fiduciary and plan administrator. The Plan uses Blue Cross Blue Shield of Massachusetts, Inc. (Blue Cross) as a third-party administrator. Blue Cross manages the “processing, review, and payment” of claims for respondent. Liberty Mut. Ins. Co. v. Donegan, 746 F. 3d 497, 502 (CA2 2014) (case below). In its contract with Blue Cross, respondent agreed to “hold [Blue Cross] harmless for any charges, including legal fees, judgments, administrative expenses and benefit payment requirements, . . . arising from or in connection with [the Plan] or due to [respondent’s] failure to comply with any laws or regulations.” App. 82. The Plan is a voluntary reporter under the Vermont regulation because it covers some 137 Vermonters, which is fewer than the 200-person cutoff for mandated reporting. Blue Cross, however, serves several thousand Vermonters, and so it is a mandated reporter. Blue Cross, therefore, must report the information it possesses about the Plan’s members in Vermont. In August 2011, Vermont issued a subpoena ordering Blue Cross to transmit to a state-appointed contractor all the files it possessed on member eligibility, medical claims, and pharmacy claims for Vermont members. Id., at 33. (For clarity, the Court uses “Vermont” to refer not only to the State but also to state officials acting in their official capacity.) The penalty for noncompliance, Vermont threatened, would be a fine of up to $2,000 a day and a suspension of Blue Cross’ authorization to operate in Vermont for as long as six months. Id., at 31. Respondent, concerned in part that the disclosure of confidential information regarding its members might violate its fiduciary duties under the Plan, instructed Blue Cross not to comply. Respondent then filed this action in the United States District Court for the District of Vermont. It sought a declaration that ERISA pre-empts application of Vermont’s statute and regulation to the Plan and an injunction forbidding Vermont from trying to acquire data about the Plan or its members. Vermont filed a motion to dismiss, which the District Court treated as one for summary judgment, see Fed. Rule Civ. Proc. 12(d), and respondent filed a cross-motion for summary judgment. The District Court granted summary judgment to Vermont. It first held that respondent, despite being a mere voluntary reporter, had standing to sue because it was faced with either allegedly violating its “fiduciary and administrative responsibilities to the Plan” or assuming liability for Blue Cross’ withholding of the data from Vermont. Liberty Mut. Ins. Co. v. Kimbell, No. 2:11–cv–204 (D Vt., Nov. 9, 2012), p. 12. The District Court then concluded that the State’s reporting scheme was not pre-empted. Although that scheme “may have some indirect effect on health benefit plans,” the court reasoned that the “effect is so peripheral that the regulation cannot be considered an attempt to interfere with the administration or structure of a welfare benefit plan.” Id., at 31–32. The Court of Appeals for the Second Circuit reversed. The panel was unanimous in concluding that respondent had standing, but it divided on the merits of the pre-emption challenge. The panel majority explained that “one of ERISA’s core functions—reporting—[cannot] be laden with burdens, subject to incompatible, multiple and variable demands, and freighted with risk of fines, breach of duty, and legal expense.” 746 F. 3d, at 510. The Vermont regime, the court held, does just that. Id., at 508–510. This Court granted certiorari to address the important issue of ERISA pre-emption. 576 U. S. ___ (2015). II The text of ERISA’s express pre-emption clause is the necessary starting point. It is terse but comprehensive. ERISA pre-empts “any and all State laws insofar as they may now or hereafter relate to any employee benefit plan.” 29 U. S. C. §1144(a). The Court has addressed the potential reach of this clause before. In Travelers, the Court observed that “[i]f ‘relate to’ were taken to extend to the furthest stretch of its indeterminacy, then for all practical purposes pre-emption would never run its course.” 514 U. S., at 655. That is a result “no sensible person could have intended.” California Div. of Labor Standards Enforcement v. Dillingham Constr., N. A., Inc., 519 U. S. 316, 336 (1997) (Scalia, J., concurring). So the need for workable standards has led the Court to reject “uncritical literalism” in applying the clause. Travelers, 514 U. S., at 656. Implementing these principles, the Court’s case law to date has described two categories of state laws that ERISA pre-empts. First, ERISA pre-empts a state law if it has a “ ‘reference to’ ” ERISA plans. Ibid. To be more precise, “[w]here a State’s law acts immediately and exclusively upon ERISA plans . . . or where the existence of ERISA plans is essential to the law’s operation . . . , that ‘reference’ will result in pre-emption.” Dillingham, supra, at 325. Second, ERISA pre-empts a state law that has an impermissible “connection with” ERISA plans, meaning a state law that “governs . . . a central matter of plan administration” or “interferes with nationally uniform plan administration.” Egelhoff v. Egelhoff, 532 U. S. 141, 148 (2001) . A state law also might have an impermissible connection with ERISA plans if “acute, albeit indirect, economic effects” of the state law “force an ERISA plan to adopt a certain scheme of substantive coverage or effectively restrict its choice of insurers.” Travelers, supra, at 668. When considered together, these formulations ensure that ERISA’s express pre-emption clause receives the broad scope Congress intended while avoiding the clause’s susceptibility to limitless application. III Respondent contends that Vermont’s law falls in the second category of state laws that are pre-empted by ERISA: laws that govern, or interfere with the uniformity of, plan administration and so have an impermissible “ ‘connection with’ ” ERISA plans. Egelhoff, supra, at 148; Travelers, 514 U. S., at 656. When presented with these contentions in earlier cases, the Court has considered “the objectives of the ERISA statute as a guide to the scope of the state law that Congress understood would survive,” ibid., and “the nature of the effect of the state law on ERISA plans,” Dillingham, supra, at 325. Here, those considerations lead the Court to conclude that Vermont’s regime, as applied to ERISA plans, is pre-empted. A ERISA does not guarantee substantive benefits. The statute, instead, seeks to make the benefits promised by an employer more secure by mandating certain oversight systems and other standard procedures. Travelers, 514 U. S., at 651. Those systems and procedures are intended to be uniform. Id., at 656 (ERISA’s pre-emption clause “indicates Congress’s intent to establish the regulation of employee welfare benefit plans ‘as exclusively a federal concern’ ” (quoting Alessi v. Raybestos-Manhattan, Inc., 451 U. S. 504, 523 (1981) )). “Requiring ERISA administrators to master the relevant laws of 50 States and to contend with litigation would undermine the congressional goal of ‘minimiz[ing] the administrative and financial burden[s]’ on plan administrators—burdens ultimately borne by the beneficiaries.” Egelhoff, supra, at 149–150 (quoting Ingersoll-Rand Co. v. McClendon, 498 U. S. 133, 142 (1990) ); see also Fort Halifax Packing Co. v. Coyne, 482 U. S. 1, 9 (1987) . ERISA’s reporting, disclosure, and recordkeeping requirements for welfare benefit plans are extensive. ERISA plans must present participants with a plan description explaining, among other things, the plan’s eligibility requirements and claims-processing procedures. §§1021(a)(1), 1022, 1024(b)(1). Plans must notify participants when a claim is denied and state the basis for the denial. §1133(1). Most important for the pre-emption question presented here, welfare benefit plans governed by ERISA must file an annual report with the Secretary of Labor. The report must include a financial statement listing assets and liabilities for the previous year and, further, receipts and disbursements of funds. §§1021(b), 1023(b)(1), 1023(b)(3)(A)–(B), 1024(a). The information on assets and liabilities as well as receipts and disbursements must be provided to plan participants on an annual basis as well. §§1021(a)(2), 1023(b)(3)(A)–(B), 1024(b)(3). Because welfare benefit plans are in the business of providing benefits to plan participants, a plan’s reporting of data on disbursements by definition incorporates paid claims. See Dept. of Labor, Schedule H (Form 5500)Financial Information (2015) (requiring reporting of “[b]enefit claims payable” and “[b]enefit payment and payments to provide benefits”), online at http://www.dol.gov/ebsa/pdf/2015-5500-Schedule-H.pdf (as last visited Feb. 26, 2016). The Secretary of Labor has authority to establish additional reporting and disclosure requirements for ERISA plans. ERISA permits the Secretary to use the data disclosed by plans “for statistical and research purposes, and [to] compile and publish such studies, analyses, reports, and surveys based thereon as he may deem appropriate.” §1026(a). The Secretary also may, “in connection” with any research, “collect, compile, analyze, and publish data, information, and statistics relating to” plans. §1143(a)(1); see also §1143(a)(3) (approving “other studies relating to employee benefit plans, the matters regulated by this subchapter, and the enforcement procedures provided for under this subchapter”). ERISA further permits the Secretary of Labor to “requir[e] any information or data from any [plan] where he finds such data or information is necessary to carry out the purposes of” the statute, §1024(a)(2)(B), and, when investigating a possible statutory violation, “to require the submission of reports, books, and records, and the filing of data” related to other requisite filings, §1134(a)(1). The Secretary has the general power to promulgate regulations “necessary or appropriate” to administer the statute, §1135, and to provide exemptions from any reporting obligations, §1024(a)(3). It should come as no surprise, then, that plans must keep detailed records so compliance with ERISA’s reporting and disclosure requirements may be “verified, explained, or clarified, and checked for accuracy and completeness.” §1027. The records to be retained must “include vouchers, worksheets, receipts, and applicable resolutions.” Ibid.; see also §1135 (allowing the Secretary to “provide for the keeping of books and records, and for the inspection of such books and records”). These various requirements are not mere formalities. Violation of any one of them may result in both civil and criminal liability. See §§1131–1132. As all this makes plain, reporting, disclosure, and recordkeeping are central to, and an essential part of, the uniform system of plan administration contemplated by ERISA. The Court, in fact, has noted often that these requirements are integral aspects of ERISA. See, e.g., Dillingham, 519 U. S., at 327; Travelers, supra, at 651; Ingersoll-Rand, supra, at 137; Massachusetts v. Morash, 490 U. S. 107, 113, 115 (1989) ; Fort Halifax, supra, at 9; Metropolitan Life Ins. Co. v. Massachusetts, 471 U. S. 724, 732 (1985) . Vermont’s reporting regime, which compels plans to report detailed information about claims and plan members, both intrudes upon “a central matter of plan administration” and “interferes with nationally uniform plan administration.” Egelhoff, 532 U. S., at 148. The State’s law and regulation govern plan reporting, disclosure, and—by necessary implication—recordkeeping. These matters are fundamental components of ERISA’s regulation of plan administration. Differing, or even parallel, regulations from multiple jurisdictions could create wasteful administrative costs and threaten to subject plans to wide-ranging liability. See, e.g., 18 V. S. A. §9410(g) (supplying penalties for violation of Vermont’s reporting rules); CVR §10 (same). Pre-emption is necessary to prevent the States from imposing novel, inconsistent, and burdensome reporting requirements on plans. The Secretary of Labor, not the States, is authorized to administer the reporting requirements of plans governed by ERISA. He may exempt plans from ERISA report-ing requirements altogether. See §1024(a)(3); 29 CFR §2520.104–44 (2005) (exempting self-insured health plans from the annual financial reporting requirement). And, he may be authorized to require ERISA plans to report data similar to that which Vermont seeks, though that question is not presented here. Either way, the uniform rule design of ERISA makes it clear that these decisions are for federal authorities, not for the separate States. B Vermont disputes the pre-emption of its reporting regime on several fronts. The State argues that respondent has not demonstrated that the reporting regime in fact has caused it to suffer economic costs. Brief for Petitioner 52–54. But respondent’s challenge is not based on the theory that the State’s law must be pre-empted solely because of economic burdens caused by the state law. See Travelers, 514 U. S., at 668. Respondent argues, rather, that Vermont’s scheme regulates a central aspect of plan administration and, if the scheme is not pre-empted, plans will face the possibility of a body of disuniform state reporting laws and, even if uniform, the necessity to accommodate multiple governmental agencies. A plan need not wait to bring a pre-emption claim until confronted with numerous inconsistent obligations and encumbered with any ensuing costs. Vermont contends, furthermore, that ERISA does not pre-empt the state statute and regulation because the state reporting scheme has different objectives. This Court has recognized that “[t]he principal object of [ERISA] is to protect plan participants and beneficiaries.” Boggs v. Boggs, 520 U. S. 833, 845 (1997) . And “[i]n enacting ERISA, Congress’ primary concern was with the mismanagement of funds accumulated to finance employee benefits and the failure to pay employees benefits from accumulated funds.” Morash, supra, at 115. The State maintains that its program has nothing to do with the financial solvency of plans or the prudent behavior of fiduciaries. See Brief for Petitioner 29. This does not suffice to avoid federal pre-emption. “[P]re-emption claims turn on Congress’s intent.” Travelers, 514 U. S., at 655. The purpose of a state law, then, is relevant only as it may relate to the “scope of the state law that Congress understood would survive,” id., at 656, or “the nature of the effect of the state law on ERISA plans,” Dillingham, supra, at 325. In Travelers, for example, the Court noted that “[b]oth the purpose and the effects of” the state law at issue “distinguish[ed] it from” laws that “function as a regulation of an ERISA plan itself.” 514 U. S., at 658–659. The perceived difference here in the objectives of the Vermont law and ERISA does not shield Vermont’s reporting regime from pre-emption. Vermont orders health insurers, including ERISA plans, to report detailed information about the administration of benefits in a systematic manner. This is a direct regulation of a fundamental ERISA function. Any difference in purpose does not transform this direct regulation of “a central matter of plan administration,” Egelhoff, supra, at 148, into an innocuous and peripheral set of additional rules. The Vermont regime cannot be saved by invoking the State’s traditional power to regulate in the area of public health. The Court in the past has “addressed claims of pre-emption with the starting presumption that Congress does not intend to supplant state law,” in particular state laws regulating a subject of traditional state power. Travelers, supra, at 654–655. ERISA, however, “certainly contemplated the pre-emption of substantial areas of traditional state regulation.” Dillingham, 519 U. S., at 330. ERISA pre-empts a state law that regulates a key facet of plan administration even if the state law exercises a traditional state power. See Egelhoff, 532 U. S., at 151–152. The fact that reporting is a principal and essential feature of ERISA demonstrates that Congress intended to pre-empt state reporting laws like Vermont’s, including those that operate with the purpose of furthering public health. The analysis may be different when applied to a state law, such as a tax on hospitals, see De Buono v. NYSA–ILA Medical and Clinical Services Fund, 520 U. S. 806 (1997) , the enforcement of which necessitates incidental reporting by ERISA plans; but that is not the law before the Court. Any presumption against pre-emption, whatever its force in other instances, cannot validate a state law that enters a fundamental area of ERISA regulation and thereby counters the federal purpose in the way this state law does. IV Respondent suggests that the Patient Protection and Affordable Care Act (ACA), which created new reporting obligations for employer-sponsored health plans and incorporated those requirements into the body of ERISA, further demonstrates that ERISA pre-empts Vermont’s reporting regime. See 29 U. S. C. §1185d; 42 U. S. C. §§300gg–15a, 17; §18031(e)(3). The ACA, however, specified that it shall not “be construed to preempt any State law that does not prevent the application of the provisions” of the ACA. 42 U. S. C. §18041(d). This anti-pre-emption provision might prevent any new ACA-created reporting obligations from pre-empting state reporting regimes like Vermont’s, notwithstanding the incorporation of these requirements in the heart of ERISA. But see 29 U. S. C. §1191(a)(2) (providing that the new ACA provisions shall not be construed to affect or modify the ERISA pre-emption clause as applied to group health plans); 42 U. S. C. §300gg–23(a)(2) (same). The Court has no need to resolve this issue. ERISA’s pre-existing reporting, disclosure, and recordkeeping provisions—upon which the Court’s conclusion rests—maintain their pre-emptive force whether or not the new ACA reporting obligations also pre-empt state law. * * * ERISA’s express pre-emption clause requires invalidation of the Vermont reporting statute as applied to ERISA plans. The state statute imposes duties that are inconsistent with the central design of ERISA, which is to provide a single uniform national scheme for the administration of ERISA plans without interference from laws of the several States even when those laws, to a large extent, impose parallel requirements. The judgment of the Court of Appeals for the Second Circuit is Affirmed.

578.US.2015_14-613

After petitioner Marvin Green complained to his employer, the United States Postal Service, that he was denied a promotion because he was black, his supervisors accused him of the crime of intentionally delaying the mail. In an agreement signed December 16, 2009, the Postal Service agreed not to pursue criminal charges, and Green agreed either to retire or to accept another position in a remote location for much less money. Green chose to retire and submitted his resignation paperwork on February 9, 2010, effective March 31. On March 22—41 days after resigning and 96 days after signing the agreement—Green reported an unlawful constructive discharge to an Equal Employment Opportunity counselor, an administrative prerequisite to filing a complaint alleging discrimination or retaliation in violation of Title VII of the Civil Rights Act of 1964. See 29 CFR §1614.105(a)(1). Green eventually filed suit in Federal District Court, which dismissed his complaint as untimely because he had not contacted the counselor within 45 days of the “matter alleged to be discriminatory,” ibid. The Tenth Circuit affirmed, holding that the 45-day limitations period began to run on December 16, the date Green signed the agreement. Held: 1. Because part of the “matter alleged to be discriminatory” in a constructive-discharge claim is an employee’s resignation, the 45-day limitations period for such action begins running only after an employee resigns. Pp. 4–15. (a) Where, as here, the regulatory text itself is not unambiguously clear, the Court relies on the standard rule for limitations periods, which provides that a limitations period ordinarily begins to run “ ‘when the plaintiff has a complete and present cause of action,’ ” Graham County Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U. S. 409 . Applied here, that rule offers three persuasive reasons to include the employee’s resignation in the limitations period. Pp. 4–10. (i) First, resignation is part of the “complete and present cause of action” in a constructive-discharge claim, which comprises two basic elements: discriminatory conduct such that a reasonable employee would have felt compelled to resign and actual resignation, Pennsylvania State Police v. Suders, 542 U. S. 129 . Until he resigns, an employee does not have a “complete and present cause of action” for constructive discharge. Under the standard rule, only after the employee has a complete and present cause of action does that trigger the limitations period. In this respect, a constructive-discharge claim is no different from an ordinary wrongful-discharge claim, which accrues only after the employee is fired. Pp. 6–8. (ii) Second, although the standard rule may be subject to exception where clearly indicated by the text creating the limitations period, nothing in Title VII or the regulation suggests such displacement. To the contrary, it is natural to read “matter alleged to be discriminatory” as including the allegation forming the basis of the claim, which confirms the standard rule’s applicability. Pp. 8–9. (iii) Third, practical considerations also confirm the merit of applying the standard rule. Starting the clock ticking before a plaintiff can actually file suit does little to further the limitations period’s goals and actively negates Title VII’s remedial structure. A “limitations perio[d] should not commence to run so soon that it becomes difficult for a layman to invoke the protection of the civil rights statutes.” Delaware State College v. Ricks, 449 U. S. 250 , n. 16. Nothing in the regulation suggests a two-step process in which an employee would have to file a complaint after an employer’s discriminatory conduct, only to be forced to amend that complaint to allege constructive discharge after resigning. Requiring that a complaint be filed before resignation occurs would also, e.g., ignore that an employee may not be in a position to leave his job immediately. Pp. 9–10. (b) Arguments against applying the standard rule here are rejected. Suders stands not for the proposition that a constructive discharge is tantamount to a formal discharge for remedial purposes only, but for the rule that constructive discharge is a claim distinct from the underlying discriminatory act, 542 U. S., at 149. Nor was Green’s resignation the mere inevitable consequence of the Postal Service’s discriminatory conduct. Ricks, 449 U. S. 250 , distinguished. Finally, the important goal of promoting conciliation through early, informal contact with a counselor does not warrant treating a constructive discharge different from an actual discharge for purposes of the limitations period. Pp. 10–15. 2. A constructive-discharge claim accrues—and the limitations period begins to run—when the employee gives notice of his resignation, not on the effective date thereof. The Tenth Circuit is left to determine, in the first instance, the date that Green in fact gave notice. P. 16. 760 F. 3d 1135, vacated and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in the judgment. Thomas, J., filed a dissenting opinion.

Title VII of the Civil Rights Act of 1964, 78Stat. 253, as amended, 42 U. S. C. §2000e et seq., prohibits employers from discriminating on the basis of race, color, religion, sex, or national origin, or retaliating against their employees for opposing or seeking relief from such discrimination. Before a federal civil servant can sue his employer for violating Title VII, he must, among other things, “initiate contact” with an Equal Employment Opportunity counselor at his agency “within 45 days of the date of the matter alleged to be discriminatory.” 29 CFR §1614.105(a)(1) (2015). If an employee claims he has been fired for discriminatory reasons, the “matter alleged to be discriminatory” includes the discharge itself and the 45-day limitations period begins running only after the employee is fired. We address here when the limitations period begins to run for an employee who was not fired, but resigns in the face of intolerable discrimination—a “constructive” discharge. We hold that, in such circumstances, the “matter alleged to be discriminatory” includes the employee’s resignation, and that the 45-day clock for a constructive discharge begins running only after the employee resigns. I We recite the following facts in the light most favorable to petitioner Marvin Green, against whom the District Court entered summary judgment. Green is a black man who worked for the Postal Service for 35 years. In 2008, he was serving as the postmaster for Englewood, Colorado when he applied for a promotion to the vacant postmaster position in nearby Boulder. He was passed over. Shortly thereafter, Green complained he was denied the promotion because of his race. Green’s relations with his supervisors crumbled following his complaint. Tensions peaked on December 11, 2009, when two of Green’s supervisors accused him of intentionally delaying the mail—a criminal offense. See 18 U. S. C. §1703. They informed Green that the Postal Service’s Office of the Inspector General (OIG) was investigating the charge and that OIG agents had arrived to interview him as part of their investigation. After Green met with the OIG agents, his supervisors gave him a letter reassigning him to off-duty status until the matter was resolved. Even though the OIG agents reported to Green’s supervisors that no further investigation was warranted, the supervisors continued to represent to Green that “the OIG is all over this” and that the “criminal” charge “could be a life changer.” App. 53. On December 16, 2009, Green and the Postal Service signed an agreement whose meaning remains disputed. Relevant here, the Postal Service promised not to pursue criminal charges in exchange for Green’s promise to leave his post in Englewood. The agreement also apparently gave Green a choice: effective March 31, 2010, he could either retire or report for duty in Wamsutter, Wyoming—population 451—at a salary considerably lower than what he earned in his Denver suburb. Green chose to retire and submitted his resignation to the Postal Service on February 9, 2010, effective March 31. On March 22—41 days after submitting his resignation paperwork to the Postal Service on February 9, but 96 days after signing the settlement agreement on December 16—Green contacted an Equal Employment Opportunity (EEO) counselor to report an unlawful constructive discharge. He contended that his supervisors had threatened criminal charges and negotiated the resulting agreement in retaliation for his original complaint.[1] He alleged that the choice he had been given effectively forced his resignation in violation of Title VII. Green eventually filed suit in the Federal District Court for the District of Colorado, alleging, inter alia, that the Postal Service constructively discharged him. The Postal Service moved for summary judgment, arguing that Green had failed to make timely contact with an EEO counselor within 45 days of the “matter alleged to be discriminatory,” as required by 29 CFR §1614.105(a)(1). The District Court granted the Postal Service’s motion for summary judgment. The Tenth Circuit affirmed, holding that the “matter alleged to be discriminatory” encompassed only the Postal Service’s discriminatory actions and not Green’s independent decision to resign on February 9. Green v. Donahue, 760 F. 3d 1135 (2014). Therefore, the 45-day limitations period started running when both parties signed the settlement agreement on December 16, 2009. Accordingly, because 96 days passed between the agreement and when Green contacted an EEO counselor on March 22, 2010, his constructive-discharge claim was time barred. Two other Courts of Appeals agree with the Tenth Circuit’s view that the limitations period begins to run for a constructive-discharge claim after the employer’s last discriminatory act.[2] As the Tenth Circuit recognized, however, other Courts of Appeals have held that the limitations period for a constructive-discharge claim does not begin to run until the employee resigns.[3] We granted certiorari to resolve this split. 575 U. S. ___ (2015). Because no party here supports the Tenth Circuit’s holding that an employee’s resignation is not part of the “matter alleged to be discriminatory,” we appointed Catherine M. A. Carroll to defend that aspect of the judgment below. 576 U. S. ___ (2015). She has ably discharged her duties and the Court thanks her for her service. II Before a federal civil servant can sue his employer in court for discriminating against him in violation of Title VII, he must first exhaust his administrative remedies. 42 U. S. C. §2000e–16(c). To exhaust those remedies, the Equal Employment Opportunity Commission (EEOC) has promulgated regulations that require, among other things, that a federal employee consult with an EEO counselor prior to filing a discrimination lawsuit. Specifically, he “must initiate contact with a Counselor within 45 days of the date of the matter alleged to be discriminatory or, in the case of personnel action, within 45 days of the effective date of the action.” 29 CFR §1614.105(a)(1).[4] The timeliness of Green’s claim therefore turns on our interpretation of this EEOC regulation implementing Title VII.[5] Although we begin our interpretation of the regulation with its text, the text in this case is not particularly helpful. Nowhere does §1614.105 indicate whether a “matter alleged to be discriminatory” in a constructive-discharge claim includes the employee’s resignation, as Green contends, or only the employer’s discriminatory conduct, as amica contends. The word “matter” simply means “an allegation forming the basis of a claim or defense,” Black’s Law Dictionary 1126 (10th ed. 2014)—a term that could readily apply to a discrimination-precipitated resignation. So the “matter alleged to be discriminatory” could refer to all of the allegations underlying a claim of discrimination, including the employee’s resignation, or only to those allegations concerning the employer’s discriminatory conduct. We therefore must turn to other canons of interpretation. The most helpful canon in this context is “the ‘standard rule’ ” for limitations periods. Graham County Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U. S. 409, 418 (2005) . Ordinarily, a “ ‘limitations period commences when the plaintiff has a complete and present cause of action.’ ” Ibid. “[A] cause of action does not become ‘complete and present’ for limitations purposes until the plaintiff can file suit and obtain relief.” Bay Area Laundry and Dry Cleaning Pension Trust Fund v. Ferbar Corp. of Cal., 522 U. S. 192, 201 (1997) . Although the standard rule can be displaced such that the limitations period begins to run before a plaintiff can file a suit, we “will not infer such an odd result in the absence of any such indication” in the text of the limitations period. Reiter v. Cooper, 507 U. S. 258, 267 (1993) . Applying this default rule, we are persuaded that the “matter alleged to be discriminatory” in a constructive-discharge claim necessarily includes the employee’s resignation for three reasons. First, in the context of a constructive-discharge claim, a resignation is part of the “complete and present cause of action” necessary before a limitations period ordinarily begins to run. Second, nothing in the regulation creating the limitations period here, §1614.105, clearly indicates an intent to displace this standard rule. Third, practical considerations confirm the merit of applying the standard rule here. We therefore interpret the term “matter alleged to be discriminatory” for a constructive-discharge claim to include the date Green resigned. A The standard rule for limitations periods requires us first to determine what is a “complete and present cause of action” for a constructive-discharge claim. We hold that such a claim accrues only after an employee resigns. The constructive-discharge doctrine contemplates a situation in which an employer discriminates against an employee to the point such that his “working conditions become so intolerable that a reasonable person in the employee’s position would have felt compelled to resign.” Pennsylvania State Police v. Suders, 542 U. S. 129, 141 (2004) . When the employee resigns in the face of such circumstances, Title VII treats that resignation as tantamount to an actual discharge. Id., at 142–143. A claim of constructive discharge therefore has two basic elements. A plaintiff must prove first that he was discriminated against by his employer to the point where a reasonable person in his position would have felt compelled to resign. Id., at 148. But he must also show that he actually resigned. Ibid. (“A constructive discharge involves both an employee’s decision to leave and precipitating conduct . . .” (emphasis added)). In other words, an employee cannot bring a constructive-discharge claim until he is constructively discharged. Only after both elements are satisfied can he file suit to obtain relief. Under the standard rule for limitations periods, the limitations period should begin to run for a constructive-discharge claim only after a plaintiff resigns. At that point—and not before—he can file a suit for constructive discharge. So only at that point—and not before—does he have a “complete and present” cause of action. And only after he has a complete and present cause of action does a limitations period ordinarily begin to run. Cf. Mac’s Shell Service, Inc. v. Shell Oil Products Co., 559 U. S. 175 –190 (2010) (the limitations period for a constructive termination of a franchise agreement starts running when the agreement is constructively terminated). In this respect, a claim that an employer constructively discharged an employee is no different from a claim that an employer actually discharged an employee. An ordinary wrongful discharge claim also has two basic elements: discrimination and discharge. See St. Mary’s Honor Center v. Hicks, 509 U. S. 502, 506 (1993) ; 1 B. Lindemann, P. Grossman, & C. Weirich, Employment Discrimination Law 21–33 (5th ed. 2012) (Lindemann) (“The sine qua non of a discharge case is, of course, a discharge”). The claim accrues when the employee is fired. At that point—and not before—he has a “complete and present cause of action.” So at that point—and not before—the limitations period begins to run. With claims of either constructive discharge or actual discharge, the standard rule thus yields the same result: a limitations period should not begin to run until after the discharge itself. In light of this rule, we interpret the term “matter alleged to be discriminatory” in §1614.105 to refer to all of the elements that make up a constructive-discharge claim—including an employee’s resignation. B Although the standard rule dictates that a limitations period should commence only after a claim accrues, there is an exception to that rule when the text creating the limitations period clearly indicates otherwise. See, e.g., Dodd v. United States, 545 U. S. 353, 360 (2005) . Nothing in the text of Title VII or the regulation, however, suggests that the standard rule should be displaced here. To the contrary, the language of the regulation confirms our application of the default rule. As noted previously, the word “matter” generally refers to “an allegation forming the basis of a claim or defense.” Black’s Law Dictionary 1126. The natural reading of “matter alleged to be discriminatory” thus refers to the allegation forming the basis of the discrimination claim—here, a claim of constructive discharge. And as discussed above, a constructive discharge claim requires two basic allegations: discriminatory conduct by the employer that leads to resignation of the employee. So long as those acts are part of the same, single claim under consideration, they are part of the “matter alleged to be discriminatory,” whatever the role of discrimination in each individual element of the claim. Cf. National Railroad Passenger Corporation v. Morgan, 536 U. S. 101 –121 (2002) (holding that a hostile-work-environment claim is a single “unlawful employment practice” that includes every act composing that claim, whether those acts are independently actionable or not). C Finally, we are also persuaded that applying the standard rule for limitations periods to constructive discharge makes a good deal of practical sense. Starting the limitations clock ticking before a plaintiff can actually sue for constructive discharge serves little purpose in furthering the goals of a limitations period—and it actively negates Title VII’s remedial structure. Cf. Zipes v. Trans World Airlines, Inc., 455 U. S. 385, 398 (1982) (holding that the Title VII limitations period should be construed to “honor the remedial purpose of the legislation as a whole without negating the particular purpose of the filing requirement”). This Court has recognized “that the limitations perio[d] should not commence to run so soon that it becomes difficult for a layman to invoke the protection of the civil rights statutes.” Delaware State College v. Ricks, 449 U. S. 250 , n. 16 (1980). If the limitations period begins to run following the employer’s precipitating discriminatory conduct, but before the employee’s resignation, the employee will be forced to file a discrimination complaint after the employer’s conduct and later amend the complaint to allege constructive discharge after he resigns. Nothing in the regulation suggests it intended to require a layperson, while making this difficult decision, to follow such a two-step process in order to preserve any remedy if he is constructively discharged. Moreover, forcing an employee to lodge a complaint before he can bring a claim for constructive discharge places that employee in a difficult situation. An employee who suffered discrimination severe enough that a reasonable person in his shoes would resign might nevertheless force himself to tolerate that discrimination for a period of time. He might delay his resignation until he can afford to leave. Or he might delay in light of other circumstances, as in the case of a teacher waiting until the end of the school year to resign. Tr. 17. And, if he feels he must stay for a period of time, he may be reluctant to complain about discrimination while still employed. A complaint could risk termination—an additional adverse consequence that he may have to disclose in future job applications. III Amica and the dissent read “matter alleged to be discriminatory” as having a clear enough meaning to displace our reliance on the standard rule for limitations periods. They argue that “matter” is not equivalent to “claim” or “cause of action,” and that the use of the phrase “matter alleged to be discriminatory” is a sufficiently clear statement that the standard claim accrual rule should not apply. According to amica and the dissent, “matter” refers only to the discriminatory acts of the Postal Service, not Green’s resignation. We disagree. There is nothing inherent in the phrase “matter alleged to be discriminatory” that clearly limits it to employer conduct. Rather, as discussed above, the term can reasonably be interpreted to include the factual basis for a claim. Green is not alleging just that the Postal Service discriminated against him. He claims that the discrimination left him no choice but to resign. Amica and the dissent dispute that a constructive discharge is a separate claim. According to amica and the dissent, the constructive-discharge doctrine merely allows a plaintiff to expand any underlying discrimination claim to include the damages from leaving his job, thereby increasing his available remedies. See 1 Lindemann 21–49 (constructive discharge allows plaintiff to seek backpay, front pay, or reinstatement). In support of this argument, amica and the dissent emphasize this Court’s statement in Suders that “[u]nder the constructive discharge doctrine, an employee’s reasonable decision to resign because of unendurable working conditions is assimilated to a formal discharge for remedial purposes.” 542 U. S., at 141 (emphasis added); see also id., at 148 (“[A] constructive discharge is functionally the same as an actual termination in damages-enhancing respects”). But the Court did not hold in Suders that a constructive discharge is tantamount to a formal discharge for remedial purposes exclusively. To the contrary, it expressly held that constructive discharge is a claim distinct from the underlying discriminatory act. Id., at 149 (holding that a hostile-work-environment claim is a “lesser included component” of the “graver claim of hostile-environment constructive discharge”). This holding was no mere dictum. See id., at 142 (“[A] claim for constructive discharge lies under Title VII”). We see no reason to excise an employee’s resignation from his constructive-discharge claim for purposes of the limitations period. The concurrence sets out a theory that there are two kinds of constructive discharge for purposes of the limitations period: constructive discharge “claims” where the employer “makes conditions intolerable with the specific discriminatory intent of forcing the employee to resign,” and constructive discharge “damages” where the employer does not intend to force the employee to quit, but the discriminatory conditions of employment are so intolerable that the employee quits anyway. Post, at 6–11 (Alito, J., concurring in judgment). According to the concurrence, the limitations period does not begin to run until an employee resigns under the “claim” theory of constructive discharge, but begins at the last discriminatory act before resignation under the “damages” theory. This sometimes-a-claim-sometimes-not theory of constructive discharge is novel and contrary to the constructive discharge doctrine. The whole point of allowing an employee to claim “constructive” discharge is that in circumstances of discrimination so intolerable that a reasonable person would resign, we treat the employee’s resignation as though the employer actually fired him. Suders, 542 U. S., at 141–143.[6] We do not also require an employee to come forward with proof—proof that would often be difficult to allege plausibly—that not only was the dis-crimination so bad that he had to quit, but also that his quitting was his employer’s plan all along. Amica and the dissent also argue that their interpretation is more consistent with this Court’s prior precedent on when the limitations period begins to run for discrimination claims. Under their interpretation, Green’s resignation was not part of the discriminatory “matter,” but was instead the mere inevitable consequence of the Postal Service’s discriminatory conduct, and therefore cannot be used to extend the limitations period. See Brief for Court-Appointed Amica Curiae in Support of Judgment Below 21–27 (Brief for Amica Curiae) (citing Ledbetter v. Goodyear Tire & Rubber Co., 550 U. S. 618 (2007) , overruled by statute, Lilly Ledbetter Fair Pay Act of 2009, 123Stat. 5; Delaware State College v. Ricks, 449 U. S. 250 ; United Air Lines, Inc. v. Evans, 431 U. S. 553 (1977) ); post, at 3–7 (Thomas, J., dissenting) (citing Ricks, 449 U. S. 250 , and Chardon v. Fernandez, 454 U. S. 6 (1981) (per curiam)). Similarly, the concurrence argues these cases require that an act done with discriminatory intent must occur within the limitations period. Post, at 4 (opinion of Alito, J.). But these cases are consistent with the standard rule that a limitations period begins to run after a claim accrues, not after an inevitable consequence of that claim. In Ricks, for example, the Court considered the discrimination claim of a college faculty member who was denied tenure and given a 1-year “ ‘terminal’ ” contract for his last year teaching. 449 U. S., at 258. The plaintiff’s claim accrued—and he could have sued—when the college informed him he would be denied tenure and gave him “explicit notice that his employment would end” when his 1-year contract expired. Ibid. The Court held that the limitations period began to run on that date, and not after his 1-year contract expired. That final year of teaching was merely an inevitable consequence of the tenure denial the plaintiff claimed was discriminatory. Green’s resignation, by contrast, is not merely an inevitable consequence of the discrimination he suffered; it is an essential part of his constructive-discharge claim. That is, Green could not sue for constructive discharge until he actually resigned. Of course, Green could not resign and then wait until the consequences of that resignation became most painful to complain. For example, he could not use the date of the expiration of his health insurance after his resignation to extend the limitations period. But the “inevitable consequence” principle of Ricks, Ledbetter, and Evans does not change the focus of the limitations period, which remains on the claim of discrimination itself. See Lewis v. Chicago, 560 U. S. 205, 214 (2010) (holding Evans and its progeny “establish only that a Title VII plaintiff must show a present violation within the limitations period” (internal quotation marks omitted)); National Railroad Passenger Corporation v. Morgan, 536 U. S., at 115–121 (holding limitations period for hostile-work-environment claim runs from the last act composing the claim).[7] For a constructive discharge, the claim does not exist until the employee resigns. Finally, amica contends that her interpretation of the regulation better advances the EEOC’s goal of promoting conciliation for federal employees through early, informal contact with an EEO counselor. See Exec. Order No. 11478, §4, 34 Fed. Reg. 12986 (1969) (counseling for federal employees “shall encourage the resolution of employee problems on an informal basis”). The dissent suggests that our holding will make a discrimination victim the master of his complaint, permitting him to “ ‘exten[d] the limitation[s period] indefinitely’ ” by waiting to resign. Post, at 7 (opinion of Thomas, J.). The concurrence claims that an employee who relies on the limitations period in waiting to resign is “doubly out of luck” if his otherwise-meritorious discrimination claim is time barred and he cannot show the discrimination was so intolerable that it amounted to a constructive discharge. Post, at 13 (opinion of Alito, J.). These concerns are overblown. Amica may be right that it is more difficult to achieve conciliation after an employee resigns. But the same is true for a federal civil servant who is fired by his agency for what the employee believes to be a discriminatory purpose. And neither decision is necessarily permanent—a resignation or a termination may be undone after an employee contacts a counselor. Conciliation, while important, does not warrant treating a constructive discharge different from an actual discharge for purposes of the limitations period. As for the dissent’s fear, we doubt that a victim of employment discrimination will continue to work in an intolerable environment merely because he can thereby extend the limitations period for a claim of constructive discharge. If anything, a plaintiff who wishes to prevail on the merits of his constructive discharge claim has the opposite incentive. A claim of constructive discharge requires proof of a causal link between the allegedly intolerable conditions and the resignation. See 1 Lindemann 21–45, and n. 106. And as for the concurrence’s double-loser concern, no plaintiff would be well advised to delay pursuing what he believes to be a meritorious non-constructive-discharge-discrimination claim on the ground that a timely filed constructive discharge claim could resuscitate other time-lapsed claims. The 45-day limitations period begins running on any separate underlying claim of discrimination when that claim accrues, regardless of whether the plaintiff eventually claims constructive discharge. The limitations-period analysis is always conducted claim by claim. IV Our decision that a resignation triggers the limitations period for a constructive-discharge claim raises the question of when precisely an employee resigns. Here, Green and the Government agree that an employee resigns when he gives his employer definite notice of his intent to resign. If an employee gives “two weeks’ notice”—telling his employer he intends to leave after two more weeks of employment—the limitations period begins to run on the day he tells his employer, not his last day at work. (This issue was not addressed by the Tenth Circuit and, accordingly, amica takes no position on it. See Brief for Amica Curiae 42.) We agree. A notice rule flows directly from this Court’s precedent. In Ricks, 449 U. S., at 250, and Chardon v. Fernandez, 454 U. S. 6 , the Court explained that an ordinary wrongful-discharge claim accrues—and the limitations period begins to run—when the employer notifies the employee he is fired, not on the last day of his employment. Ricks, 449 U. S., at 258–259; Chardon, 454 U. S., at 8. Likewise, here, we hold that a constructive-discharge claim accrues—and the limitations period begins to run—when the employee gives notice of his resignation, not on the effective date of that resignation. One factual issue remains: when exactly Green gave the Postal Service notice of his resignation. The Government argues that Green resigned on December 16, 2009—when he signed the settlement agreement—and that his claim is therefore still time barred. Green argues that he did not resign until February 9, 2010—when he submitted his retirement paperwork—and that his claim is therefore timely. We need not resolve this issue. Having concluded that the limitations period for Green’s constructive-discharge claim runs from the date he gave notice of his resignation, we leave it to the Tenth Circuit to determine when this in fact occurred. * * * For these reasons, we vacate the judgment of the Tenth Circuit and remand the case for further proceedings consistent with this opinion. So ordered.Notes 1 We assume without deciding that it is unlawful for a federal agency to retaliate against a civil servant for complaining of discrimination. See Gómez-Pérez v. Potter, 553 U. S. 474 , n. 4 (2008); Brief for Respondent 2. 2 Mayers v. Laborers’ Health and Safety Fund of North America, 478 F. 3d 364, 370 (CADC 2007) (per curiam); Davidson v. Indiana-American Water Works, 953 F. 2d 1058, 1059 (CA7 1992). 3 Flaherty v. Metromail Corp., 235 F. 3d 133, 138 (CA2 2000); Draper v. Coeur Rochester, Inc., 147 F. 3d 1104, 1111 (CA9 1998); Hukkanen v. Operating Engineers, 3 F. 3d 281, 285 (CA8 1993); Young v. National Center for Health Servs. Research, 828 F. 2d 235, 238 (CA4 1987). 4 This regulation, applicable to federal employees only, has a statutory analog for private-sector Title VII plaintiffs, who are required to file a charge with the EEOC within 180 or 300 days “after the alleged unlawful employment practice occurred.” 42 U. S. C. §2000e–5(e)(1). Al-though the language is different, the EEOC treats the federal and private-sector employee limitations periods as identical in operation. See EEOC Compliance Manual: Threshold Issues §2–IV(C)(1), n. 179. 5 Green does not contend that his alleged constructive discharge is a “personnel action.” See Brief for Petitioner 17–18; Green v. Donahoe, 760 F. 3d 1135, 1144, n. 3 (CA10 2014). We therefore address the “matter alleged to be discriminatory” clause only. 6 The concurrence suggests that its theory is consistent with statements in the Suders opinion that constructive discharge is akin to an actual discharge “ ‘for remedial purposes’ ” and in “ ‘damages-enhancing respects.’ ” Post, at 10 (opinion of Alito, J.) (quoting Suders, 542 U. S., at 141, 148). This ignores the more obvious explanation for this qualification: The Court was distinguishing between the merits of a claim of constructive discharge generally, where resignation is imputed as a discriminatory act of the employer, and the affirmative defense avail-able to an employer in a hostile work environment claim specifically, which allows an employer to defend against a hostile work environment claim in certain circumstances if it took no “ ‘official act’ ” against the employee. Id., at 143–146. The Court in Suders recognized that it would be bizarre to always impute resignation as an “official act” of the employer in a constructive discharge hostile work environment case and prohibit the employer from relying on the no-“official-act” defense, because it would make it easier to prove the “graver” claim of a constructive discharge hostile work environment than to prove a hostile work environment claim. Id., at 148–149. Thus, the Court declined to hold that resignation in a constructive discharge case was categorically an “official act” in all instances. Ibid. In other words, the Court sought a measure of parity between constructive discharge and ordinary discrimination—parity that we extend to the limitations period here. 7 The dissent relies on Morgan’s other holding that, unlike a hostile-work-environment claim that may comprise many discriminatory acts, discrete claims of discrimination based on independent discriminatory acts cannot be aggregated to extend the limitations period. See post, at 3 (opinion of Thomas, J.) (citing 536 U. S., at 109–113). But this just proves the point: The analysis for the limitations period turns on the nature of the specific legal claim at issue. In Morgan, the Court noted that even if a claim of discrimination based on a single discriminatory act is time barred, that same act could still be used as part of the basis for a hostile-work-environment claim, so long as one other act that was part of that same hostile-work-environment claim occurred within the limitations period. Id., at 117 (“It is precisely because the entire hostile work environment encompasses a single unlawful employment practice that we do not hold, as have some of the Circuits, that the plaintiff may not base a suit on individual acts that occurred outside the statute of limitations . . . ”).

579.US.2015_14-1513

Section 284 of the Patent Act provides that, in a case of infringement, courts “may increase the damages up to three times the amount found or assessed.” 35 U. S. C. §284. The Federal Circuit has adopted a two-part test for determining whether damages may be increased pursuant to §284. First, a patent owner must “show by clear and convincing evidence that the infringer acted despite an objectively high likelihood that its actions constituted infringement of a valid patent.” In re Seagate Technology, LLC, 497 F. 3d 1360, 1371. Second, the patentee must demonstrate, also by clear and convincing evidence, that the risk of infringement “was either known or so obvious that it should have been known to the accused infringer.” Ibid. Under Federal Circuit precedent, an award of enhanced damages is subject to trifurcated appellate review. The first step of Seagate—objective recklessness—is reviewed de novo; the second—subjective knowledge—for substantial evidence; and the ultimate decision—whether to award enhanced damages—for abuse of discretion. In each of these cases, petitioners were denied enhanced damages under the Seagate framework. Held: The Seagate test is not consistent with §284. Pp. 7–15. (a) The pertinent language of §284 contains no explicit limit or condition on when enhanced damages are appropriate, and this Court has emphasized that the “word ‘may’ clearly connotes discretion.” Martin v. Franklin Capital Corp., 546 U. S. 132 . At the same time, however, “[d]iscretion is not whim.” Id., at 139. Although there is “no precise rule or formula” for awarding damages under §284, a district court’s “discretion should be exercised in light of the considerations” underlying the grant of that discretion. Octane Fitness, LLC v. ICON Health & Fitness, Inc., 572 U. S. ___, ___. Here, 180 years of enhanced damage awards under the Patent Act establish that they are not to be meted out in a typical infringement case, but are instead designed as a sanction for egregious infringement behavior. Pp. 7–9. (b) In many respects, the Seagate test rightly reflects this historic guidance. It is, however, “unduly rigid, and . . . impermissibly encumbers the statutory grant of discretion to district courts.” Octane Fitness, 572 U. S., at ___. Pp. 9–13. (1) By requiring an objective recklessness finding in every case, the Seagate test excludes from discretionary punishment many of the most culpable offenders, including the “wanton and malicious pirate” who intentionally infringes a patent—with no doubts about its validity or any notion of a defense—for no purpose other than to steal the patentee’s business. Seymour v. McCormick, 16 How. 480, 488. Under Seagate, a district court may not even consider enhanced damages for such a pirate, unless the court first determines that his infringement was “objectively” reckless. In the context of such deliberate wrongdoing, however, it is not clear why an independent showing of objective recklessness should be a prerequisite to enhanced damages. Octane Fitness arose in a different context but is instructive here. There, a two-part test for determining when a case was “exceptional”—and therefore eligible for an award of attorney’s fees—was rejected because a claim of “subjective bad faith” alone could “warrant a fee award.” 572 U. S., at ___. So too here: A patent infringer’s subjective willfulness, whether intentional or knowing, may warrant enhanced damages, without regard to whether his infringement was objectively reckless. The Seagate test further errs by making dispositive the ability of the infringer to muster a reasonable defense at trial, even if he did not act on the basis of that defense or was even aware of it. Culpability, however, is generally measured against the actor’s knowledge at the time of the challenged conduct. In sum, §284 allows district courts to punish the full range of culpable behavior. In so doing, they should take into account the particular circumstances of each case and reserve punishment for egregious cases typified by willful misconduct. Pp. 9–11. (2) Seagate’s requirement that recklessness be proved by clear and convincing evidence is also inconsistent with §284. Once again, Octane Fitness is instructive. There, a clear and convincing standard for awards of attorney’s fees was rejected because the statute at issue supplied no basis for imposing a heightened standard. Here, too, §284 “imposes no specific evidentiary burden, much less such a high one,” 572 U. S., at ___. And the fact that Congress erected a higher standard of proof elsewhere in the Patent Act, but not in §284, is telling. “[P]atent-infringement litigation has always been governed by a preponderance of the evidence standard.” Id., at ___. Enhanced damages are no exception. P. 12. (3) Having eschewed any rigid formula for awarding enhanced damages under §284, this Court likewise rejects the Federal Circuit’s tripartite appellate review framework. In Highmark Inc. v. Allcare Health Management System, Inc., 572 U. S. ___, the Court built on the Octane Fitness holding—which confirmed district court discretion to award attorney’s fees—and rejected a similar multipart standard of review in favor of abuse of discretion review. The same conclusion follows naturally from the holding here: Because §284 “commits the determination” whether enhanced damages are appropriate to the district court’s discretion, “that decision is to be reviewed on appeal for abuse of discretion.” Id., at ___. Nearly two centuries of enhanced damage awards have given substance to the notion that district courts’ discretion is limited, and the Federal Circuit should review their exercise of that discretion in light of longstanding considerations that have guided both Congress and the courts. Pp. 12–13. (c) Respondents’ additional arguments are unpersuasive. They claim that Congress ratified the Seagate test when it reenacted §284 in 2011 without pertinent change, but the reenacted language unambiguously confirmed discretion in the district courts. Neither isolated snippets of legislative history nor a reference to willfulness in another recently enacted section reflects an endorsement of Seagate’s test. Respondents are also concerned that allowing district courts unlimited discretion to award enhanced damages could upset the balance between the protection of patent rights and the interest in technological innovation. That concern—while serious—cannot justify imposing an artificial construct such as the Seagate test on the limited discretion conferred under §284. Pp. 13–15. No. 14–1513, 769 F. 3d 1371; No. 14–1520, 782 F. 3d 649, vacated and remanded. Roberts, C. J., delivered the opinion for a unanimous Court. Breyer, J., filed a concurring opinion, in which Kennedy and Alito, JJ., joined.Notes 1 Together with No. 14–1520, Stryker Corp. et al. v. Zimmer, Inc., et al., also on certiorari to the same court.

Section 284 of the Patent Act provides that, in a case of infringement, courts “may increase the damages up to three times the amount found or assessed.” 35 U. S. C. §284. In In re Seagate Technology, LLC, 497 F. 3d 1360 (2007) (en banc), the United States Court of Appeals for the Federal Circuit adopted a two-part test for determining when a district court may increase damages pursuant to §284. Under Seagate, a patent owner must first “show by clear and convincing evidence that the infringer acted despite an objectively high likelihood that its actions constituted infringement of a valid patent.” Id., at 1371. Second, the patentee must demonstrate, again by clear and convincing evidence, that the risk of infringement “was either known or so obvious that it should have been known to the accused infringer.” Ibid. The question before us is whether this test is consistent with §284. We hold that it is not. I A Enhanced damages are as old as U. S. patent law. The Patent Act of 1793 mandated treble damages in any successful infringement suit. See Patent Act of 1793, §5, 1Stat. 322. In the Patent Act of 1836, however, Congress changed course and made enhanced damages discretionary, specifying that “it shall be in the power of the court to render judgment for any sum above the amount found by [the] verdict . . . not exceeding three times the amount thereof, according to the circumstances of the case.” Pat-ent Act of 1836, §14, 5Stat. 123. In construing that new provision, this Court explained that the change was prompted by the “injustice” of subjecting a “defendant who acted in ignorance or good faith” to the same treatment as the “wanton and malicious pirate.” Seymour v. McCormick, 16 How. 480, 488 (1854). There “is no good reason,” we observed, “why taking a man’s property in an invention should be trebly punished, while the measure of damages as to other property is single and actual damages.” Id., at 488–489. But “where the injury is wanton or malicious, a jury may inflict vindictive or exemplary damages, not to recompense the plaintiff, but to punish the defendant.” Id., at 489. The Court followed the same approach in other decisions applying the 1836 Act, finding enhanced damages appropriate, for instance, “where the wrong [had] been done, under aggravated circumstances,” Dean v. Mason, 20 How. 198, 203 (1858), but not where the defendant “appeared in truth to be ignorant of the existence of the patent right, and did not intend any infringement,” Hogg v. Emerson, 11 How. 587, 607 (1850). See also Livingston v. Woodworth, 15 How. 546, 560 (1854) (“no ground” to inflict “penalty” where infringers were not “wanton”). In 1870, Congress amended the Patent Act, but preserved district court discretion to award up to treble damages “according to the circumstances of the case.” Patent Act of 1870, §59, 16Stat. 207. We continued to describe enhanced damages as “vindictive or punitive,” which the court may “inflict” when “the circumstances of the case appear to require it.” Tilghman v. Proctor, 125 U. S. 136 –144 (1888); Topliff v. Topliff, 145 U. S. 156, 174 (1892) (infringer knowingly sold copied technology of his former employer). At the same time, we reiterated that there was no basis for increased damages where “[t]here is no pretence of any wanton and wilful breach” and “nothing that suggests punitive damages, or that shows wherein the defendant was damnified other than by the loss of the profits which the plaintiff received.” Cincinnati Siemens-Lungren Gas Illuminating Co. v. Western Siemens-Lungren Co., 152 U. S. 200, 204 (1894) . Courts of Appeals likewise characterized enhanced damages as justified where the infringer acted deliberately or willfully, see, e.g., Baseball Display Co. v. Star Ballplayer Co., 35 F. 2d 1, 3–4 (CA3 1929) (increased damages award appropriate “because of the deliberate and willful infringement”); Power Specialty Co. v. Connecticut Light & Power Co., 80 F. 2d 874, 878 (CA2 1936) (“wanton, deliberate, and willful” infringement); Brown Bag Filling Mach. Co. v. Drohen, 175 F. 576, 577 (CA2 1910) (“a bald case of piracy”), but not where the infringement “was not wanton and deliberate,” Rockwood v. General Fire Extinguisher Co., 37 F. 2d 62, 66 (CA2 1930), or “conscious and deliberate,” Goodyear Tire & Rubber Co. v. Overman Cushion Tire Co., 95 F. 2d 978, 986 (CA6 1938). Some early decisions did suggest that enhanced dam-ages might serve to compensate patentees as well as to punish infringers. See, e.g., Clark v. Wooster, 119 U. S. 322, 326 (1886) (noting that “[t]here may be damages beyond” licensing fees “but these are more properly the subjects” of enhanced damage awards). Such statements, however, were not for the ages, in part because the merger of law and equity removed certain procedural obstacles to full compensation absent enhancement. See generally 7 Chisum on Patents §20.03[4][b][iii], pp. 20–343 to 20–344 (2011). In the main, moreover, the references to compensation concerned costs attendant to litigation. See Clark, 119 U. S., at 326 (identifying enhanced damages as compensation for “the expense and trouble the plaintiff has been put to”); Day v. Woodworth, 13 How. 363, 372 (1852) (enhanced damages appropriate when defendant was “stubbornly litigious” or “caused unnecessary expense and trouble to the plaintiff”); Teese v. Huntingdon, 23 How. 2, 8–9 (1860) (discussing enhanced damages in the context of “counsel fees”). That concern dissipated with the enactment in 1952 of 35 U. S. C. §285, which authorized district courts to award reasonable attorney’s fees to prevailing parties in “exceptional cases” under the Patent Act. See Octane Fitness, LLC v. ICON Health & Fitness Inc., 572 U. S. ___, ___ (2014) (slip op., at 7). It is against this backdrop that Congress, in the 1952 codification of the Patent Act, enacted §284. “The stated purpose” of the 1952 revision “was merely reorganization in language to clarify the statement of the statutes.” Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U. S. 476, 505, n. 20 (1964) (internal quotation marks omitted). This Court accordingly described §284—consistent with the history of enhanced damages under the Patent Act—as providing that “punitive or ‘increased’ damages” could be recovered “in a case of willful or bad-faith infringement.” Id., at 508; see also Dowling v. United States, 473 U. S. 207 , n. 19 (1985) (“willful infringement”); Florida Prepaid Postsecondary Ed. Expense Bd. v. College Savings Bank, 527 U. S. 627 , n. 11 (1999) (describing §284 damages as “punitive”). B In 2007, the Federal Circuit decided Seagate and fashioned the test for enhanced damages now before us. Under Seagate, a plaintiff seeking enhanced damages must show that the infringement of his patent was “willful.” 497 F. 3d, at 1368. The Federal Circuit announced a two-part test to establish such willfulness: First, “a patentee must show by clear and convincing evidence that the infringer acted despite an objectively high likelihood that its actions constituted infringement of a valid patent,” without regard to “[t]he state of mind of the accused infringer.” Id., at 1371. This objectively defined risk is to be “determined by the record developed in the infringement proceedings.” Ibid. “Objective recklessness will not be found” at this first step if the accused infringer, during the infringement proceedings, “raise[s] a ‘substantial question’ as to the validity or noninfringement of the patent.” Bard Peripheral Vascular, Inc. v. W. L. Gore & Assoc., Inc., 776 F. 3d 837, 844 (CA Fed. 2015). That categorical bar applies even if the defendant was unaware of the arguable defense when he acted. See Seagate, 497 F. 3d, at 1371; Spine Solutions, Inc. v. Medtronic Sofamor Danek USA, Inc., 620 F. 3d 1305, 1319 (CA Fed. 2010). Second, after establishing objective recklessness, a patentee must show—again by clear and convincing evidence—that the risk of infringement “was either known or so obvious that it should have been known to the accused infringer.” Seagate, 497 F. 3d, at 1371. Only when both steps have been satisfied can the district court proceed to consider whether to exercise its discretion to award enhanced damages. Ibid. Under Federal Circuit precedent, an award of enhanced damages is subject to trifurcated appellate review. The first step of Seagate—objective recklessness—is reviewed de novo; the second—subjective knowledge—for substantial evidence; and the ultimate decision—whether to award enhanced damages—for abuse of discretion. See Bard Peripheral Vascular, Inc. v. W. L. Gore & Assoc., Inc., 682 F. 3d 1003, 1005, 1008 (CA Fed. 2012); Spectralytics, Inc. v. Cordis Corp., 649 F. 3d 1336, 1347 (CA Fed. 2011). C 1 Petitioner Halo Electronics, Inc., and respondents Pulse Electronics, Inc., and Pulse Electronics Corporation (collectively, Pulse) supply electronic components. 769 F. 3d 1371, 1374–1375 (CA Fed. 2014). Halo alleges that Pulse infringed its patents for electronic packages containing transformers designed to be mounted to the surface of circuit boards. Id., at 1374. In 2002, Halo sent Pulse two letters offering to license Halo’s patents. Id., at 1376. After one of its engineers concluded that Halo’s patents were invalid, Pulse continued to sell the allegedly infringing products. Ibid. In 2007, Halo sued Pulse. Ibid. The jury found that Pulse had infringed Halo’s patents, and that there was a high probability it had done so willfully. Ibid. The District Court, however, declined to award enhanced damages under §284, after determining that Pulse had at trial presented a defense that “was not objectively baseless, or a ‘sham.’ ” App. to Pet. for Cert. in No. 14–1513, p. 64a (quoting Bard, 682 F. 3d, at 1007). Thus, the court concluded, Halo had failed to show objective recklessness under the first step of Seagate. App. to Pet. for Cert. in No. 14–1513, at 65a. The Federal Circuit affirmed. 769 F. 3d 1371 (2014). 2 Petitioners Stryker Corporation, Stryker Puerto Rico, Ltd., and Stryker Sales Corporation (collectively, Stryker) and respondents Zimmer, Inc., and Zimmer Surgical, Inc. (collectively, Zimmer), compete in the market for orthopedic pulsed lavage devices. App. to Pet. for Cert. in No. 14–1520, p. 49a. A pulsed lavage device is a combination spray gun and suction tube, used to clean tissue during surgery. Ibid. In 2010, Stryker sued Zimmer for patent infringement. 782 F. 3d 649, 653 (CA Fed. 2015). The jury found that Zimmer had willfully infringed Stryker’s patents and awarded Stryker $70 million in lost profits. Ibid. The District Court added $6.1 million in supplemental damages and then trebled the total sum under §284, resulting in an award of over $228 million. App. in No. 14–1520, pp. 483–484. Specifically, the District Court noted, the jury had heard testimony that Zimmer had “all-but instructed its design team to copy Stryker’s products,” App. to Pet. for Cert. in No. 14–1520, at 77a, and had chosen a “high-risk/high-reward strategy of competing immediately and aggressively in the pulsed lavage market,” while “opt[ing] to worry about the potential legal consequences later,” id., at 52a. “[T]reble damages [were] appropriate,” the District Court concluded, “[g]iven the one-sidedness of the case and the flagrancy and scope of Zimmer’s infringement.” Id., at 119a. The Federal Circuit affirmed the judgment of infringement but vacated the award of treble damages. 782 F. 3d, at 662. Applying de novo review, the court concluded that enhanced damages were unavailable because Zimmer had asserted “reasonable defenses” at trial. Id., at 661–662. We granted certiorari in both cases, 577 U. S. ___ (2015), and now vacate and remand. II A The pertinent text of §284 provides simply that “the court may increase the damages up to three times the amount found or assessed.” 35 U. S. C. §284. That language contains no explicit limit or condition, and we have emphasized that the “word ‘may’ clearly connotes discretion.” Martin v. Franklin Capital Corp., 546 U. S. 132, 136 (2005) (quoting Fogerty v. Fantasy, Inc., 510 U. S. 517, 533 (1994) ). At the same time, “[d]iscretion is not whim.” Martin, 546 U. S., at 139. “[I]n a system of laws discretion is rarely without limits,” even when the statute “does not specify any limits upon the district courts’ discretion.” Flight Attendants v. Zipes, 491 U. S. 754, 758 (1989) . “[A] motion to a court’s discretion is a motion, not to its inclination, but to its judgment; and its judgment is to be guided by sound legal principles.” Martin, 546 U. S., at 139 (quoting United States v. Burr, 25 F. Cas. 30, 35 (No. 14,692d) (CC Va. 1807) (Marshall, C. J.); alteration omitted). Thus, although there is “no precise rule or formula” for awarding damages under §284, a district court’s “discretion should be exercised in light of the considerations” underlying the grant of that discretion. Octane Fitness, 572 U. S., at ___ (slip op., at 8) (quoting Fogerty, 510 U. S., at 534). Awards of enhanced damages under the Patent Act over the past 180 years establish that they are not to be meted out in a typical infringement case, but are instead designed as a “punitive” or “vindictive” sanction for egregious infringement behavior. The sort of conduct warranting enhanced damages has been variously described in our cases as willful, wanton, malicious, bad-faith, deliberate, consciously wrongful, flagrant, or—indeed—characteristic of a pirate. See supra, at 2–5. District courts enjoy discretion in deciding whether to award enhanced damages, and in what amount. But through nearly two centuries of discretionary awards and review by appellate tribunals, “the channel of discretion ha[s] narrowed,” Friendly, Indiscretion About Discretion, 31 Emory L. J. 747, 772 (1982), so that such damages are generally reserved for egregious cases of culpable behavior. B The Seagate test reflects, in many respects, a sound recognition that enhanced damages are generally appropriate under §284 only in egregious cases. That test, however, “is unduly rigid, and it impermissibly encumbers the statutory grant of discretion to district courts.” Octane Fitness, 572 U. S., at ___ (slip op., at 7) (construing §285 of the Patent Act). In particular, it can have the effect of insulating some of the worst patent infringers from any liability for enhanced damages. 1 The principal problem with Seagate’s two-part test is that it requires a finding of objective recklessness in every case before district courts may award enhanced damages. Such a threshold requirement excludes from discretionary punishment many of the most culpable offenders, such as the “wanton and malicious pirate” who intentionally infringes another’s patent—with no doubts about its validity or any notion of a defense—for no purpose other than to steal the patentee’s business. Seymour, 16 How., at 488. Under Seagate, a district court may not even consider enhanced damages for such a pirate, unless the court first determines that his infringement was “objectively” reckless. In the context of such deliberate wrongdoing, how-ever, it is not clear why an independent showing of objective recklessness—by clear and convincing evidence, no less—should be a prerequisite to enhanced damages. Our recent decision in Octane Fitness arose in a different context but points in the same direction. In that case we considered §285 of the Patent Act, which allows district courts to award attorney’s fees to prevailing parties in “exceptional” cases. 35 U. S. C. §285. The Federal Circuit had adopted a two-part test for determining when a case qualified as exceptional, requiring that the claim asserted be both objectively baseless and brought in subjective bad faith. We rejected that test on the ground that a case presenting “subjective bad faith” alone could “sufficiently set itself apart from mine-run cases to warrant a fee award.” 572 U. S., at ___ (slip op., at 9). So too here. The subjective willfulness of a patent infringer, intentional or knowing, may warrant enhanced damages, without regard to whether his infringement was objectively reckless. The Seagate test aggravates the problem by making dispositive the ability of the infringer to muster a reasonable (even though unsuccessful) defense at the infringement trial. The existence of such a defense insulatesthe infringer from enhanced damages, even if he did not act on the basis of the defense or was even aware of it. Under that standard, someone who plunders a patent—in-fringing it without any reason to suppose his conduct is arguably defensible—can nevertheless escape any come-uppance under §284 solely on the strength of his attorney’s ingenuity. But culpability is generally measured against the knowledge of the actor at the time of the challenged conduct. See generally Restatement (Second) of Torts §8A (1965) (“intent” denotes state of mind in which “the actor desires to cause consequences of his act” or “believes” them to be “substantially certain to result from it”); W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts §34, p. 212 (5th ed. 1984) (describing willful, wanton, and reckless as “look[ing] to the actor’s real or supposed state of mind”); see also Kolstad v. American Dental Assn., 527 U. S. 526, 538 (1999) (“Most often . . . eligibility for punitive awards is characterized in terms of a defendant’s motive or intent”). In Safeco Ins. Co. of America v. Burr, 551 U. S. 47 (2007) , we stated that a person is reckless if he acts “knowing or having reason to know of facts which would lead a reasonable man to realize” his actions are unreasonably risky. Id., at 69 (emphasis added and internal quotation marks omitted). The Court found that the defendant had not recklessly violated the Fair Credit Reporting Act because the defendant’s interpretation had “a foundation in the statutory text” and the defendant lacked “the benefit of guidance from the courts of appeals or the Federal Trade Commission” that “might have warned it away from the view it took.” Id., at 69–70. Nothing in Safeco suggests that we should look to facts that the defendant neither knew nor had reason to know at the time he acted.[1]* Section 284 allows district courts to punish the full range of culpable behavior. Yet none of this is to say that enhanced damages must follow a finding of egregious misconduct. As with any exercise of discretion, courts should continue to take into account the particular circumstances of each case in deciding whether to award damages, and in what amount. Section 284 permits district courts to exercise their discretion in a manner free from the inelastic constraints of the Seagate test. Consistent with nearly two centuries of enhanced damages under patent law, however, such punishment should generally be reserved for egregious cases typified by willful misconduct. 2 The Seagate test is also inconsistent with §284 because it requires clear and convincing evidence to prove recklessness. On this point Octane Fitness is again instructive. There too the Federal Circuit had adopted a clear and convincing standard of proof, for awards of attorney’s fees under §285 of the Patent Act. Because that provision supplied no basis for imposing such a heightened standard of proof, we rejected it. See Octane Fitness, 572 U. S., at ___ (slip op., at 11). We do so here as well. Like §285, §284 “imposes no specific evidentiary burden, much less such a high one.” Ibid. And the fact that Congress expressly erected a higher standard of proof elsewhere in the Patent Act, see 35 U. S. C. §273(b), but not in §284, is telling. Furthermore, nothing in historical practice supports a heightened standard. As we explained in Octane Fitness, “patent-infringement litigation has always been governed by a preponderance of the evidence standard.” 572 U. S., at ___ (slip op., at 11). Enhanced damages are no exception. 3 Finally, because we eschew any rigid formula for awarding enhanced damages under §284, we likewise reject the Federal Circuit’s tripartite framework for appellate review. In Highmark Inc. v. Allcare Health Management System, Inc., 572 U. S. ___ (2014), we built on our Octane Fitness holding to reject a similar multipart standard of review. Because Octane Fitness confirmed district court discretion to award attorney fees, we concluded that such decisions should be reviewed for abuse of discretion. Highmark, 572 U. S., at ___ (slip op., at 1). The same conclusion follows naturally from our holding here. Section 284 gives district courts discretion in meting out enhanced damages. It “commits the determination” whether enhanced damages are appropriate “to the discretion of the district court” and “that decision is to be reviewed on appeal for abuse of discretion.” Id., at ___ (slip op., at 4). That standard allows for review of district court decisions informed by “the considerations we have identified.” Octane Fitness, 572 U. S., at ___ (slip op., at 8) (internal quotation marks omitted). The appellate review framework adopted by the Federal Circuit reflects a concern that district courts may award enhanced damages too readily, and distort the balance between the protection of patent rights and the interest in technological innovation. Nearly two centuries of exercising discretion in awarding enhanced damages in patent cases, however, has given substance to the notion that there are limits to that discretion. The Federal Circuit should review such exercises of discretion in light of the longstanding considerations we have identified as having guided both Congress and the courts. III For their part, respondents argue that Congress ratified the Seagate test when it passed the America Invents Act of 2011 and reenacted §284 without pertinent change. See Brief for Respondents in No. 14–1513 27 (citing Lorillard v. Pons, 434 U. S. 575, 580 (1978) ). But the language Congress reenacted unambiguously confirmed discretion in the district courts. Congress’s retention of §284 could just as readily reflect an intent that enhanced damages be awarded as they had been for nearly two centuries, through the exercise of such discretion, informed by settled practices. Respondents point to isolated snippets of legislative history referring to Seagate as evidence of congressional endorsement of its framework, but other morsels—such as Congress’s failure to adopt a proposed codification similar to Seagate—point in the opposite direction. See, e.g., H. R. 1260, 111th Cong., 1st Sess. §5(e) (2009). Respondents also seize on an addition to the Act addressing opinions of counsel. Section 298 provides that “[t]he failure of an infringer to obtain the advice of counsel” or “the failure of the infringer to present such advice to the court or jury, may not be used to prove that the accused infringer willfully infringed.” 35 U. S. C. §298. Respondents contend that the reference to willfulness reflects an endorsement of Seagate’s willfulness test. But willfulness has always been a part of patent law, before and after Seagate. Section 298 does not show that Congress ratified Seagate’s particular conception of willfulness. Rather, it simply addressed the fallout from the Federal Circuit’s opinion in Underwater Devices Inc. v. Morrison-Knudsen Co., 717 F. 2d 1380 (1983), which had imposed an “affirmative duty” to obtain advice of counsel prior to initiating any possible infringing activity, id., at 1389–1390. See, e.g., H. R. Rep. No. 112–98, pt. 1, p. 53 (2011). At the end of the day, respondents’ main argument for retaining the Seagate test comes down to a matter of policy. Respondents and their amici are concerned that allowing district courts unlimited discretion to award up to treble damages in infringement cases will impede innovation as companies steer well clear of any possible interference with patent rights. They also worry that the ready availability of such damages will embolden “trolls.” Trolls, in the patois of the patent community, are entities that hold patents for the primary purpose of enforcing them against alleged infringers, often exacting outsized licensing fees on threat of litigation. Respondents are correct that patent law reflects “a careful balance between the need to promote innovation” through patent protection, and the importance of facilitating the “imitation and refinement through imitation” that are “necessary to invention itself and the very lifeblood of a competitive economy.” Bonito Boats, Inc. v. Thunder Craft Boats, Inc., 489 U. S. 141, 146 (1989) . That balance can indeed be disrupted if enhanced damages are awarded in garden-variety cases. As we have explained, however, they should not be. The seriousness of respondents’ policy concerns cannot justify imposing an artificial construct such as the Seagate test on the discretion conferred under §284. * * * Section 284 gives district courts the discretion to award enhanced damages against those guilty of patent infringement. In applying this discretion, district courts are “to be guided by [the] sound legal principles” developed over nearly two centuries of application and interpretation of the Patent Act. Martin, 546 U. S., at 139 (internal quotation marks omitted). Those principles channel the exercise of discretion, limiting the award of enhanced damages to egregious cases of misconduct beyond typical infringement. The Seagate test, in contrast, unduly confines the ability of district courts to exercise the discretion conferred on them. Because both cases before us were decided under the Seagate framework, we vacate the judgments of the Federal Circuit and remand the cases for proceedings consistent with this opinion. It is so ordered.Notes 1 * Respondents invoke a footnote in Safeco where we explained that in considering whether there had been a knowing or reckless violation of the Fair Credit Reporting Act, a showing of bad faith was not relevant absent a showing of objective recklessness. See 551 U. S., at 70, n. 20. But our precedents make clear that “bad-faith infringement” is an independent basis for enhancing patent damages. Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U. S. 476, 508 (1964) ; see supra, at 2–5, 9–10; see also Safeco, 551 U. S., at 57 (noting that “ ‘willfully’ is a word of many meanings whose construction is often dependent on the context in which it appears” (some internal quotation marks omitted)).

578.US.2015_14-232

After the 2010 census, Arizona’s independent redistricting commission (Commission), comprising two Republicans, two Democrats, and one Independent, redrew Arizona’s legislative districts, with guidance from legal counsel, mapping specialists, a statistician, and a Voting Rights Act specialist. The initial plan had a maximum population deviation from absolute equality of districts of 4.07%, but the Commission adopted a revised plan with an 8.8% deviation on a 3-to-2 vote, with the Republican members dissenting. After the Department of Justice approved the revised plan as consistent with the Voting Rights Act, appellants filed suit, claiming that the plan’s population variations were inconsistent with the Fourteenth Amendment. A three-judge Federal District Court entered judgment for the Commission, concluding that the “deviations were primarily a result of good-faith efforts to comply with the Voting Rights Act . . . even though partisanship played some role.” Held: The District Court did not err in upholding Arizona’s redistricting plan. Pp. 3–11. (a) The Fourteenth Amendment’s Equal Protection Clause requires States to “make an honest and good faith effort to construct [legislative] districts . . . as nearly of equal population as is practicable,” Reynolds v. Sims, 377 U. S. 533 , but mathematical perfection is not required. Deviations may be justified by “legitimate considerations,” id., at 579, including “traditional districting principles such as compactness [and] contiguity,” Shaw v. Reno, 509 U. S. 630 , as well as a state interest in maintaining the integrity of political subdivisions, Mahan v. Howell, 410 U. S. 315 , a competitive balance among political parties, Gaffney v. Cummings, 412 U. S. 735 , and, before Shelby County v. Holder, 570 U. S. ___, compliance with §5 of the Voting Rights Act. It was proper for the Commission to proceed on the last basis here. In addition, “minor deviations from mathematical equality”—i.e., deviations “under 10%,” Brown v. Thomson, 462 U. S. 835 —do not, by themselves, “make out a prima facie case of invidious discrimination under the Fourteenth Amendment [requiring] justification by the State,” Gaffney, supra, at 745. Because the deviation here is under 10%, appellants cannot rely upon the numbers to show a constitutional violation. Instead, they must show that it is more probable than not that the deviation reflects the predominance of illegitimate reapportionment factors rather than “legitimate considerations.” Pp. 3–5. (b) Appellants have failed to meet that burden here, where the record supports the District Court’s conclusion that the deviations predominantly reflected Commission efforts to achieve compliance with the Voting Rights Act, not to secure political advantage for the Democratic Party. To meet the Voting Rights Act’s nonretrogression requirement, a new plan, when compared to the current plan (benchmark plan), must not diminish the number of districts in which minority groups can “elect their preferred candidates of choice” (ability-to-elect districts). A State can obtain legal assurance that it has satisfied this requirement if it submits its proposed plan to the Justice Department and the Department does not object to the plan. The record shows that the Commission redrew the initial map to ensure that the plan had 10 ability-to-elect districts, the same number as the benchmark plan. But after a statistician reported that the Justice Department still might not agree with the plan, the Commission changed additional boundaries, causing District 8, a Republican leaning district, to become more politically competitive. Because this record well supports the District Court’s finding that the Commission was trying to comply with the Voting Rights Act, appellants have not shown that it is more probable than not that illegitimate considerations were the predominant motivation for the deviations. They have thus failed to show that the plan violates the Equal Protection Clause. Pp. 5–9. (c) Appellants’ additional arguments are unpersuasive. While Arizona’s Democratic-leaning districts may be somewhat underpopulated and its Republican-leaning districts somewhat overpopulated, these variations may reflect only the tendency of Arizona’s 2010 minority populations to vote disproportionately for Democrats and thus can be explained by the Commission’s efforts to maintain at least 10 ability-to-elect districts. Cox v. Larios, 542 U. S. 947 , in which the Court affirmed a District Court’s conclusion that a Georgia reapportionment plan violated the Equal Protection Clause where its deviation, though less than 10%, resulted from the use of illegitimate factors, is inapposite because appellants have not carried their burden of showing the use of illegitimate factors here. And because Shelby County was decided after Arizona’s plan was created, it has no bearing on the issue whether the State’s attempt to comply with the Voting Rights Act is a legitimate state interest. Pp. 9–11. 993 F. Supp. 2d 1042, affirmed. Breyer, J., delivered the opinion for a unanimous Court.

Appellants, a group of Arizona voters, challenge a re-districting plan for the State’s legislature on the ground that the plan’s districts are insufficiently equal in population. See Reynolds v. Sims, 377 U. S. 533, 577 (1964) . Because the maximum population deviation between the largest and the smallest district is less than 10%, the appellants cannot simply rely upon the numbers to show that the plan violates the Constitution. See Brown v. Thomson, 462 U. S. 835, 842 (1983) . Nor have appellants adequately supported their contentions with other evidence. We consequently affirm a 3-judge Federal District Court decision upholding the plan. I In 2000, Arizona voters, using the initiative process, amended the Arizona Constitution to provide for an independent redistricting commission. See Arizona State Legislature v. Arizona Independent Redistricting Comm’n, 576 U. S. ___, ___ (2015) (slip op., at 35) (upholding the amendment as consistent with federal constitutional and statutory law). Each decade, the Arizona Commission on Appellate Court Appointments creates three slates of individuals: one slate of 10 Republicans, one slate of 10 Democrats, and one slate of 5 individuals not affiliated with any political party. The majority and minority leader of the Arizona Legislature each select one Redistricting Commission member from the first two lists. These four selected individuals in turn choose one member from the third, nonpartisan list. See Ariz. Const., Art. IV, pt. 2, §§1(5)–(8). Thus, the membership of the Commission consists of two Republicans, two Democrats, and one independent. After each decennial census, the Commission redraws Arizona’s 30 legislative districts. The first step in the process is to create “districts of equal population in a grid-like pattern across the state.” §1(14). It then adjusts the grid to “the extent practicable” in order to take into account the need for population equality; to maintain geographic compactness and continuity; to show respect for “communities of interest”; to follow locality boundaries; and to use “visible geographic features” and “undivided . . . tracts.” §§1(14)(B)–(E). The Commission will “favo[r]” political “competitive[ness]” as long as its efforts to do so “create no significant detriment to the other goals.” Id., §1(14)(F). Finally, it must adjust boundaries “as necessary” to comply with the Federal Constitution and with the federal Voting Rights Act. §1(14)(A). After the 2010 census, the legislative leadership selected the Commission’s two Republican and two Democratic members, who in turn selected an independent member, Colleen Mathis. Mathis was then elected chairwoman. The Commission hired two counsel, one of whom they thought of as leaning Democrat and one as leaning Republican. It also hired consultants, including mapping specialists, a statistician, and a Voting Rights Act specialist. With the help of its staff, it drew an initial plan, based upon the gridlike map, with district boundaries that produced a maximum population deviation (calculated as the difference between the most populated and least populated district) of 4.07%. After changing several boundaries, including those of Districts 8, 24, and 26, the Commission adopted a revised plan by a vote of 3 to 2, with the two Republican members voting against it. In late April 2012, the Department of Justice approved the plan as consistent with the Voting Rights Act. The next day, appellants filed this lawsuit, primarily claiming that the plan’s population variations were inconsistent with the Fourteenth Amendment. A 3-judge Federal District Court heard the case. See 28 U. S. C. §2284(a) (providing for the convention of such a court whenever an action is filed challenging the constitutional-ity of apportionment of legislative districts). After a 5-day bench trial, the court, by a vote of 2 to 1, entered judgment for the Commission. The majority found that “the population deviations were primarily a result of good-faith efforts to comply with the Voting Rights Act . . . even though partisanship played some role.” 993 F. Supp. 2d 1042, 1046 (Ariz. 2014). Appellants sought direct review in this Court. See 28 U. S. C. §1253. We noted probable jurisdiction on June 30, 2015, and we now affirm. II A The Fourteenth Amendment’s Equal Protection Clause requires States to “make an honest and good faith effort to construct [legislative] districts . . . as nearly of equal population as is practicable.” Reynolds, 377 U. S., at 577. The Constitution, however, does not demand mathematical perfection. In determining what is “practicable,” we have recognized that the Constitution permits deviation when it is justified by “legitimate considerations incident to the effectuation of a rational state policy.” Id., at 579. In related contexts, we have made clear that in addition to the “traditional districting principles such as compactness [and] contiguity,” Shaw v. Reno, 509 U. S. 630, 647 (1993) , those legitimate considerations can include a state interest in maintaining the integrity of political subdivisions, Mahan v. Howell, 410 U. S. 315, 328 (1973) , or the competitive balance among political parties, Gaffney v. Cummings, 412 U. S. 735, 752 (1973) . In cases decided before Shelby County v. Holder, 570 U. S. ___ (2013), Members of the Court expressed the view that compliance with §5 of the Voting Rights Act is also a legitimate state consideration that can justify some deviation from perfect equality of population. See League of United Latin American Citizens v. Perry, 548 U. S. 399, 518 (2006) (Scalia, J., concurring in judgment in part and dissenting in part, joined in relevant part by Roberts, C. J., Thomas & Alito, JJ.); id., at 475, n. 12 (Stevens, J., concurring in part and dissenting in part, joined in relevant part by Breyer, J.); id., at 485 n. 2 (Souter, J., concurring in part and dissenting in part, joined by Ginsburg, J.); see also Vieth v. Jubelirer, 541 U. S. 267, 284 (2004) (plurality opinion) (listing examples of traditional redistricting criteria, including “compliance with requirements of the [Voting Rights Act]”). It was proper for the Commission to proceed on that basis here. We have further made clear that “minor deviations from mathematical equality” do not, by themselves, “make out a prima facie case of invidious discrimination under the Fourteenth Amendment so as to require justification by the State.” Gaffney, supra, at 745. We have defined as “minor deviations” those in “an apportionment plan with a maximum population deviation under 10%.” Brown, 462 U. S., at 842. And we have refused to require States to justify deviations of 9.9%, White v. Regester, 412 U. S. 755, 764 (1973) , and 8%, Gaffney, 412 U. S., at 751. See also Fund for Accurate and Informed Representation, Inc. v. Weprin, 506 U. S. 1017 (1992) (summarily affirming a District Court’s finding that there was no prima facie case where the maximum population deviation was 9.43%). In sum, in a case like this one, those attacking a state-approved plan must show that it is more probable than not that a deviation of less than 10% reflects the predominance of illegitimate reapportionment factors rather than the “legitimate considerations” to which we have referred in Reynolds and later cases. Given the inherent difficulty of measuring and comparing factors that may legitimately account for small deviations from strict mathematical equality, we believe that attacks on deviations under 10% will succeed only rarely, in unusual cases. And we are not surprised that the appellants have failed to meet their burden here. B Appellants’ basic claim is that deviations in their apportionment plan from absolute equality of population reflect the Commission’s political efforts to help the Democratic Party. We believe that appellants failed to prove this claim because, as the district court concluded, the deviations predominantly reflected Commission efforts to achieve compliance with the federal Voting Rights Act, not to secure political advantage for one party. Appellants failed to show to the contrary. And the record bears out this conclusion. Cf. Anderson v. Bessemer City, 470 U. S. 564, 573 (1985) (explaining that a district court’s factual finding as to whether discrimination occurred will not be set aside by an appellate court unless clearly erroneous). The Voting Rights Act, among other things, forbids the use of new reapportionment plans that “would lead to a retrogression in the position of racial minorities with respect to their effective exercise of the electoral franchise.” Reno v. Bossier Parish School Bd., 520. U. S. 471, 478 (1997). A plan leads to impermissible retrogression when, compared to the plan currently in effect (typically called a “benchmark plan”), the new plan diminishes the number of districts in which minority groups can “elect their preferred candidates of choice” (often called “ability-to-elect” districts). See 52 U. S. C. §10304(b). A State can obtain legal assurance that it has satisfied the non-retrogression requirement if it submits its proposed plan to the Federal Department of Justice, and the Department does not object to the plan within 60 days. See 28 C. F. R. §§51.9, 51.52(b) (2015). While Shelby County struck down the §4(b) coverage formula, that decision came after the maps in this case were drawn. The record in this case shows that the gridlike map that emerged after the first step of the redistricting process had a maximum population deviation from absolute equality of districts of 4.07%. After consulting with their Voting Rights Act expert, their mapping consultant, and their statisticians, all five Commissioners agreed that they must try to obtain Justice Department Voting Rights Act “preclearance” and that the former benchmark plan contained 10 ability-to-elect districts. They consequently set a goal of 10 such districts for the new plan. They then went through an iterative process, involving further consultation, to adjust the plan’s initial boundaries in order to enhance minority voting strength. In October 2011 (by a vote of 4 to 1), they tentatively approved a draft plan with adjusted boundaries. They believed it met their goal of 10 ability-to-elect districts. And they published the plan for public comment. In the meantime, however, the Commission received a report from one of its statisticians suggesting that the Department of Justice might not agree that the new proposed plan contained 10 ability-to-elect districts. It was difficult to know for certain because the Justice Department did not tell States how many ability-to-elect districts it believed were present in a benchmark plan, and neither did it typically explain precisely and specifically how it would calculate the number that exist in a newly submitted plan. See 76 Fed. Reg. 7470–7471 (2011). At the same time, the ability-to-elect analysis was complex, involving more than simply adding up census figures. The Department of Justice instead conducted a “functional analysis of the electoral behavior within the particular . . . election district,” id., at 7471, and so might, for example, count as ability-to-elect districts “crossover” districts in which white voters combine their votes with minorities, see Bartlett v. Strickland, 556 U. S. 1 –14 (2009). Its calculations might take into account group voting patterns, electoral participation, election history, and voter turnout. See 76 Fed. Reg., 7471. The upshot was not randomdecision-making but the process did create an inevitable degree of uncertainty. And that uncertainty could lead a redistricting commission, as it led Arizona’s, to make serious efforts to make certain that the districts it believed were ability-to-elect districts did in fact meet the criteria that the Department might reasonably apply. Cf. Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___ (2015) (slip op., at 22) (“The law cannot insist that a state legislature, when redistricting, determine precisely what percent minority population §5 demands [because] the standards of §5 are complex . . . . [To do so would] lay a trap for an unwary legislature, condemning its redistricting plan as either . . . unconstitutional racial gerrymandering [or] . . . retrogressive under §5”). As a result of the statistician’s report, the Commission became concerned about certain of its proposed boundaries. One of the Commission’s counsel advised that it would be “prudent to stay the course in terms of the ten districts that are in the draft map and look to . . . strengthen them if there is a way to strengthen them.” 993 F. Supp. 2d, at 1058 (internal quotation marks omitted). Subsequently, the Commission adopted several changes to the boundaries of Districts 24 and 26. It reduced the populations of those districts, thereby increasing the percentage of Hispanic voters in each. The Commission approved these changes unanimously. Changes in the boundaries of District 8, however, proved more controversial. District 8 leaned Republican. A Democrat-appointed Commissioner asked the mapping specialist to look into modifications that might make District 8 politically more competitive. The specialist returned with a draft that shifted the boundary line between District 8 and District 11 so as to keep several communities with high minority populations together in District 8. The two Republican-appointed Commissioners objected that doing so would favor Democrats by “hyperpacking” Republicans into other districts; they added that the Commission should either favor political competitiveness throughout the State or not at all. Id., at 1059 (internal quotation marks omitted). The Democrat-appointed proponent of the change replied that District 8 had historically provided minority groups a good opportunity to elect their candidate of choice—an opportunity that the changes would preserve. The Voting Rights Act specialist then said that by slightly increasing District 8’s minority population, the Commission might be able to claim an 11th ability-to-elect district; and that fact would “unquestionably enhance the submission and enhance chances for preclearance.” Ibid. (internal quotation marks omitted). The Commission’s counsel then added that having another possible ability-to-elect district could be helpful because District 26 was not as strong an ability-to-elect district as the others. See ibid. Only then, after the counsel and consultants argued for District 8 changes for the sake of Voting Rights Act preclearance, did Chairwoman Mathis support those changes. On that basis, the Commission ultimately approved the changes to District 8 by a vote of 3 to 2 (with the two Republican-appointed commissioners dissenting). The total population deviation among districts in this final map was 8.8%. While the Commission ultimately concluded that District 8 was not a true ability-to-elect district, the State’s submission to the Department of Justice cited the changes to District 8 in support of the argument for preclearance. On April 26, 2012, the Department of Justice precleared the submitted plan. On the basis of the facts that we have summarized, the District Court majority found that “the population deviations were primarily a result of good-faith efforts to comply with the Voting Rights Act . . . even though partisanship played some role.” 993 F. Supp. 2d, at 1046. This conclusion was well supported in the record. And as a result, appellants have not shown that it is more probable than not that illegitimate considerations were the predominant motivation behind the plan’s deviations from mathematically equal district populations—deviations that were under 10%. Consequently, they have failed to show that the Commission’s plan violates the Equal Protection Clause as interpreted in Reynolds and subsequent cases. C The appellants make three additional arguments. First, they support their claim that the plan reflects unreason-able use of partisan considerations by pointing to the fact that almost all the Democratic-leaning districts are somewhat underpopulated and almost all the Republican-leaning districts are somewhat overpopulated. That is likely true. See 993 F. Supp. 2d, at 1049 (providing a chart with percentage deviation figures by district). But that fact may well reflect the tendency of minority populations in Arizona in 2010 to vote disproportionately for Democrats. If so, the variations are explained by the Commission’s efforts to maintain at least 10 ability-to-elect districts. The Commission may have relied on data from its statisticians and Voting Rights Act expert to create districts tailored to achieve preclearance in which minority voters were a larger percentage of the district population. That might have necessitated moving other voters out of those districts, thereby leaving them slightly underpopulated. The appellants point to nothing in the record to suggest the contrary. Second, the appellants point to Cox v. Larios, 542 U. S. 947 (2004) , in which we summarily affirmed a district court’s judgment that Georgia’s reapportionment of representatives to state legislative districts violated the Equal Protection Clause, even though the total population deviation was less than 10%. In Cox, however, unlike the present case, the district court found that those attacking the plan had shown that it was more probable than not that the use of illegitimate factors significantly explained deviations from numerical equality among districts. The district court produced many examples showing that population deviation as well as the shape of many districts “did not result from any attempt to create districts that were compact or contiguous, or to keep counties whole, or to preserve the cores of prior districts.” Id., at 949. No legitimate purposes could explain them. It is appellants’ inability to show that the present plan’s deviations and boundary shapes result from the predominance of simi-larly illegitimate factors that makes Cox inapposite here. Even assuming, without deciding, that partisanship is an illegitimate redistricting factor, appellants have not carried their burden. Third, appellants point to Shelby County v. Holder, 570 U. S. ___ (2013), in which this Court held unconstitutional sections of the Voting Rights Act that are relevant to this case. Appellants contend that, as a result of that holding, Arizona’s attempt to comply with the Act could not have been a legitimate state interest. The Court decided Shelby County, however, in 2013. Arizona created the plan at issue here in 2010. At the time, Arizona was subject tothe Voting Rights Act, and we have never suggested the contrary. * * * For these reasons the judgment of the District Court is affirmed. It is so ordered.

578.US.2015_14-1280

Petitioner Heffernan was a police officer working in the office of Paterson, New Jersey’s chief of police. Both the chief of police and Heffernan’s supervisor had been appointed by Paterson’s incumbent mayor, who was running for re-election against Lawrence Spagnola, a good friend of Heffernan’s. Heffernan was not involved in Spagnola’s campaign in any capacity. As a favor to his bedridden mother, Heffernan agreed to pick up and deliver to her a Spagnola campaign yard sign. Other police officers observed Heffernan speaking to staff at a Spagnola distribution point while holding the yard sign. Word quickly spread throughout the force. The next day, Heffernan’s supervisors demoted him from detective to patrol officer as punishment for his “overt involvement” in Spagnola’s campaign. Heffernan filed suit, claiming that the police chief and the other respondents had demoted him because, in their mistaken view, he had engaged in conduct that constituted protected speech. They had thereby “depriv[ed]” him of a “right . . . secured by the Constitution.” 42 U. S. C. §1983. The District Court, however, found that Heffernan had not been deprived of any constitutionally protected right because he had not engaged in any First Amendment conduct. Affirming, the Third Circuit concluded that Heffernan’s claim was actionable under §1983 only if his employer’s action was prompted by Heffernan’s actual, rather than his perceived, exercise of his free-speech rights. Held: 1. When an employer demotes an employee out of a desire to prevent the employee from engaging in protected political activity, the employee is entitled to challenge that unlawful action under the First Amendment and §1983 even if, as here, the employer’s actions are based on a factual mistake about the employee’s behavior. To answer the question whether an official’s factual mistake makes a critical legal difference, the Court assumes that the activities that Heffernan’s supervisors mistakenly thought he had engaged in are of a kind that they cannot constitutionally prohibit or punish. Section 1983 does not say whether the “right” protected primarily focuses on the employee’s actual activity or on the supervisor’s motive. Neither does precedent directly answer the question. In Connick v. Myers, 461 U. S. 138 , Garcetti v. Ceballos, 547 U. S. 410 , and Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U. S. 563 , there were no factual mistakes: The only question was whether the undisputed reason for the adverse action was in fact protected by the First Amendment. However, in Waters v. Churchill, 511 U. S. 661 , a government employer’s adverse action was based on a mistaken belief that an employee had not engaged in protected speech. There, this Court determined that the employer’s motive, and particularly the facts as the employer reasonably understood them, mattered in determining that the employer had not violated the First Amendment. The government’s motive likewise matters here, where respondents demoted Heffernan on the mistaken belief that he had engaged in protected speech. A rule of law finding liability in these circumstances tracks the First Amendment’s language, which focuses upon the Government’s activity. Moreover, the constitutional harm—discouraging employees from engaging in protected speech or association—is the same whether or not the employer’s action rests upon a factual mistake. Finally, a rule of law imposing liability despite the employer’s factual mistake is not likely to impose significant extra costs upon the employer, for the employee bears the burden of proving an improper employer motive. Pp. 3–8. 2. For the purposes of this opinion, the Court has assumed that Heffernan’s employer demoted him out of an improper motive. However, the lower courts should decide in the first instance whether respondents may have acted under a neutral policy prohibiting police officers from overt involvement in any political campaign and whether such a policy, if it exists, complies with constitutional standards. P. 8. 777 F. 3d 147, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

The First Amendment generally prohibits government officials from dismissing or demoting an employee because of the employee’s engagement in constitutionally protected political activity. See Elrod v. Burns, 427 U. S. 347 (1976) ; Branti v. Finkel, 445 U. S. 507 (1980) ; but cf. Civil Service Comm’n v. Letter Carriers, 413 U. S. 548, 564 (1973) . In this case a government official demoted an employee because the official believed, but incorrectly believed, that the employee had supported a particular candidate for mayor. The question is whether the official’s factual mistake makes a critical legal difference. Even though the employee had not in fact engaged in protected political activity, did his demotion “deprive” him of a “right . . . secured by the Constitution”? 42 U. S. C. §1983. We hold that it did. I To decide the legal question presented, we assume the following, somewhat simplified, version of the facts: In 2005, Jeffrey Heffernan, the petitioner, was a police officer in Paterson, New Jersey. He worked in the office of the Chief of Police, James Wittig. At that time, the mayor of Paterson, Jose Torres, was running for reelection against Lawrence Spagnola. Torres had appointed to their current positions both Chief Wittig and a subordinate who directly supervised Heffernan. Heffernan was a good friend of Spagnola’s. During the campaign, Heffernan’s mother, who was bedridden, asked Heffernan to drive downtown and pick up a large Spagnola sign. She wanted to replace a smaller Spagnola sign, which had been stolen from her front yard. Heffernan went to a Spagnola distribution point and picked up the sign. While there, he spoke for a time to Spagnola’s campaign manager and staff. Other members of the police force saw him, sign in hand, talking to campaign workers. Word quickly spread throughout the force. The next day, Heffernan’s supervisors demoted Heffernan from detective to patrol officer and assigned him to a “walking post.” In this way they punished Heffernan for what they thought was his “overt involvement” in Spag-nola’s campaign. In fact, Heffernan was not involved in the campaign but had picked up the sign simply to help his mother. Heffernan’s supervisors had made a factual mistake. Heffernan subsequently filed this lawsuit in federal court. He claimed that Chief Wittig and the other respondents had demoted him because he had engaged in conduct that (on their mistaken view of the facts) constituted protected speech. They had thereby “depriv[ed]” him of a “right . . . secured by the Constitution.” Rev. Stat. §1979, 42 U. S. C. §1983. The District Court found that Heffernan had not engaged in any “ First Amendment conduct,” 2 F. Supp. 3d 563, 580 (NJ 2014); and, for that reason, the respondents had not deprived him of any constitutionally protected right. The Court of Appeals for the Third Circuit affirmed. It wrote that “a free-speech retaliation claim is actionable under §1983 only where the adverse action at issue was prompted by an employee’s actual, rather than perceived, exercise of constitutional rights.” 777 F. 3d 147, 153 (2015) (citing Ambrose v. Robinson, 303 F. 3d 488, 496 (CA3 2002); emphasis added). Heffernan filed a petition for certiorari. We agreed to decide whether the Third Circuit’s legal view was correct. Compare 777 F. 3d, at 153 (case below), with Dye v. Office of Racing Comm’n, 702 F. 3d 286, 300 (CA6 2012) (similar factual mistake does not affect the validity of the government employee’s claim). II With a few exceptions, the Constitution prohibits a government employer from discharging or demoting an employee because the employee supports a particular political candidate. See Elrod v. Burns, supra; Branti v. Finkel, supra. The basic constitutional requirement reflects the First Amendment’s hostility to government action that “prescribe[s] what shall be orthodox in politics.” West Virginia Bd. of Ed. v. Barnette, 319 U. S. 624, 642 (1943) . The exceptions take account of “practical realities” such as the need for “efficiency” and “effective[ness]” in government service. Waters v. Churchill, 511 U. S. 661, 672, 675 (1994) ; see also Civil Service Comm’n, supra, at 564 (neutral and appropriately limited policy may prohibit government employees from engaging in partisan activity), and Branti, supra, at 518 (political affiliation requirement permissible where affiliation is “an appropriate requirement for effective performance of the public office involved”). In order to answer the question presented, we assume that the exceptions do not apply here. But see infra, at 8. We assume that the activities that Heffernan’s supervisors thought he had engaged in are of a kind that they cannot constitutionally prohibit or punish, see Rutan v. Republican Party of Ill., 497 U. S. 62, 69 (1990) (“joining, working for or contributing to the political party and candidates of their own choice”), but that the supervisors were mistaken about the facts. Heffernan had not engaged in those protected activities. Does Heffernan’s constitutional case consequently fail? The text of the relevant statute does not answer the question. The statute authorizes a lawsuit by a person “depriv[ed]” of a “right . . . secured by the Constitution.” 42 U. S. C. §1983. But in this context, what precisely is that “right?” Is it a right that primarily focuses upon (the employee’s) actual activity or a right that primarily fo-cuses upon (the supervisor’s) motive, insofar as that motive turns on what the supervisor believes that activity to be? The text does not say. Neither does precedent directly answer the question. In some cases we have used language that suggests the “right” at issue concerns the employee’s actual activity. In Connick v. Myers, 461 U. S. 138 (1983) , for example, we said that a court should first determine whether the plaintiff spoke “ ‘as a citizen’ ” on a “ ‘matter[] of public concern,’ ” id., at 143. We added that, if the employee has not engaged in what can “be fairly characterized as constituting speech on a matter of public concern, it is unnecessary for us to scrutinize the reasons for her discharge.” Id., at 146. We made somewhat similar statements in Garcetti v. Ceballos, 547 U. S. 410, 418 (2006) , and Pickering v. Board of Ed. of Township High School Dist. 205, Will Cty., 391 U. S. 563 (1968) . These cases, however, did not present the kind of question at issue here. In Connick, for example, no factual mistake was at issue. The Court assumed that both the employer and the employee were at every stage in agreement about the underlying facts: that the employer dismissed the employee because of her having circulated within the office a document that criticized how the office was being run (that she had in fact circulated). The question was whether the circulation of that document amounted to constitutionally protected speech. If not, the Court need go no further. Neither was any factual mistake at issue in Pickering. The Court assumed that both the employer (a school board) and the employee understood the cause for dismissal, namely, a petition that the employee had indeed circulated criticizing his employer’s practices. The question concerned whether the petition was protectedspeech. Garcetti is substantially similar. In each of these cases, the only way to show that the employer’s motive was unconstitutional was to prove that the controver-sial statement or activity—in each case the undisputed reason for the firing—was in fact protected by the First Amendment. Waters v. Churchill, 511 U. S. 661 (1994) , is more to the point. In that case the Court did consider the consequences of an employer mistake. The employer wrongly, though reasonably, believed that the employee had spoken only on personal matters not of public concern, and the employer dismissed the employee for having engaged in that unprotected speech. The employee, however, had in fact used words that did not amount to personal “gossip” (as the employer believed) but which focused on matters of public concern. The Court asked whether, and how, the employer’s factual mistake mattered. The Court held that, as long as the employer (1) had reasonably believed that the employee’s conversation had involved personal matters, not matters of public concern, and (2) had dismissed the employee because of that mistaken belief, the dismissal did not violate the First Amendment. Id., at 679–680. In a word, it was the employer’s motive, and in particular the facts as the employer reasonably understood them, that mattered. In Waters, the employer reasonably but mistakenly thought that the employee had not engaged in protected speech. Here the employer mistakenly thought that the employee had engaged in protected speech. If the employer’s motive (and in particular the facts as the employer reasonably understood them) is what mattered in Waters, why is the same not true here? After all, in the law, what is sauce for the goose is normally sauce for the gander. We conclude that, as in Waters, the government’s reason for demoting Heffernan is what counts here. When an employer demotes an employee out of a desire to prevent the employee from engaging in political activity that the First Amendment protects, the employee is entitled to challenge that unlawful action under the First Amendment and 42 U. S. C. §1983—even if, as here, the employer makes a factual mistake about the employee’s behavior. We note that a rule of law finding liability in these circumstances tracks the language of the First Amend-ment more closely than would a contrary rule. Unlike, say, the Fourth Amendment, which begins by speaking of the “right of the people to be secure in their persons, houses, papers, and effects . . . ,” the First Amendment beginsby focusing upon the activity of the Government. It says that “Congress shall make no law . . . abridging the freedom of speech.” The Government acted upon a constitutionally harmful policy whether Heffernan did or did not in fact engage in political activity. That which stands for a “law” of “Congress,” namely, the police department’s reason for taking action, “abridge[s] the freedom of speech” of employees aware of the policy. And Heffernan was di-rectly harmed, namely, demoted, through application of that policy. We also consider relevant the constitutional implications of a rule that imposes liability. The constitutional harm at issue in the ordinary case consists in large part of discouraging employees—both the employee discharged (or demoted) and his or her colleagues—from engaging in protected activities. The discharge of one tells the others that they engage in protected activity at their peril. See, e.g., Elrod, 427 U. S., at 359 (retaliatory employment action against one employee “unquestionably inhibits protected belief and association” of all employees). Hence, we do not require plaintiffs in political affiliation cases to “prove that they, or other employees, have been coerced into changing, either actually or ostensibly, their political allegiance.” Branti, 445 U. S., at 517. The employer’s factual mistake does not diminish the risk of causing precisely that same harm. Neither, for that matter, is that harm diminished where an employer announces a policy of demoting those who, say, help a particular candidate in the mayoral race, and all employees (including Heffernan), fearful of demotion, refrain from providing any such help. Cf. Gooding v. Wilson, 405 U. S. 518, 521 (1972) (explaining that overbreadth doctrine is necessary “because persons whose expression is constitutionally protected may well refrain from exercising their rights for fear of criminal sanctions”). The upshot is that a discharge or demotion based upon an employer’s belief that the employee has engaged in protected activity can cause the same kind, and degree, of constitutional harm whether that belief does or does not rest upon a factual mistake. Finally, we note that, contrary to respondents’ asser-tions, a rule of law that imposes liability despite the employer’s factual mistake will not normally impose significant extra costs upon the employer. To win, the employee must prove an improper employer motive. In a case like this one, the employee will, if anything, find it more difficult to prove that motive, for the employee will have to point to more than his own conduct to show an employer’s intent to discharge or to demote him for engaging in what the employer (mistakenly) believes to have been different (and protected) activities. We concede that, for that very reason, it may be more complicated and costly for the employee to prove his case. But an employee bringing suit will ordinarily shoulder that more complicated burden voluntarily in order to recover the damages he seeks. III We now relax an assumption underlying our decision. We have assumed that the policy that Heffernan’s employers implemented violated the Constitution. Supra, at 3. There is some evidence in the record, however, suggesting that Heffernan’s employers may have dismissed him pursuant to a different and neutral policy prohibiting police officers from overt involvement in any political campaign. See Brief for United States as Amicus Curiae 27–28. Whether that policy existed, whether Heffernan’s supervisors were indeed following it, and whether it complies with constitutional standards, see Civil Service Comm’n, 413 U. S., at 564, are all matters for the lower courts to decide in the first instance. Without expressing views on the matter, we reverse the judgment of the Third Circuit and remand the case for such further proceedings consistent with this opinion. It is so ordered.

578.US.2015_14-614

The Federal Power Act (FPA) vests in the Federal Energy Regulatory Commission (FERC) exclusive jurisdiction over wholesale sales of electricity in the interstate market, but “leaves to the States alone, the regulation of [retail electricity sales].” FERC v. Electric Power Supply Assn., 577 U. S. ___, ___. In Maryland and other States that have deregulated their energy markets, “load serving entities” (LSEs) purchase electricity at wholesale from independent power generators for delivery to retail consumers. Interstate wholesale transactions in deregulated markets typically occur through (1) bilateral contracting, where LSEs agree to purchase a certain amount of electricity from generators at a certain rate over a certain period of time; and (2) competitive wholesale auctions administered by Regional Transmission Organizations (RTOs) and Independent System Operators (ISOs), nonprofit entities that manage certain segments of the electricity grid. PJM Interconnection (PJM), an RTO overseeing a multistate grid, operates a capacity auction. The capacity auction is designed to identify need for new generation and to accommodate long-term bilateral contracts for capacity. PJM predicts demand three years into the future and assigns a share of that demand to each participating LSE. Owners of capacity to produce electricity in three years’ time then bid that capacity into the auction for sale to PJM at rates the sellers set in their bids. PJM accepts bids until it has purchased enough capacity to satisfy anticipated demand. All accepted capacity sellers receive the highest accepted rate, called the “clearing price.” LSEs then must purchase, from PJM, enough electricity to satisfy their assigned share of overall projected demand. FERC extensively regulates the structure of the capacity auction to ensure that it efficiently balances supply and demand, producing a just and reasonable clearing price. Concerned that the PJM capacity auction was failing to encourage development of sufficient new in-state generation, Maryland enacted its own regulatory program. Maryland selected, through a proposal process, petitioner CPV Maryland, LLC (CPV), to construct a new power plant and required LSEs to enter into a 20-year pricing contract (called a contract for differences) with CPV at a rate CPV specified in its proposal. Under the terms of the contract, CPV sells its capacity to PJM through the auction, but—through mandated payments from or to LSEs—receives the contract price rather than the clearing price for these sales to PJM. In a suit filed by incumbent generators (respondents here) against members of the Maryland Public Service Commission—CPV intervened as a defendant—the District Court issued a declaratory judgment holding that Maryland’s program improperly sets the rate CPV receives for interstate wholesale capacity sales to PJM. The Fourth Circuit affirmed. Held: Maryland’s program is preempted because it disregards the interstate wholesale rate FERC requires. A state law is preempted where “Congress has legislated comprehensively to occupy an entire field of regulation,” Northwest Central Pipeline Corp. v. State Corporation Comm’n of Kan., 489 U. S. 493 , as well as “ ‘where, under the circumstances of [a] particular case, [the challenged state law] stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress,’ ” Crosby v. National Foreign Trade Council, 530 U. S. 363 . Exercising its exclusive authority over interstate wholesale sales, see 16 U. S. C. §824(b)(1), FERC has approved PJM’s capacity auction as the sole ratesetting mechanism for capacity sales to PJM, and has deemed the clearing price per se just and reasonable. However, Maryland—through the contract for differences—guarantees CPV a rate distinct from the clearing price for its interstate capacity sales to PJM. By adjusting an interstate wholesale rate, Maryland’s program contravenes the FPA’s division of authority between state and federal regulators. That Maryland was attempting to encourage construction of new in-state generation does not save its program. States may regulate within their assigned domain even when their laws incidentally affect areas within FERC’s domain. But they may not seek to achieve ends, however legitimate, through regulatory means that intrude on FERC’s authority over interstate wholesale rates, as Maryland has done here. See Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U. S. 354 ; Nantahala Power & Light Co. v. Thornburg, 476 U. S. 953 . Maryland and CPV analogize the contract for differences to traditional bilateral contracts for capacity. Unlike traditional bilateral contracts, however, the contract for differences does not transfer ownership of capacity from one party to another outside the auction. Instead, Maryland’s program operates within the auction, mandating LSEs and CPV to exchange money based on the cost of CPV’s capacity sales to PJM. Maryland’s program is rejected only because it disregards an interstate wholesale rate required by FERC. Neither Maryland nor other States are foreclosed from encouraging production of new or clean generation through measures that do not condition payment of funds on capacity clearing the auction. Pp. 11–15. 753 F. 3d 467, affirmed. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Sotomayor, J., filed a concurring opinion. Thomas, J., filed an opinion concurring in part and concurring in the judgment. Notes 1 Together with No. 14–623, CPV Maryland, LLC v. Talen Energy Marketing, LLC, fka PPL EnergyPlus, LLC, et al., also on certiorari to the same court

The Federal Power Act (FPA), 41Stat. 1063, as amended, 16 U. S. C. §791a et seq., vests in the Federal Energy Regulatory Commission (FERC) exclusive jurisdiction over wholesale sales of electricity in the interstate market. FERC’s regulatory scheme includes an auction-based market mechanism to ensure wholesale rates that are just and reasonable. FERC’s scheme, in Maryland’s view, provided insufficient incentive for new electricity generation in the State. Maryland therefore enacted its own regulatory program. Maryland’s program provides subsidies, through state-mandated contracts, to a new generator, but conditions receipt of those subsidies on the new generator selling capacity into a FERC-regulated wholesale auction. In a suit initiated by competitors of Maryland’s new electricity generator, the Court of Appeals for the Fourth Circuit held that Maryland’s scheme impermissibly intrudes upon the wholesale electricity market, a domain Congress reserved to FERC alone. We affirm the Fourth Circuit’s judgment. I A Under the FPA, FERC has exclusive authority to regulate “the sale of electric energy at wholesale in interstate commerce.” §824(b)(1). A wholesale sale is defined as a “sale of electric energy to any person for resale.” §824(d). The FPA assigns to FERC responsibility for ensuring that “[a]ll rates and charges made, demanded, or received by any public utility for or in connection with the transmission or sale of electric energy subject to the jurisdiction of the Commission . . . shall be just and reasonable.” §824d(a). See also §824e(a) (if a rate or charge is found to be unjust or unreasonable, “the Commission shall determine the just and reasonable rate”). “But the law places beyond FERC’s power, and leaves to the States alone, the regulation of ‘any other sale’—most notably, any retail sale—of electricity.” FERC v. Electric Power Supply Assn., 577 U. S. ___, ___ (2016) (EPSA) (slip op., at 1) (quoting §824(b)). The States’ reserved authority includes control over in-state “facilities used for the generation of electric energy.” §824(b)(1); see Pacific Gas & Elec. Co. v. State Energy Resources Conservation and Development Comm’n, 461 U. S. 190, 205 (1983) (“Need for new power facilities, their economic feasibility, and rates and services, are areas that have been characteristically governed by the States.”). “Since the FPA’s passage, electricity has increasingly become a competitive interstate business, and FERC’s role has evolved accordingly.” EPSA, 577 U. S., at ___ (slip op., at 4). Until relatively recently, most state energy markets were vertically integrated monopolies—i.e., one entity, often a state utility, controlled electricity generation, transmission, and sale to retail consumers. Over the past few decades, many States, including Maryland, have deregulated their energy markets. In deregulated markets, the organizations that deliver electricity to retail consumers—often called “load serving entities” (LSEs)—purchase that electricity at wholesale from independent power generators. To ensure reliable transmission of electricity from independent generators to LSEs, FERC has charged nonprofit entities, called Regional Transmission Organizations (RTOs) and Independent System Operators (ISOs), with managing certain segments of the electricity grid. Interstate wholesale transactions in deregulated markets typically occur through two mechanisms. The first is bilateral contracting: LSEs sign agreements with generators to purchase a certain amount of electricity at a certain rate over a certain period of time. After the parties have agreed to contract terms, FERC may review the rate for reasonableness. See Morgan Stanley Capital Group Inc. v. Public Util. Dist. No. 1 of Snohomish Cty., 554 U. S. 527 –548 (2008) (Because rates set through good-faith arm’s-length negotiation are presumed reasonable, “FERC may abrogate a valid contract only if it harms the public interest.”). Second, RTOs and ISOs administer a number of competitive wholesale auctions: for example, a “same-day auction” for immediate delivery of electricity to LSEs facing a sudden spike in demand; a “next-day auction” to satisfy LSEs’ anticipated near-term demand; and a “capacity auction” to ensure the availability of an adequate supply of power at some point far in the future. These cases involve the capacity auction administered by PJM Interconnection (PJM), an RTO that oversees the electricity grid in all or parts of 13 mid-Atlantic and Midwestern States and the District of Columbia. The PJM capacity auction functions as follows. PJM predicts electricity demand three years ahead of time, and assigns a share of that demand to each participating LSE. Owners of capacity to produce electricity in three years’ time bid to sell that capacity to PJM at proposed rates. PJM accepts bids, beginning with the lowest proposed rate, until it has purchased enough capacity to satisfy projected demand. No matter what rate they listed in their original bids, all accepted capacity sellers receive the highest accepted rate, which is called the “clearing price.”[1] LSEs then must purchase from PJM, at the clearing price, enough electricity to satisfy their PJM-assigned share of overall projected demand. The capacity auction serves to identify need for new generation: A high clearing price in the capacity auction encourages new generators to enter the market, increasing supply and thereby lowering the clearing price in same-day and next-day auctions three years’ hence; a low clearing price discourages new entry and encourages retirement of existing high-cost generators.[2] The auction is designed to accommodate long-term bilateral contracts for capacity. If an LSE has acquired a certain amount of capacity through a long-term bilateral contract with a generator, the LSE—not the generator—is considered the owner of that capacity for purposes of the auction. The LSE sells that capacity into the auction, where it counts toward the LSE’s assigned share of PJM-projected demand, thereby reducing the net costs of the LSE’s required capacity purchases from PJM.[3] LSEs generally bid their capacity into the auction at a price of $0, thus guaranteeing that the capacity will clear at any price. Such bidders are called “price takers.” Because the fixed costs of building generating facilities often vastly exceed the variable costs of producing electricity, many generators also function as price takers. FERC extensively regulates the structure of the PJM capacity auction to ensure that it efficiently balances supply and demand, producing a just and reasonable clearing price. See EPSA, 577 U. S., at ___ (slip op., at 5) (the clearing price is “the price an efficient market would produce”). Two FERC rules are particularly relevant to these cases. First, the Minimum Offer Price Rule (MOPR) requires new generators to bid capacity into the auction at or above a price specified by PJM, unless those generators can prove that their actual costs fall below the MOPR price. Once a new generator clears the auction at the MOPR price, PJM deems that generator an efficient entrant and exempts it from the MOPR going forward, allowing it to bid its capacity into the auction at any price it elects, including $0. Second, the New Entry Price Adjustment (NEPA) guarantees new generators, under certain circumstances, a stable capacity price for their first three years in the market. The NEPA’s guarantee eliminates, for three years, the risk that the new generator’s entry into the auction might so decrease the clearing price as to prevent that generator from recovering its costs. B Around 2009, Maryland electricity regulators became concerned that the PJM capacity auction was failing to encourage development of sufficient new in-state generation. Because Maryland sits in a particularly congested part of the PJM grid, importing electricity from other parts of the grid into the State is often difficult. To address this perceived supply shortfall, Maryland regulators proposed that FERC extend the duration of the NEPA from three years to ten. FERC rejected the proposal. PJM, 126 FERC ¶62,563 (2009). “[G]iving new suppliers longer payments and assurances unavailable to existing suppliers,” FERC reasoned, would improperly favor new generation over existing generation, throwing the auction’s market-based price-setting mechanism out of balance. Ibid. See also PJM, 128 FERC ¶61,789 (2009) (order on petition for rehearing) (“Both new entry and retention of existing efficient capacity are necessary to ensure reliability and both should receive the same price so that the price signals are not skewed in favor of new entry.”). Shortly after FERC rejected Maryland’s NEPA proposal, the Maryland Public Service Commission promulgated the Generation Order at issue here. Under the order, Maryland solicited proposals from various companies for construction of a new gas-fired power plant at a particular location, and accepted the proposal of petitioner CPV Maryland, LLC (CPV). Maryland then required LSEs to enter into a 20-year pricing contract (the parties refer to this contract as a “contract for differences”) with CPV at a rate CPV specified in its accepted proposal.[4] Unlike a traditional bilateral contract for capacity, the contract for differences does not transfer ownership of capacity from CPV to the LSEs. Instead, CPV sells its capacity on the PJM market, but Maryland’s program guarantees CPV the contract price rather than the auction clearing price. If CPV’s capacity clears the PJM capacity auction and the clearing price falls below the price guaranteed in the contract for differences, Maryland LSEs pay CPV the difference between the contract price and the clearing price. The LSEs then pass the costs of these required payments along to Maryland consumers in the form of higher retail prices. If CPV’s capacity clears the auction and the clearing price exceeds the price guaranteed in the contract for differences, CPV pays the LSEs the difference between the contract price and the clearing price, and the LSEs then pass the savings along to consumers in the form of lower retail prices. Because CPV sells its capacity exclusively in the PJM auction market, CPV receives no payment from Maryland LSEs or PJM if its capacity fails to clear the auction. But CPV is guaranteed a certain rate if its capacity does clear, so the contract’s terms encourage CPV to bid its capacity into the auction at the lowest possible price.[5] Prior to enactment of the Maryland program, PJM had exempted new state-supported generation from the MOPR, allowing such generation to bid capacity into the auction at $0 without first clearing at the MOPR price. Responding to a complaint filed by incumbent generators in the Maryland region who objected to Maryland’s program (and the similar New Jersey program), FERC eliminated this exemption. PJM, 135 FERC ¶61,106 (2011). See also 137 FERC ¶61,145 (2011) (order on petition for rehearing) (“Our intent is not to pass judgment on state and local policies and objectives with regard to the development of new capacity resources, or unreasonably interfere with those objectives. We are forced to act, however, when subsidized entry supported by one state’s or locality’s policies has the effect of disrupting the competitive price signals that PJM’s [capacity auction] is designed to produce, and that PJM as a whole, including other states, rely on to attract sufficient capacity.”); New Jersey Bd. of Pub. Util. v. FERC, 744 F. 3d 74, 79–80 (CA3 2014) (upholding FERC’s elimination of the state-supported generation exemption). In the first year CPV bid capacity from its new plant into the PJM capacity auction, that capacity cleared the auction at the MOPR rate, so CPV was thereafter eligible to function as a price taker. In addition to seeking the elimination of the state-supported generation exemption, incumbent generators—respondents here—brought suit in the District of Maryland against members of the Maryland Public Service Commission in their official capacities. The incumbent generators sought a declaratory judgment that Maryland’s program violates the Supremacy Clause by setting a wholesale rate for electricity and by interfering with FERC’s capacity-auction policies.[6] CPV intervened as a defendant. After a six-day bench trial, the District Court issued a declaratory judgment holding that Maryland’s program improperly sets the rate CPV receives for interstate wholesale capacity sales to PJM. PPL Energyplus, LLC v. Nazarian, 974 F. Supp. 2d 790, 840 (Md. 2013). “While Maryland may retain traditional state authority to regulate the development, location, and type of power plants within its borders,” the District Court explained, “the scope of Maryland’s power is necessarily limited by FERC’s exclusive authority to set wholesale energy and capacity prices.” Id., at 829.[7] The Fourth Circuit affirmed. Relying on this Court’s decision in Mississippi Power & Light Co. v. Mississippi ex rel. Moore, 487 U. S. 354, 370 (1988) , the Fourth Circuit observed that state laws are preempted when they “den[y] full effect to the rates set by FERC, even though [they do] not seek to tamper with the actual terms of an interstate transaction.” PPL EnergyPlus, LLC v. Nazarian, 753 F. 3d 467, 476 (2014). Maryland’s program, the Fourth Circuit reasoned, “functionally sets the rate that CPV receives for its sales in the PJM auction,” “a FERC-approved market mechanism.” Id., at 476–477. “[B]y adopting terms and prices set by Maryland, not those sanctioned by FERC,” the Fourth Circuit concluded, Maryland’s program “strikes at the heart of the agency’s statutory power.” Id., at 478.[8] The Fourth Circuit cautioned that it “need not express an opinion on other state efforts to encourage new generation, such as direct subsidies or tax rebates, that may or may not differ in important ways from the Maryland initiative.” Ibid. The Fourth Circuit then held that Maryland’s program impermissibly conflicts with FERC policies. Maryland’s program, the Fourth Circuit determined, “has the potential to seriously distort the PJM auction’s price signals,” undermining the incentive structure FERC has approved for construction of new generation. Ibid. Moreover, the Fourth Circuit explained, Maryland’s program “conflicts with NEPA” by providing a 20-year price guarantee to a new entrant—even though FERC refused Maryland’s request to extend the duration of the NEPA past three years. Id., at 479. We granted certiorari, 577 U. S. ___ (2015), and now affirm. II The Supremacy Clause makes the laws of the United States “the supreme Law of the Land; . . . any Thing in the Constitution or Laws of any State to the Contrary notwithstanding.” U. S. Const., Art. VI, cl. 2. Put simply, federal law preempts contrary state law. “Our inquiry into the scope of a [federal] statute’s pre-emptive effect is guided by the rule that the purpose of Congress is the ultimate touchstone in every pre-emption case.” Altria Group, Inc. v. Good, 555 U. S. 70, 76 (2008) (internal quotation marks omitted). A state law is preempted where “Congress has legislated comprehensively to occupy an entire field of regulation, leaving no room for the States to supplement federal law,” Northwest Central Pipeline Corp. v. State Corporation Comm’n of Kan., 489 U. S. 493, 509 (1989) , as well as “where, under the circumstances of a particular case, the challenged state law stands as an obstacle to the accomplishment and execution of the full purposes and objectives of Congress,” Crosby v. National Foreign Trade Council, 530 U. S. 363, 373 (2000) (brackets and internal quotation marks omitted). We agree with the Fourth Circuit’s judgment that Maryland’s program sets an interstate wholesale rate, contravening the FPA’s division of authority between state and federal regulators. As earlier recounted, see supra, at 2, the FPA allocates to FERC exclusive jurisdiction over “rates and charges . . . received . . . for or in connection with” interstate wholesale sales. §824d(a). Exercising this authority, FERC has approved the PJM capacity auction as the sole ratesetting mechanism for sales of capacity to PJM, and has deemed the clearing price per se just and reasonable. Doubting FERC’s judgment, Maryland—through the contract for differences—requires CPV to participate in the PJM capacity auction, but guarantees CPV a rate distinct from the clearing price for its interstate sales of capacity to PJM. By adjusting an interstate wholesale rate, Maryland’s program invades FERC’s regulatory turf. See EPSA, 577 U. S., at ___ (slip op., at 26) (“The FPA leaves no room either for direct state regulation of the prices of interstate wholesales or for regulation that would indirectly achieve the same result.” (internal quotation marks omitted)).[9] That Maryland was attempting to encourage construction of new in-state generation does not save its program. States, of course, may regulate within the domain Congress assigned to them even when their laws incidentally affect areas within FERC’s domain. See Oneok, Inc. v. Learjet, Inc., 575 U. S. ___, ___ (2015) (slip op., at 11) (whether the Natural Gas Act (NGA) preempts a particular state law turns on “the target at which the state law aims”).[10] But States may not seek to achieve ends, however legitimate, through regulatory means that intrude on FERC’s authority over interstate wholesale rates, as Maryland has done here. See ibid. (distinguishing between “measures aimed directly at interstate purchasers and wholesalers for resale, and those aimed at subjects left to the States to regulate” (internal quotation marks omitted)).[11] The problem we have identified with Maryland’s program mirrors the problems we identified in Mississippi Power & Light and Nantahala Power & Light Co. v. Thornburg, 476 U. S. 953 (1986) . In each of those cases, a State determined that FERC had failed to ensure the reasonableness of a wholesale rate, and the State therefore prevented a utility from recovering—through retail rates—the full cost of wholesale purchases. See Mississippi Power & Light, 487 U. S., at 360–364; Nantahala, 476 U. S., at 956–962. This Court invalidated the States’ attempts to second-guess the reasonableness of interstate wholesale rates. “ ‘Once FERC sets such a rate,’ ” we observed in Mississippi Power & Light, “ ‘a State may not conclude in setting retail rates that the FERC-approved wholesale rates are unreasonable. A State must rather give effect to Congress’ desire to give FERC plenary authority over interstate wholesale rates, and to ensure that the States do not interfere with this authority.’ ” 487 U. S., at 373 (quoting Nantahala, 476 U. S., at 966). True, Maryland’s program does not prevent a utility from recovering through retail sales a cost FERC mandated it incur—Maryland instead guarantees CPV a certain rate for capacity sales to PJM regardless of the clearing price. But Mississippi Power & Light and Nantahala make clear that States interfere with FERC’s authority by disregarding interstate wholesale rates FERC has deemed just and reasonable, even when States exercise their traditional authority over retail rates or, as here, in-state generation. The contract for differences, Maryland and CPV respond, is indistinguishable from traditional bilateral contracts for capacity, which FERC has long accommodated in the auction. See supra, at 4–5, and n. 3. But the contract at issue here differs from traditional bilateral contracts in this significant respect: The contract for differences does not transfer ownership of capacity from one party to another outside the auction. Instead, the contract for differences operates within the auction; it mandates that LSEs and CPV exchange money based on the cost of CPV’s capacity sales to PJM. Notably, because the contract for differences does not contemplate the sale of capacity outside the auction, Maryland and CPV took the position, until the Fourth Circuit issued its decision, that the rate in the contract for differences is not subject to FERC’s reasonableness review. See §824(b)(1) (FERC has jurisdiction over contracts for “the sale of electric energy at wholesale in interstate commerce.” (emphasis added)).[12] Our holding is limited: We reject Maryland’s program only because it disregards an interstate wholesale rate required by FERC. We therefore need not and do not address the permissibility of various other measures States might employ to encourage development of new or clean generation, including tax incentives, land grants, direct subsidies, construction of state-owned generation facilities, or re-regulation of the energy sector. Nothing in this opinion should be read to foreclose Maryland and other States from encouraging production of new or clean generation through measures “untethered to a generator’s wholesale market participation.” Brief for Respondents 40. So long as a State does not condition payment of funds on capacity clearing the auction, the State’s program would not suffer from the fatal defect that renders Maryland’s program unacceptable.[13] * * * For the reasons stated, the judgment of the Court of Appeals for the Fourth Circuit is Affirmed.Notes 1 For example, if four power plants bid to sell capacity at, respectively, $10/unit, $20/unit, $30/unit, and $40/unit, and the first three plants provide enough capacity to satisfy projected demand, PJM will purchase capacity only from those three plants, each of which will receive $30/unit, the clearing price. 2 Because PJM operates the electricity grid in a very large region of the country, PJM divides its overall grid into geographic subregions and makes adjustments to the clearing price to reflect operating conditions in those subregions. For instance, PJM may pay a higher rate in or near areas where transmission-line congestion limits the amount of electricity that can be imported from other areas. The elevated clearing price might encourage a company to site a new power plant in a subregion where the need for local generation is great rather than elsewhere in PJM’s grid. 3 To take a simplified example, assume an LSE has signed a long-term bilateral contract with a generator to purchase 50 units of electricity annually at a price of $40/unit (total annual cost: $2,000). In a given year when the auction clearing price is $50/unit, assume PJM requires the LSE to purchase 100 units of electricity to satisfy its share of projected demand. The LSE bids the 50 units of capacity it already owns into the PJM auction, and PJM pays the LSE $2,500 for those 50 units. Although the LSE then must pay PJM $5,000 for the 100 units it must purchase to satisfy projected demand, the net cost to the LSE of auction participation is only $2,500. Note that the effective price the LSE pays for 50 of the 100 units it must purchase from PJM—the amount purchased through the long-term contract—is the contract price, not the clearing price. That is, the LSE pays the utility $2,000 for 50 units of capacity, receives $2,500 from PJM after selling that capacity into the auction, and then pays $2,500 to PJM to purchase 50 units of capacity, resulting in a net cost of $2,000—the contract price—for those 50 units. The LSE, of course, must pay the full clearing price—$50/unit—for the other 50 units it is obliged to purchase to satisfy its full share of projected demand. 4 New Jersey implemented a similar program around the same time. The duration of the price guarantee for the New Jersey program is 15 years rather than Maryland’s 20. 5 Two simplified examples illustrate how Maryland’s program interacts with the PJM capacity auction. First, consider a hypothetical situation where the clearing price falls below the price guaranteed in the contract for differences. Assume that CPV’s plant produces 10,000 units of electricity a year, and that the 20-year price guaranteed under the contract is $30/unit. Assume further that, in a given year during the duration of the price guarantee, the clearing price is $20/unit, and CPV’s capacity clears the auction. CPV receives payments from Maryland LSEs of $10/unit, or $100,000, and payments from PJM of $20/unit, or $200,000. The rate CPV receives from the capacity auction is therefore $30/unit—the contract price—not $20/unit—the clearing price. Under PJM auction rules, Maryland LSEs then must purchase from PJM, at the clearing price of $20/unit, enough capacity to satisfy their assigned shares of anticipated demand. Assume that PJM requires Maryland LSEs to purchase 40,000 units of capacity. Total capacity-auction expenses for Maryland LSEs would therefore include both the payment to CPV ($100,000) and the full cost of purchasing capacity from PJM ($800,000), or $900,000. Absent Maryland’s program, the LSEs’ capacity-auction expenses would have included only the total cost of capacity purchases from PJM, or $800,000. 6 Because neither CPV nor Maryland has challenged whether plaintiffs may seek declaratory relief under the Supremacy Clause, the Court assumes without deciding that they may. See Brief for Public Utility Law Project of New York, Inc., as Amicus Curiae 21 (arguing that the incumbent generators should have been required to exhaust administrative remedies before filing suit). 7 Respondents also raised arguments under the Dormant Commerce Clause and 42 U. S. C. §1983. The District Court rejected those arguments, PPL Energyplus, LLC v. Nazarian, 974 F. Supp. 2d 790, 841–855 (Md. 2013), the Fourth Circuit did not address them, and they are irrelevant at this stage. 8 For the same reason, the Third Circuit found New Jersey’s similar program preempted. PPL Energyplus, LLC v. Solomon, 766 F. 3d 241, 246 (2014). 9 According to Maryland and CPV, the payments guaranteed under Maryland’s program are consideration for CPV’s compliance with various state-imposed conditions, i.e., the requirements that CPV build a certain type of generator, at a particular location, that would produce a certain amount of electricity over a particular period of time. The payments, Maryland and CPV continue, are therefore separate from the rate CPV receives for its wholesale sales of capacity to PJM. But because the payments are conditioned on CPV’s capacity clearing the auction—and, accordingly, on CPV selling that capacity to PJM—the payments are certainly “received . . . in connection with” interstate wholesale sales to PJM. 16 U. S. C. §824d(a). 10 Although Oneok, Inc. v. Learjet, Inc., 575 U. S. ___ (2015), involved the NGA rather than the FPA, the relevant provisions of the two statutes are analogous. This Court has routinely relied on NGA cases in determining the scope of the FPA, and vice versa. See, e.g., id., at 14–15 (discussing FPA cases while determining the preemptive scope of the NGA). 11 Maryland’s program, Maryland and CPV assert, is consistent with federal law because FERC has accommodated the program by eliminating the MOPR’s state-supported generation exception. Even assuming that this change has prevented Maryland’s program from distorting the auction’s price signals, however—a point the parties dispute—Maryland cannot regulate in a domain Congress assigned to FERC and then require FERC to accommodate Maryland’s intrusion. See Northwest Central Pipeline Corp. v. State Corporation Comm’n of Kan., 489 U. S. 493, 518 (1989) (“The NGA does not require FERC to regulate around a state rule the only purpose of which is to influence purchasing decisions of interstate pipelines, however that rule is labeled.”). 12 Our opinion does not call into question whether generators and LSEs may enter into long-term financial hedging contracts based on the auction clearing price. Such contracts, also frequently termed contracts for differences, do not involve state action to the same degree as Maryland’s program, which compels private actors (LSEs) to enter into contracts for differences—like it or not—with a generator that must sell its capacity to PJM through the auction. 13 Because the reasons we have set out suffice to invalidate Maryland’s program, we do not resolve whether, as the incumbent generators also assert, Maryland’s program is preempted because it counteracts FERC’s refusal to extend the NEPA’s duration, or because it interferes with the capacity auction’s price signals.

577.US.2015_14-7505

Under Florida law, the maximum sentence a capital felon may receive on the basis of a conviction alone is life imprisonment. He may be sentenced to death, but only if an additional sentencing proceeding “results in findings by the court that such person shall be punished by death.” Fla. Stat. §775.082(1). In that proceeding, the sentencing judge first conducts an evidentiary hearing before a jury. §921.141(1). Next, the jury, by majority vote, renders an “advisory sentence.” §921.141(2). Notwithstanding that recommendation, the court must independently find and weigh the aggravating and mitigating circumstances before entering a sentence of life or death. §921.141(3). A Florida jury convicted petitioner Timothy Hurst of first-degree murder for killing a co-worker and recommended the death penalty. The court sentenced Hurst to death, but he was granted a new sentencing hearing on appeal. At resentencing, the jury again recommended death, and the judge again found the facts necessary to sentence Hurst to death. The Florida Supreme Court affirmed, rejecting Hurst’s argument that his sentence violated the Sixth Amendment in light of Ring v. Arizona, 536 U. S. 584 , in which this Court found unconstitutional an Arizona capital sentencing scheme that permitted a judge rather than the jury to find the facts necessary to sentence a defendant to death. Held: Florida’s capital sentencing scheme violates the Sixth Amendment in light of Ring. Pp. 4–10. (a) Any fact that “expose[s] the defendant to a greater punishment than that authorized by the jury’s guilty verdict” is an “element” that must be submitted to a jury. Apprendi v. New Jersey, 530 U. S. 466 . Applying Apprendi to the capital punishment context, the Ring Court had little difficulty concluding that an Arizona judge’s independent factfinding exposed Ring to a punishment greater than the jury’s guilty verdict authorized. 536 U. S., at 604. Ring’s analysis applies equally here. Florida requires not the jury but a judge to make the critical findings necessary to impose the death penalty. That Florida provides an advisory jury is immaterial. See Walton v. Arizona, 497 U. S. 639 . As with Ring, Hurst had the maximum authorized punishment he could receive increased by a judge’s own factfinding. Pp. 4–6. (b) Florida’s counterarguments are rejected. Pp. 6–10. (1) In arguing that the jury’s recommendation necessarily included an aggravating circumstance finding, Florida fails to appreciate the judge’s central and singular role under Florida law, which makes the court’s findings necessary to impose death and makes the jury’s function advisory only. The State cannot now treat the jury’s advisory recommendation as the necessary factual finding required by Ring. Pp. 6–7. (2) Florida’s reliance on Blakely v. Washington, 542 U. S. 296 , is misplaced. There, this Court stated that under Apprendi, a judge may impose any sentence authorized “on the basis of the facts . . . admitted by the defendant,” 542 U. S., at 303. Florida alleges that Hurst’s counsel admitted the existence of a robbery, but Blakely applied Apprendi to facts admitted in a guilty plea, in which the defendant necessarily waived his right to a jury trial, while Florida has not explained how Hurst’s alleged admissions accomplished a similar waiver. In any event, Hurst never admitted to either aggravating circumstance alleged by the State. Pp. 7–8. (3) That this Court upheld Florida’s capital sentencing scheme in Hildwin v. Florida, 490 U. S. 638 , and Spaziano v. Florida, 468 U. S. 447 , does not mean that stare decisis compels the Court to do so here, see Alleyne v. United States, 570 U. S. ___, ___ (Sotomayor, J., concurring). Time and subsequent cases have washed away the logic of Spaziano and Hildwin. Those decisions are thus overruled to the extent they allow a sentencing judge to find an aggravating circumstance, independent of a jury’s factfinding, that is necessary for imposition of the death penalty. Pp. 8–9. (4) The State’s assertion that any error was harmless is not addressed here, where there is no reason to depart from the Court’s normal pattern of leaving such considerations to state courts. P. 10. 147 So. 3d 435, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Thomas, Ginsburg, and Kagan, JJ., joined. Breyer, J., filed an opinion concurring in the judgment. Alito, J., filed a dissenting opinion.

A Florida jury convicted Timothy Lee Hurst of murdering his co-worker, Cynthia Harrison. A penalty-phase jury recommended that Hurst’s judge impose a death sentence. Notwithstanding this recommendation, Florida law required the judge to hold a separate hearing and determine whether sufficient aggravating circumstances existed to justify imposing the death penalty. The judge so found and sentenced Hurst to death. We hold this sentencing scheme unconstitutional. The Sixth Amendment requires a jury, not a judge, to find each fact necessary to impose a sentence of death. A jury’s mere recommendation is not enough. I On May 2, 1998, Cynthia Harrison’s body was discovered in the freezer of the restaurant where she worked—bound, gagged, and stabbed over 60 times. The restaurant safe was unlocked and open, missing hundreds of dollars. The State of Florida charged Harrison’s co-worker, Timothy Lee Hurst, with her murder. See 819 So. 2d 689, 692–694 (Fla. 2002). During Hurst’s 4-day trial, the State offered substantial forensic evidence linking Hurst to the murder. Witnesses also testified that Hurst announced in advance that he planned to rob the restaurant; that Hurst and Harrison were the only people scheduled to work when Harrison was killed; and that Hurst disposed of blood-stained evidence and used stolen money to purchase shoes and rings. Hurst responded with an alibi defense. He claimed he never made it to work because his car broke down. Hurst told police that he called the restaurant to let Harrison know he would be late. He said she sounded scared and he could hear another person—presumably the real murderer—whispering in the background. At the close of Hurst’s defense, the judge instructed the jury that it could find Hurst guilty of first-degree murder under two theories: premeditated murder or felony murder for an unlawful killing during a robbery. The jury convicted Hurst of first-degree murder but did not specify which theory it believed. First-degree murder is a capital felony in Florida. See Fla. Stat. §782.04(1)(a) (2010). Under state law, the maximum sentence a capital felon may receive on the basis of the conviction alone is life imprisonment. §775.082(1). “A person who has been convicted of a capital felony shall be punished by death” only if an additional sentencing proceeding “results in findings by the court that such person shall be punished by death.” Ibid. “[O]therwise such person shall be punished by life imprisonment and shall be ineligible for parole.” Ibid. The additional sentencing proceeding Florida employs is a “hybrid” proceeding “in which [a] jury renders an advisory verdict but the judge makes the ultimate sentencingdeterminations.” Ring v. Arizona, 536 U. S. 584, 608, n. 6 (2002) . First, the sentencing judge conducts an evidentiary hearing before a jury. Fla. Stat. §921.141(1) (2010). Next, the jury renders an “advisory sentence” of life or death without specifying the factual basis of its recommendation. §921.141(2). “Notwithstanding the recommendation of a majority of the jury, the court, after weighing the aggravating and mitigating circumstances, shall enter a sentence of life imprisonment or death.” §921.141(3). If the court imposes death, it must “set forth in writing its findings upon which the sentence of death is based.” Ibid. Although the judge must give the jury recommendation “great weight,” Tedder v. State, 322 So. 2d 908, 910 (Fla. 1975) (per curiam), the sentencing order must “reflect the trial judge’s independent judgment about the existence of aggravating and mitigating factors,” Blackwelder v. State, 851 So. 2d 650, 653 (Fla. 2003) ( per curiam). Following this procedure, Hurst’s jury recommended a death sentence. The judge independently agreed. See 819 So. 2d, at 694–695. On postconviction review, however, the Florida Supreme Court vacated Hurst’s sentence for reasons not relevant to this case. See 18 So. 3d 975 (2009). At resentencing in 2012, the sentencing judge conducted a new hearing during which Hurst offered mitigating evidence that he was not a “major participant” in the murder because he was at home when it happened. App. 505–507. The sentencing judge instructed the advisory jury that it could recommend a death sentence if it found at least one aggravating circumstance beyond a reason-able doubt: that the murder was especially “heinous, atrocious, or cruel” or that it occurred while Hurst was committing a robbery. Id., at 211–212. The jury recommended death by a vote of 7 to 5. The sentencing judge then sentenced Hurst to death. In her written order, the judge based the sentence in part on her independent determination that both the heinous-murder and robbery aggravators existed. Id., at 261–263. She assigned “great weight” to her findings as well as to the jury’s recommendation of death. Id., at 271. The Florida Supreme Court affirmed 4 to 3. 147 So. 3d 435 (2014). As relevant here, the court rejected Hurst’s argument that his sentence violated the Sixth Amendment in light of Ring, 536 U. S. 584 . Ring, the court recognized, “held that capital defendants are entitled to a jury determination of any fact on which the legislature conditions an increase in the maximum punishment.” 147 So. 3d, at 445. But the court considered Ring inapplicable in light of this Court’s repeated support of Florida’s capital sentencing scheme in pre-Ring cases. 147 So. 3d, at 446–447 (citing Hildwin v. Florida, 490 U. S. 638 (1989) (per curiam)); see also Spaziano v. Florida, 468 U. S. 447 –465 (1984). Specifically, in Hildwin, this Court held that the Sixth Amendment “does not require that the specific findings authorizing the imposition of the sentence of death be made by the jury.” 490 U. S., at 640–641. The Florida court noted that we have “never expressly overruled Hildwin, and did not do so in Ring.” 147 So. 3d, at 446–447. Justice Pariente, joined by two colleagues, dissented from this portion of the court’s opinion. She reiterated her view that “Ring requires any fact that qualifies a capital defendant for a sentence of death to be found by a jury.” Id., at 450 (opinion concurring in part and dissenting in part). We granted certiorari to resolve whether Florida’s capital sentencing scheme violates the Sixth Amendment in light of Ring. 575 U. S. ___ (2015). We hold that it does, and reverse. II The Sixth Amendment provides: “In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial, by an impartial jury. . . .” This right, in conjunction with the Due Process Clause, requires that each element of a crime be proved to a jury beyond a reasonable doubt. Alleyne v. United States, 570 U. S. ___, ___ (2013) (slip op., at 3). In Apprendi v. New Jersey, 530 U. S. 466, 494 (2000) , this Court held that any fact that “expose[s] the defendant to a greater punishment than that authorized by the jury’s guilty verdict” is an “element” that must be submitted to a jury. In the years since Apprendi, we have applied its rule to instances involving plea bargains, Blakely v. Washington, 542 U. S. 296 (2004) , sentencing guidelines, United States v. Booker, 543 U. S. 220 (2005) , criminal fines, Southern Union Co. v. United States, 567 U. S. ___ (2012), mandatory minimums, Alleyne, 570 U. S., at ___, and, in Ring, 536 U. S. 584 , capital punishment. In Ring, we concluded that Arizona’s capital sentencing scheme violated Apprendi’s rule because the State allowed a judge to find the facts necessary to sentence a defendant to death. An Arizona jury had convicted Timothy Ring of felony murder. 536 U. S., at 591. Under state law, “Ring could not be sentenced to death, the statutory maximum penalty for first-degree murder, unless further findings were made.” Id., at 592. Specifically, a judge could sentence Ring to death only after independently finding at least one aggravating circumstance. Id., at 592–593. Ring’s judge followed this procedure, found an aggravating circumstance, and sentenced Ring to death. The Court had little difficulty concluding that “ ‘the required finding of an aggravated circumstance exposed Ring to a greater punishment than that authorized by the jury’s guilty verdict.’ ” Id., at 604 (quoting Apprendi, 530 U. S., at 494; alterations omitted). Had Ring’s judge not engaged in any factfinding, Ring would have received a life sentence. Ring, 536 U. S., at 597. Ring’s death sentence therefore violated his right to have a jury find the facts behind his punishment. The analysis the Ring Court applied to Arizona’s sentencing scheme applies equally to Florida’s. Like Arizona at the time of Ring, Florida does not require the jury to make the critical findings necessary to impose the death penalty. Rather, Florida requires a judge to find these facts. Fla. Stat. §921.141(3). Although Florida incorporates an advisory jury verdict that Arizona lacked, we have previously made clear that this distinction is immaterial: “It is true that in Florida the jury recommends a sentence, but it does not make specific factual findings with regard to the existence of mitigating or aggravating circumstances and its recommendation is not binding on the trial judge. A Florida trial court no more has the assistance of a jury’s findings of fact with respect to sentencing issues than does a trial judge in Arizona.” Walton v. Arizona, 497 U. S. 639, 648 (1990) ; accord, State v. Steele, 921 So. 2d 538, 546 (Fla. 2005) (“[T]he trial court alone must make detailed findings about the existence and weight of aggravating circumstances; it has no jury findings on which to rely”). As with Timothy Ring, the maximum punishment Timothy Hurst could have received without any judge-made findings was life in prison without parole. As with Ring, a judge increased Hurst’s authorized punishment based on her own factfinding. In light of Ring, we hold that Hurst’s sentence violates the Sixth Amendment. III Without contesting Ring’s holding, Florida offers a bevy of arguments for why Hurst’s sentence is constitutional. None holds water. A Florida concedes that Ring required a jury to find every fact necessary to render Hurst eligible for the death pen-alty. But Florida argues that when Hurst’s sentencing jury recommended a death sentence, it “necessarily included a finding of an aggravating circumstance.” Brief for Respondent 44. The State contends that this finding qualified Hurst for the death penalty under Florida law, thus satisfying Ring. “[T]he additional requirement that a judge also find an aggravator,” Florida concludes, “only provides the defendant additional protection.” Brief for Respondent 22. The State fails to appreciate the central and singular role the judge plays under Florida law. As described above and by the Florida Supreme Court, the Florida sentencing statute does not make a defendant eligible for death until “findings by the court that such person shall be punished by death.” Fla. Stat. §775.082(1) (emphasis added). The trial court alone must find “the facts . . . [t]hat sufficient aggravating circumstances exist” and “[t]hat there are insufficient mitigating circumstances to outweigh the aggravating circumstances.” §921.141(3); see Steele, 921 So. 2d, at 546. “[T]he jury’s function under the Florida death penalty statute is advisory only.” Spaziano v. State, 433 So. 2d 508, 512 (Fla. 1983). The State cannot now treat the advisory recommendation by the jury as the necessary factual finding that Ring requires. B Florida launches its second salvo at Hurst himself, arguing that he admitted in various contexts that an aggravating circumstance existed. Even if Ring normally requires a jury to hear all facts necessary to sentence a defendant to death, Florida argues, “Ring does not require jury findings on facts defendants have admitted.” Brief for Respondent 41. Florida cites our decision in Blakely v. Washington, 542 U. S. 296 (2004) , in which we stated that under Apprendi, a judge may impose any sentence authorized “on the basis of the facts reflected in the jury verdict or admitted by the defendant.” 542 U. S., at 303 (emphasis deleted). In light of Blakely, Florida points to various instances in which Hurst’s counsel allegedly admitted the existence of a robbery. Florida contends that these “admissions” made Hurst eligible for the death penalty. Brief for Respondent 42–44. Blakely, however, was a decision applying Apprendi to facts admitted in a guilty plea, in which the defendant necessarily waived his right to a jury trial. See 542 U. S., at 310–312. Florida has not explained how Hurst’s alleged admissions accomplished a similar waiver. Florida’s argument is also meritless on its own terms. Hurst never admitted to either aggravating circumstance alleged by the State. At most, his counsel simply refrained from challenging the aggravating circumstances in parts of his appellate briefs. See, e.g., Initial Brief for Appellant in No. SC12–1947 (Fla.), p. 24 (“not challeng[ing] the trial court’s findings” but arguing that death was nevertheless a disproportionate punishment). C The State next argues that stare decisis compels us to uphold Florida’s capital sentencing scheme. As the Flor-ida Supreme Court observed, this Court “repeatedly has reviewed and upheld Florida’s capital sentencing statute over the past quarter of a century.” Bottoson v. Moore, 833 So. 2d 693, 695 (2002) (per curiam) (citing Hildwin, 490 U. S. 638 ; Spaziano, 468 U. S. 447 ). “In a comparable situation,” the Florida court reasoned, “the United States Supreme Court held: ‘If a precedent of this Court has direct application in a case, yet appears to rest on reasons rejected in some other line of decisions, the [other courts] should follow the case which directly controls, leaving to this Court the prerogative of overruling its own decisions.’ ” Bottoson, 833 So. 2d, at 695 (quoting Rodriguez de Quijas v. Shearson/American Express, Inc., 490 U. S. 477, 484 (1989) ); see also 147 So. 3d, at 446–447 (casebelow). We now expressly overrule Spaziano and Hildwin in relevant part. Spaziano and Hildwin summarized earlier precedent to conclude that “the Sixth Amendment does not require that the specific findings authorizing the imposition of the sentence of death be made by the jury.” Hildwin, 490 U. S., at 640–641. Their conclusion was wrong, and irreconcilable with Apprendi. Indeed, today is not the first time we have recognized as much. In Ring, we held that another pre-Apprendi decision—Walton, 497 U. S. 639 —could not “survive the reasoning of Apprendi.” 536 U. S., at 603. Walton, for its part, was a mere application of Hildwin’s holding to Arizona’s capital sentencing scheme. 497 U. S., at 648. “Although ‘ “the doctrine of stare decisis is of fundamental importance to the rule of law[,]” . . . [o]ur precedents are not sacrosanct.’ . . . ‘[W]e have overruled prior decisions where the necessity and propriety of doing so has been established.’ ” Ring, 536 U. S., at 608 (quoting Patterson v. McLean Credit Union, 491 U. S. 164, 172 (1989) ). And in the Apprendi context, we have found that “stare decisis does not compel adherence to a decision whose ‘underpinnings’ have been ‘eroded’ by subsequent developments of constitutional law.” Alleyne, 570 U. S., at ___ (Sotomayor, J., concurring) (slip op., at 2); see also United States v. Gaudin, 515 U. S. 506 –520 (1995) (over-ruling Sinclair v. United States, 279 U. S. 263 (1929) ); Ring, 536 U. S., at 609 (overruling Walton, 497 U. S., at 639); Alleyne, 570 U. S., at ___ (slip op., at 15) (overruling Harris v. United States, 536 U. S. 545 (2002) ). Time and subsequent cases have washed away the logic of Spaziano and Hildwin. The decisions are overruled to the extent they allow a sentencing judge to find an aggravating circumstance, independent of a jury’s factfinding, that is necessary for imposition of the death penalty. D Finally, we do not reach the State’s assertion that any error was harmless. See Neder v. United States, 527 U. S. 1 –19 (1999) (holding that the failure to submit an uncontested element of an offense to a jury may be harmless). This Court normally leaves it to state courts to consider whether an error is harmless, and we see no reason to depart from that pattern here. See Ring, 536 U. S., at 609, n. 7. * * * The Sixth Amendment protects a defendant’s right to an impartial jury. This right required Florida to base Timothy Hurst’s death sentence on a jury’s verdict, not a judge’s factfinding. Florida’s sentencing scheme, which required the judge alone to find the existence of an aggravating circumstance, is therefore unconstitutional. The judgment of the Florida Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. So ordered.

578.US.2015_15-145

Chrysalis Manufacturing Corp. incurred a debt to petitioner Husky International Electronics, Inc., of nearly $164,000. Respondent Daniel Lee Ritz, Jr., Chrysalis’ director and part owner at the time, drained Chrysalis of assets available to pay the debt by transferring large sums of money to other entities Ritz controlled. Husky sued Ritz to recover on the debt. Ritz then filed for Chapter 7 bankruptcy, prompting Husky to file a complaint in Ritz’ bankruptcy case, seeking to hold him personally liable and contending that the debt was not dischargeable because Ritz’ intercompany-transfer scheme constituted “actual fraud” under the Bankruptcy Code’s discharge exceptions. 11 U. S. C. §523(a)(2)(A). The District Court held that Ritz was personally liable under state law but also held that the debt was not “obtained by . . . actual fraud” under §523(a)(2)(A) and thus could be discharged in bankruptcy. The Fifth Circuit affirmed, holding that a misrepresentation from a debtor to a creditor is a necessary element of “actual fraud” and was lacking in this case, because Ritz made no false representations to Husky regarding the transfer of Chrysalis’ assets. Held: The term “actual fraud” in §523(a)(2)(A) encompasses fraudulent conveyance schemes, even when those schemes do not involve a false representation. Pp. 3–11. (a) It is sensible to presume that when Congress amended the Bankruptcy Code in 1978 and added to debts obtained by “false pretenses or false representations” an additional bankruptcy discharge exception for debts obtained by “actual fraud,” it did not intend the term “actual fraud” to mean the same thing as the already-existing term “false representations.” See United States v. Quality Stores, Inc., 572 U. S. ___, ___. Even stronger evidence that “actual fraud” encompasses the kind of conduct alleged to have occurred here is found in the phrase’s historical meaning. At common law, “actual fraud” meant fraud committed with wrongful intent, Neal v. Clark, 95 U. S. 704 . And the term “fraud” has, since the beginnings of bankruptcy practice, been used to describe asset transfers that, like Ritz’ scheme, impair a creditor’s ability to collect a debt. One of the first bankruptcy Acts, the Fraudulent Conveyances Act of 1571, 13 Eliz., ch. 5, identified as “fraud” conveyances made with “[i]ntent to delay hynder or defraude [c]reditors.” The degree to which that statute remains embedded in fraud-related laws today, see, e.g., BFP v. Resolution Trust Corporation, 511 U. S. 531 , clarifies that the common-law term “actual fraud” is broad enough to incorporate fraudulent conveyances. The common law also indicates that fraudulent conveyances do not require a misrepresentation from a debtor to a creditor, see id., at 541, as they lie not in dishonestly inducing a creditor to extend a debt but in the acts of concealment and hindrance. Pp. 3–6. (b) Interpreting “actual fraud” in §523(a)(2)(A) to encompass fraudulent conveyances would not, as Ritz contends, render duplicative two of §523’s other discharge exceptions, §§523(a)(4), (6), given that “actual fraud” captures much conduct not covered by those other provisions. Nor does this interpretation create a redundancy in §727(a)(2), which is meaningfully different from §523(a)(2)(A). It is also not incompatible with §523(a)(2)(A)’s “obtained by” requirement. Even though the transferor of a fraudulent conveyance does not obtain assets or debts through the fraudulent conveyance, the transferee—who, with the requisite intent, also commits fraud—does. At minimum, those debts would not be dischargeable under §523(a)(2)(A). Finally, reading the phrase “actual fraud” to restrict, rather than expand, the discharge exception’s reach would untenably require reading the disjunctive “or” in the phrase “false pretenses, a false representation, or actual fraud” to mean “by.” Pp. 7–10. 787 F. 3d 312, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion.

The Bankruptcy Code prohibits debtors from discharging debts “obtained by . . . false pretenses, a false representation, or actual fraud.” 11 U. S. C. §523(a)(2)(A). The Fifth Circuit held that a debt is “obtained by . . . actual fraud” only if the debtor’s fraud involves a false representation to a creditor. That ruling deepened an existing split among the Circuits over whether “actual fraud” requires a false representation or whether it encompasses other traditional forms of fraud that can be accomplished without a false representation, such as a fraudulent conveyance of property made to evade payment to creditors. We granted certiorari to resolve that split and now reverse. I Husky International Electronics, Inc., is a Colorado-based supplier of components used in electronic devices. Between 2003 and 2007, Husky sold its products to Chrysalis Manufacturing Corp., and Chrysalis incurred a debt to Husky of $163,999.38. During the same period, respondent Daniel Lee Ritz, Jr., served as a director of Chrysalis and owned at least 30% of Chrysalis’ common stock. All parties agree that between 2006 and 2007, Ritz drained Chrysalis of assets it could have used to pay its debts to creditors like Husky by transferring large sums of Chrysalis’ funds to other entities Ritz controlled. For instance—and Ritz’ actions were by no means limited to these examples—Ritz transferred $52,600 to CapNet Risk Management, Inc., a company he owned in full; $121,831 to CapNet Securities Corp., a company in which he owned an 85% interest; and $99,386.90 to Dynalyst Manufacturing Corp., a company in which he owned a 25% interest. In May 2009, Husky filed a lawsuit against Ritz seek-ing to hold him personally responsible for Chrysalis’ $163,999.38 debt. Husky argued that Ritz’ intercompany-transfer scheme was “actual fraud” for purposes of a Texas law that allows creditors to hold shareholders responsible for corporate debt. See Tex. Bus. Orgs. Code Ann. §21.223(b) (West 2012). In December 2009, Ritz filed for Chapter 7 bankruptcy in the United States Bankruptcy Court for the Southern District of Texas. Husky then initiated an adversarial proceeding in Ritz’ bankruptcy case again seeking to hold Ritz personally liable for Chrysalis’ debt. Husky also contended that Ritz could not discharge that debt in bankruptcy because the same intercompany-transfer scheme constituted “actual fraud” under 11 U. S. C. §523(a)(2)(A)’s exemption to discharge.[1] The District Court held that Ritz was personally liable for the debt under Texas law, but that the debt was not “obtained by . . . actual fraud” under §523(a)(2)(A) and could be discharged in his bankruptcy. The Fifth Circuit affirmed. It did not address whether Ritz was responsible for Chrysalis’ debt under Texas law because it agreed with the District Court that Ritz did not commit “actual fraud” under §523(a)(2)(A). Before the Fifth Circuit, Husky argued that Ritz’ asset-transfer scheme was effectuated through a series of fraudulent conveyances—or transfers intended to obstruct the collection of debt. And, Husky said, such transfers are a recognizable form of “actual fraud.” The Fifth Circuit dis-agreed, holding that a necessary element of “actual fraud” is a misrepresentation from the debtor to the creditor, as when a person applying for credit adds an extra zero to her income or falsifies her employment history. In re Ritz, 787 F. 3d 312, 316 (2015). In transferring Chrysalis’ assets, Ritz may have hindered Husky’s ability to recover its debt, but the Fifth Circuit found that he did not make any false representations to Husky regarding those assets or the transfers and therefore did not commit “actual fraud.” We reverse. The term “actual fraud” in §523(a)(2)(A) encompasses forms of fraud, like fraudulent conveyance schemes, that can be effected without a false representation. II A Before 1978, the Bankruptcy Code prohibited debtors from discharging debts obtained by “false pretenses or false representations.” §35(a)(2) (1976 ed.). In the Bankruptcy Reform Act of 1978, Congress added “actual fraud” to that list. 92Stat. 2590. The prohibition now reads: “A discharge under [Chapters 7, 11, 12, or 13] of this title does not discharge an individual debtor from any debt . . . for money, property, services, or an extension, renewal, or refinancing of credit, to the extent obtained by . . . false pretenses, a false representation, or actual fraud.” §523(a)(2)(A) (2012 ed.). When “ ‘Congress acts to amend a statute, we presume it intends its amendment to have real and substantial effect.’ ” United States v. Quality Stores, Inc., 572 U. S. ___, ___ (2014) (slip op., at 7). It is therefore sensible to start with the presumption that Congress did not intend “actual fraud” to mean the same thing as “a false representation,” as the Fifth Circuit’s holding suggests. But the historical meaning of “actual fraud” provides even stronger evidence that the phrase has long encompassed the kind of conduct alleged to have occurred here: a transfer scheme designed to hinder the collection of debt. This Court has historically construed the terms in §523(a)(2)(A) to contain the “elements that the common law has defined them to include.” Field v. Mans, 516 U. S. 59, 69 (1995) . “Actual fraud” has two parts: actual and fraud. The word “actual” has a simple meaning in the context of common-law fraud: It denotes any fraud that “involv[es] moral turpitude or intentional wrong.” Neal v. Clark, 95 U. S. 704, 709 (1878) . “Actual” fraud stands in contrast to “implied” fraud or fraud “in law,” which describe acts of deception that “may exist without the imputation of bad faith or immorality.” Ibid. Thus, anything that counts as “fraud” and is done with wrongful intent is “actual fraud.” Although “fraud” connotes deception or trickery gener-ally, the term is difficult to define more precisely. See 1 J. Story, Commentaries on Equity Jurisprudence §189, p. 221 (6th ed. 1853) (Story) (“Fraud . . . being so various in its nature, and so extensive in its application to human concerns, it would be difficult to enumerate all the instances in which Courts of Equity will grant relief under this head”). There is no need to adopt a definition for all times and all circumstances here because, from the beginning of English bankruptcy practice, courts and legislatures have used the term “fraud” to describe a debtor’s transfer of assets that, like Ritz’ scheme, impairs a creditor’s ability to collect the debt. One of the first bankruptcy acts, the Statute of 13 Elizabeth, has long been relied upon as a restatement of the law of so-called fraudulent conveyances (also known as “fraudulent transfers” or “fraudulent alienations”). See generally G. Glenn, The Law of Fraudulent Conveyances 89–92 (1931). That statute, also called the Fraudulent Conveyances Act of 1571, identified as fraud “faigned covenous and fraudulent Feoffmentes Gyftes Grauntes Alienations [and] Conveyaunces” made with “Intent to delaye hynder or defraude Creditors.” 13 Eliz. ch. 5. In modern terms, Parliament made it fraudulent to hide assets from creditors by giving them to one’s family, friends, or associates. The principles of the Statute of 13 Elizabeth—and even some of its language—continue to be in wide use today. See BFP v. Resolution Trust Corporation, 511 U. S. 531, 540 (1994) (“The modern law of fraudulent transfers had its origin in the Statute of 13 Elizabeth”); id., at 541 (“Every American bankruptcy law has incorporated a fraudulent transfer provision”); Story §353, at 393 (“[T]he statute of 13 Elizabeth . . . has been universally adopted in America, as the basis of our jurisprudence on the same subject”); Boston Trading Group, Inc. v. Burnazos, 835 F. 2d 1504, 1505–1506 (CA1 1987) (Breyer, J.) (“Mass. Gen. Laws ch. 109A, §§1–13 . . . is a uniform state law that codifies both common and statutory law stretching back at least to 1571 and the Statute of Elizabeth”). The degree to which this statute remains embedded in laws related to fraud today clarifies that thecommon-law term “actual fraud” is broad enough toincorporate a fraudulent conveyance. Equally important, the common law also indicates that fraudulent conveyances, although a “fraud,” do not require a misrepresentation from a debtor to a creditor. As a basic point, fraudulent conveyances are not an inducement-based fraud. Fraudulent conveyances typically involve “a transfer to a close relative, a secret transfer, a transfer of title without transfer of possession, or grossly inadequate consideration.” BFP, 511 U. S., at 540–541 (citing Twyne’s Case, 3 Co. Rep. 80b, 76 Eng. Rep. 809 (K. B. 1601)); O. Bump, Fraudulent Conveyances: A Treatise Upon Conveyances Made by Debtors To Defraud Creditors 31–60 (3d ed. 1882)). In such cases, the fraudulent conduct is not in dishonestly inducing a creditor to extend a debt. It is in the acts of concealment and hindrance. In the fraudulent-conveyance context, therefore, the opportunities for a false representation from the debtor to the creditor are limited. The debtor may have the opportunity to put forward a false representation if the creditor inquires into the whereabouts of the debtor’s assets, but that could hardly be considered a defining feature of this kind of fraud. Relatedly, under the Statute of 13 Elizabeth and the laws that followed, both the debtor and the recipient of the conveyed assets were liable for fraud even though the recipient of a fraudulent conveyance of course made no representation, true or false, to the debtor’s creditor. The famous Twyne’s Case, which this Court relied upon in BFP, illustrates this point. See Twyne’s Case, 76 Eng. Rep., at 823 (convicting Twyne of fraud under the Statute of 13 Elizabeth, even though he was the recipient of a debtor’s conveyance). That principle underlies the now-common understanding that a “conveyance which hinders, delays or defrauds creditors shall be void as against [the recipient] unless . . . th[at] party . . . received it in good faith and for consideration.” Glenn, Law of Fraudulent Conveyances §233, at 312. That principle also underscores the point that a false representation has never been a required element of “actual fraud,” and we decline to adopt it as one today. B Ritz concedes that fraudulent conveyances are a formof “actual fraud,”[2] but contends that 11 U. S. C. §523(a)(2)(A)’s particular use of the phrase means something else. Ritz’ strained reading of the provision finds little support. First, Ritz contends that interpreting “actual fraud” in §523(a)(2)(A) to encompass fraudulent conveyances would render duplicative two other exceptions to discharge in §523. Section 523(a)(4) exempts from discharge “any debt . . . for fraud or defalcation while acting in a fiduciary capacity, embezzlement, or larceny.” And §523(a)(6) exempts “any debt . . . for willful and malicious injury by the debtor to another entity or to the property of another entity.” Ritz makes the unremarkable point that the traditional definition of “actual fraud” will cover some of the same conduct as those exceptions: for example, a trustee who fraudulently conveys away his trust’s assets. But Ritz’ interpretation does not avoid duplication, nor does our interpretation fail to preserve a meaningful difference between §523(a)(2)(A) and §§523(a)(4), (6). Just as a fiduciary who engages in a fraudulent conveyance may find his debt exempted from discharge under either §523(a)(2)(A) or §523(a)(4), so too would a fiduciary who engages in one of the fraudulent misrepresentations that form the core of Ritz’ preferred interpretation of §523(a)(2)(A). The same is true for §523(a)(6). The debtors who commit fraudulent conveyances and the debtors who make false representations under §523(a)(2)(A) could likewise also inflict “willful and malicious injury” under §523(a)(6). There is, in short, overlap, but that overlap appears inevitable. And, of course, our interpretation of “actual fraud” in §523(a)(2)(A) also preserves meaningful distinctions between that provision and §§523(a)(4), (a)(6). Section 523(a)(4), for instance, covers only debts for fraud while acting as a fiduciary, whereas §523(a)(2)(A) has no similar limitation. Nothing in our interpretation alters that distinction. And §523(a)(6) covers debts “for willful and malicious injury,” whether or not that injury is the result of fraud, see Kawaauhau v. Geiger, 523 U. S. 57, 61 (1998) (discussing injuries resulting from “ ‘intentional torts’ ”), whereas §523(a)(2)(A) covers only fraudulent acts. Nothing in our interpretation alters that distinction either. Thus, given the clear differences between these provisions, we see no reason to craft an artificial definition of “actual fraud” merely to avoid narrow redundancies in §523 that appear unavoidable. Ritz also says that our interpretation creates redun-dancy with a separate section of the Bankruptcy Code, §727(a)(2), which prevents a debtor from discharging all of his debts if, within the year preceding the bankruptcy petition, he “transferred, removed, destroyed, mutilated, or concealed” property “with intent to hinder, delay, or defraud a creditor or an officer of the estate charged with custody of property.” Although the two provisions could cover some of the same conduct, they are meaningfully different. Section 727(a)(2) is broader than §523(a)(2)(A) in scope—preventing an offending debtor from discharging all debt in bankruptcy. But it is narrower than §523(a)(2)(A) in timing—applying only if the debtor fraudulently conveys assets in the year preceding the bank-ruptcy filing. In short, while §727(a)(2) is a blunt remedy for actions that hinder the entire bankruptcy process, §523(a)(2)(A) is a tailored remedy for behavior connected to specific debts. Ritz’ next point of resistance rests on §523(a)(2)(A)’s requirement that the relevant debt be “for money, prop-erty, services, or . . . credit . . . obtained by . . . actual fraud.” (Emphasis added.) The argument, which the dissent also emphasizes, has two parts: First, it posits that fraudulent conveyances (unlike other forms of actual fraud) cannot be used to “obtai[n]” debt because they function instead to hide valuables that a debtor already possesses. Brief for Respondent 20, 31. There is, the dissent says, no debt at the end of a fraudulent conveyance that could be said to “ ‘resul[t] from’ ” or be “ ‘traceable to’ ” the fraud. Post, at 3 (quoting Field, 516 U. S., at 61, 64). Second, it urges that “actual fraud” not be interpreted to encompass forms of fraud that are incompatible with the provision’s “obtained by” requirement. It is of course true that the transferor does not “obtai[n]” debts in a fraudulent conveyance. But the recipient of the transfer—who, with the requisite intent, also commits fraud—can “obtai[n]” assets “by” his or her participation in the fraud. See, e.g., McClellan v. Cantrell, 217 F. 3d 890 (CA 7 2000); see also supra, at 6. If that recipient later files for bankruptcy, any debts “traceable to” the fraudulent conveyance, see Field, 516 U. S., at 61; post, at 3, will be nondischargable under §523(a)(2)(A). Thus, at least sometimes a debt “obtained by” a fraudulent conveyance scheme could be nondischargeable under §523(a)(2)(A). Such circumstances may be rare because a person who receives fraudulently conveyed assets is not necessarily (or even likely to be) a debtor on the verge of bankruptcy,[3] but they make clear that fraudulent conveyances are not wholly incompatible with the “obtained by” requirement. The dissent presses further still, contending that the phrase “obtained by . . . actual fraud” requires not only that the relevant debts “resul[t] from” or be “traceable to” fraud but also that they “result from fraud at the inception of a credit transaction.” Post, at 3 (emphasis added). Nothing in the text of §523(a)(2)(A) supports that additional requirement. The dissent bases its conclusion on this Court’s opinion in Field, in which the Court noted that certain forms of bankruptcy fraud require a degreeof direct reliance by a creditor on an action taken by a debtor. But Field discussed such “reliance” only in setting forth the requirements of the form of fraud alleged in that case—namely, fraud perpetrated through a misrepresentation to a creditor. See 516 U. S., at 61. The Court was not establishing a “reliance” requirement for frauds that are not premised on such a misrepresentation. Finally, Ritz argues that Congress added the phrase “actual fraud” to §523(a)(2)(A) not to expand the exception’s reach, but to restrict it. In Ritz’ view, “actual fraud” was inserted as the last item in a disjunctive list—“false pretenses, a false representation, or actual fraud”—in order to make clear that the “false pretenses” and “false representation[s]” covered by the provision needed to be intentional. Brief for Respondent 29–31. Ritz asks us, in other words, to ignore what he believes is Congress’ “imprudent use of the word ‘or,’ ” id., at 32, and read the final item in the list to modify and limit the others. In essence, he asks us to change the word “or” to “by.” That is an argument that defeats itself. We can think of no other example, nor could petitioner point to any at oral argument, in which this Court has attempted such an unusual statutory modification. * * * Because we must give the phrase “actual fraud” in §523(a)(2)(A) the meaning it has long held, we interpret “actual fraud” to encompass fraudulent conveyance schemes, even when those schemes do not involve a false representation. We therefore reverse the judgment of the Fifth Circuit and remand the case for further proceedings consistent with this opinion. So ordered.Notes 1 Husky also alleged that Ritz’ debt should be exempted from discharge under two other exceptions, see 11 U. S. C. §523(a)(4) (excepting debts for fraud “while acting in a fiduciary capacity”); §523(a)(6) (excepting debts for “willful and malicious injury”), but does not press those claims in this petition. 2 See Tr. of Oral Arg. 30 (Justice Kagan: “[Y]ou’re not contesting that fraudulent conveyance is a form of actual fraud; is that right?” Ms. Murphy: “[Y]es, that’s right”); id., at 27 (Ms. Murphy: “[T]o be clear, we don’t dispute that fraudulent conveyance is a form of actual fraud”). 3 Ritz’ situation may be unusual in this regard because Husky contends that Ritz was both the transferor and the transferee in his fraudulent conveyance scheme, having transferred Chrysalis assets to other companies he controlled. We take no position on that contention here and leave it to the Fifth Circuit to decide on remand whether the debt to Husky was “obtained by” Ritz’ asset-transfer scheme.

577.US.2015_14-449

A Kansas jury sentenced respondent Sidney Gleason to death for killing a co-conspirator and her boyfriend to cover up the robbery of an elderly man. A Kansas jury sentenced respondents Reginald and Jonathan Carr, brothers, to death after a joint sentencing proceeding. Respondents were convicted of various charges stemming from a notorious crime spree that culminated in the brutal rape, robbery, kidnaping, and execution-style shooting of five young men and women. The Kansas Supreme Court vacated the death sentences in each case, holding that the sentencing instructions violated the Eighth Amendment by failing “to affirmatively inform the jury that mitigating circumstances need only be proved to the satisfaction of the individual juror in that juror’s sentencing decision and not beyond a reasonable doubt.” It also held that the Carrs’ Eighth Amendment right “to an individualized capital sentencing determination” was violated by the trial court’s failure to sever their sentencing proceedings. Held: 1. The Eighth Amendment does not require capital-sentencing courts to instruct a jury that mitigating circumstances need not be proved beyond a reasonable doubt. Pp. 8–13. (a) Because the Kansas Supreme Court left no doubt that its ruling was based on the Federal Constitution, Gleason’s initial argument—that this Court lacks jurisdiction to hear his case because the state court’s decision rested on adequate and independent state-law grounds—is rejected. See Kansas v. Marsh, 548 U. S. 163 . Pp. 8–9. (b) This Court’s capital-sentencing case law does not support requiring a court to instruct a jury that mitigating circumstances need not be proved beyond a reasonable doubt. See, e.g., Buchanan v. Angelone, 522 U. S. 269 . Nor was such an instruction constitutionally necessary in these particular cases to avoid confusion. Ambiguity in capital-sentencing instructions gives rise to constitutional error only if “there is a reasonable likelihood that the jury has applied the challenged instruction in a way that prevents the consideration of constitutionally relevant evidence,” Boyde v. California, 494 U. S. 370 , a bar not cleared here. Even assuming that it would be unconstitutional to require the defense to prove mitigating circumstances beyond a reasonable doubt, the record belies the defendants’ contention that the instructions caused jurors to apply such a standard of proof here. The instructions make clear that both the existence of aggravating circumstances and the conclusion that they outweigh mitigating circumstances must be proved beyond a reasonable doubt but that mitigating circumstances must merely be “found to exist,” which does not suggest proof beyond a reasonable doubt. No juror would have reasonably speculated that “beyond a reasonable doubt” was the correct burden for mitigating circumstances. Pp. 9–13. 2. The Constitution did not require severance of the Carrs’ joint sentencing proceedings. The Eighth Amendment is inapposite when a defendant’s claim is, at bottom, that evidence was improperly admitted at a capital-sentencing proceeding. The question is whether the allegedly improper evidence “so infected the sentencing proceeding with unfairness as to render the jury’s imposition of the death penalty a denial of due process.” Romano v. Oklahoma, 512 U. S. 1 . In light of all the evidence presented at the guilt and penalty phases relevant to the jury’s sentencing determination, the contention that the admission of mitigating evidence by one Carr brother could have “so infected” the jury’s consideration of the other’s sentence as to amount to a denial of due process is beyond the pale. The Court presumes that the jury followed its instructions to “give separate consideration to each defendant.” Bruton v. United States, 391 U. S. 123 , distinguished. Joint proceedings are permissible and often preferable when the joined defendants’ criminal conduct arises out of a single chain of events. Buchanan v. Kentucky, 483 U. S. 402 . Limiting instructions, like those given in the Carrs’ proceeding, “often will suffice to cure any risk of prejudice,” Zafiro v. United States, 506 U. S. 534 , that might arise from codefendants’ “antagonistic” mitigation theories, id., at 538. It is improper to vacate a death sentence based on pure “speculation” of fundamental unfairness, “rather than reasoned judgment.” Romano, supra, at 13–14. Only the most extravagant speculation would lead to the conclusion that any supposedly prejudicial evidence rendered the Carr brothers’ joint sentencing proceeding fundamentally unfair when their acts of almost inconceivable cruelty and depravity were described in excruciating detail by the sole survivor, who, for two days, relived the Wichita Massacre with the jury. Pp. 13–17. No. 14–449, 300 Kan. 340, 329 P. 3d 1195; No. 14–450, 300 Kan. 1, 331 P. 3d 544; and No. 14–452, 299 Kan. 1127, 329 P. 3d 1102, reversed and remanded. Scalia, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, Breyer, Alito, and Kagan, JJ., joined. Sotomayor, J., filed a dissenting opinion. Notes 1 Together with No. 14–450, Kansas v. Carr, and No. 14–452, Kansas v. Gleason, also on certiorari to the same court.

The Supreme Court of Kansas vacated the death sentences of Sidney Gleason and brothers Reginald and Jonathan Carr. Gleason killed one of his co-conspirators and her boyfriend to cover up the robbery of an elderly man. The Carrs’ notorious Wichita crime spree culminated in the brutal rape, robbery, kidnaping, and execution-style shooting of five young men and women. We first consider whether the Constitution required the sentencing courts to instruct the juries that mitigating circumstances “need not be proved beyond a reasonable doubt.” And second, whether the Constitution required severance of the Carrs’ joint sentencing proceedings. I A Less than one month after Sidney Gleason was paroled from his sentence for attempted voluntary manslaughter, he joined a conspiracy to rob an elderly man at knifepoint.[1] Gleason and a companion “cut up” the elderly man to get $10 to $35 and a box of cigarettes. 299 Kan. 1127, 1136, 329 P. 3d 1102, 1115 (2014). Fearing that their female co-conspirators would snitch, Gleason and his cousin, Damien Thompson, set out to kill co-conspirator Mikiala Martinez. Gleason shot and killed Martinez’s boyfriend, and then Gleason and Thompson drove Martinez to a rural location, where Thompson strangled her for five minutes and then shot her in the chest, Gleason standing by and providing the gun for the final shot. The State ultimately charged Gleason with capital murder for killing Martinez and her boyfriend, first-degree premeditated murder of the boyfriend, aggravating kidnaping of Martinez, attempted first-degree murder and aggravated robbery of the elderly man, and criminal possession of a firearm. He was convicted on all counts except the attempted first-degree murder charge. Id., at 1134–1135, 1146, 329 P. 3d, at 1114, 1120. The jury also found that the State proved beyond a reasonable doubt the existence of four aggravating circumstances and unanimously agreed to a sentence of death. Id., at 1146–1147, 329 P. 3d, at 1120–1121. B In December 2000, brothers Reginald and Jonathan Carr set out on a crime spree culminating in the Wichita Massacre.[2] On the night of December 7, Reginald Carr and an unknown man carjacked Andrew Schreiber, held a gun to his head, and forced him to make cash withdrawals at various ATMs. On the night of December 11, the brothers followed Linda Ann Walenta, a cellist for the Wichita symphony, home from orchestra practice. One of them approached her vehicle and said he needed help. When she rolled down her window, he pointed a gun at her head. When she shifted into reverse to escape, he shot her three times, ran back to his brother’s car, and fled the scene. One of the gunshots severed Walenta’s spine, and she died one month later as a result of her injuries. On the night of December 14, the brothers burst into a triplex at 12727 Birchwood, where roommates Jason, Brad, and Aaron lived. Jason’s girlfriend, Holly, and Heather, a friend of Aaron’s, were also in the house. Armed with handguns and a golf club, the brothers forced all five into Jason’s bedroom. They demanded that they strip naked and later ordered them into the bedroom closet. They took Holly and Heather from the bedroom, demanded that they perform oral sex and digitally penetrate each other as the Carrs looked on and barked orders. They forced each of the men to have sex with Holly and then with Heather. They yelled that the men would be shot if they could not have sex with the women, so Holly—fearing for Jason’s life—performed oral sex on him in the closet before he was ordered out by the brothers. Jonathan then snatched Holly from the closet. He ordered that she digitally penetrate herself. He set his gun between her knees on the floor. And he raped her. Then he raped Heather. Reginald took Brad, Jason, Holly, and Aaron one-by-one to various ATMs to withdraw cash. When the victims returned to the house, their torture continued. Holly uri-nated in the closet because of fright. Jonathan found an engagement ring hidden in the bedroom that Jason was keeping as a surprise for Holly. Pointing his gun at Jason, he had Jason identify the ring while Holly was sitting nearby in the closet. Then Reginald took Holly from the closet, said he was not going to shoot her yet, and raped her on the dining-room floor strewn with boxes of Christmas decorations. He forced her to turn around, ejaculated into her mouth, and forced her to swallow. In a nearby bathroom, Jonathan again raped Heather and then again raped Holly. At 2 a.m.—three hours after the mayhem began—the brothers decided it was time to leave the house. They attempted to put all five victims in the trunk of Aaron’s Honda Civic. Finding that they would not all fit, they jammed the three young men into the trunk. They directed Heather to the front of the car and Holly to Jason’s pickup truck, driven by Reginald. Once the vehicles arrived at a snow-covered field, they instructed Jason and Brad, still naked, and Aaron to kneel in the snow. Holly cried, “Oh, my God, they’re going to shoot us.” Holly and Heather were then ordered to kneel in the snow. Holly went to Jason’s side; Heather, to Aaron. Holly heard the first shot, heard Aaron plead with the brothers not to shoot, heard the second shot, heard the screams, heard the third shot, and the fourth. She felt the blow of the fifth shot to her head, but remained kneeling. They kicked her so she would fall face-first into the snow and ran her over in the pickup truck. But she survived, because a hair clip she had fastened to her hair that night deflected the bullet. She went to Jason, took off her sweater, the only scrap of clothing the brothers had let her wear, and tied it around his head to stop the bleeding from his eye. She rushed to Brad, then Aaron, and then Heather. Spotting a house with white Christmas lights in the distance, Holly started running toward it for help—naked, skull shattered, and without shoes, through the snow and over barbed-wire fences. Each time a car passed on the nearby road, she feared it was the brothers returning and camouflaged herself by lying down in the snow. She made it to the house, rang the doorbell, knocked. A man opened the door, and she relayed as quickly as she could the events of the night to him, and minutes later to a 911 dispatcher, fearing that she would not live. Holly lived, and retold this play-by-play of the night’s events to the jury. Investigators also testified that the brothers returned to the Birchwood house after leaving the five friends for dead, where they ransacked the place for valuables and (for good measure) beat Holly’s dog, Nikki, to death with a golf club. The State charged each of the brothers with more than 50 counts, including murder, rape, sodomy, kidnaping, burglary, and robbery, and the jury returned separate guilty verdicts. It convicted Reginald of one count of kidnaping, aggravated robbery, aggravated battery, and criminal damage to property for the Schreiber carjacking, and one count of first-degree felony murder for the Walenta shooting. Jonathan was acquitted of all counts relatedto the Schreiber carjacking but convicted of first-degree felony murder for the Walenta shooting. For the Birchwood murders, the jury convicted each brother of 4 counts of capital murder, 1 count of attempted first-degree murder, 5 counts of aggravated kidnaping, 9 counts of aggravated robbery, 20 counts of rape or attempted rape, 3 counts of aggravated criminal sodomy, 1 count each of aggravated burglary and burglary, 1 count of theft, and 1 count of cruelty to animals. The jury also convicted Reg-inald of three counts of unlawful possession of a firearm. 300 Kan. 1, 15–16, 331 P. 3d 544, 573–574 (2014). The State sought the death penalty for each of the four Birchwood murders, and the brothers were sentenced together. The State relied on the guilt-phase evidence, including Holly’s two days of testimony, as evidence of four aggravating circumstances: that the defendants knowingly or purposely killed or created a great risk of death to more than one person; that they committed the crimes for the purpose of receiving money or items of monetary value; that they committed the crimes to prevent arrest or pro-secution; and that they committed the crimes in an especially heinous, atrocious, or cruel manner. Id., at 258–259, 331 P. 3d, at 708. After hearing each brother’s case for mitigation, the jury issued separate verdicts of death for Reginald and Jonathan. It found unanimously that the State proved the existence of the four aggravating circumstances beyond a reasonable doubt and that those aggravating circumstances outweighed the mitigating circumstances, justifying four separate verdicts of death for each brother for the murders of Jason, Brad, Aaron, and Heather. App. in No. 14–449 etc., pp. 461–492. C The Kansas Supreme Court vacated the death penalties in both cases. It held that the instructions used in both Gleason’s and the Carrs’ sentencing violated the Eighth Amendment because they “failed to affirmatively inform the jury that mitigating circumstances need only be proved to the satisfaction of the individual juror in that juror’s sentencing decision and not beyond a reasonable doubt.” 299 Kan., at 1196, 329 P. 3d, at 1147 (Gleason); 300 Kan., at 303, 331 P. 3d, at 733 (Reginald Carr); 300 Kan. 340, 369–370, 329 P. 3d 1195, 1213 (2014) (Jonathan Carr). Without that instruction, according to the court, the jury “was left to speculate as to the correct burden of proof for mitigating circumstances, and reasonable jurors might have believed they could not consider mitigating circumstances not proven beyond a reasonable doubt.” 299 Kan., at 1197, 329 P. 3d, at 1148. This, the court concluded, might have caused jurors to exclude relevant mitigating evidence from their consideration. Ibid. The Kansas Supreme Court also held that the Carrs’ death sentences had to be vacated because of the trial court’s failure to sever their sentencing proceedings, thereby violating the brothers’ Eighth Amendment right “to an individualized capital sentencing determination.” 300 Kan., at 275, 331 P. 3d, at 717; 300 Kan., at 368, 329 P. 3d, at 1212. According to the court, the joint trial “inhibited the jury’s individualized consideration of [Jonathan] because of family characteristics tending to demonstrate future dangerousness that he shared with his brother”; and his brother’s visible handcuffs prejudiced the jury’s consideration of his sentence. 300 Kan., at 275, 331 P. 3d, at 717. As for Reginald, he was prejudiced, according to the Kansas Supreme Court, by Jonathan’s portrayal of him as the corrupting older brother. Id., at 276, 331 P. 3d, at 717. Moreover, Reginald was prejudiced by his brother’s cross-examination of their sister, who testified that she thought Reginald had admitted to her that he was the shooter. Id., at 279, 331 P. 3d, at 719. (She later backtracked and testified, “ ‘I don’t remember who was, you know, shot by who[m].’ ” Ibid.) The Kansas Supreme Court opined that the presumption that the jury followed its instructions to consider each defendant separately was “defeated by logic.” Id., at 280, 331 P. 3d, at 719. “[T]he defendants’ joint upbringing in the maelstrom that was their family and their influence on and interactions with one another . . . simply was not amenable to orderly separation and analysis.” Ibid., 331 P. 3d, at 719–720. The Kansas Supreme Court found itself unable to “say that the death verdict was unattributable, at least in part, to this error.” Id., at 282, 331 P. 3d, at 720. We granted certio-rari. 575 U. S. ___ (2015). II We first turn to the Kansas Supreme Court’s contention that the Eighth Amendment required these capital-sentencing courts to instruct the jury that mitigating circumstances need not be proved beyond a reasonable doubt. A Before considering the merits of that contention, we consider Gleason’s challenge to our jurisdiction. According to Gleason, the Kansas Supreme Court’s decision rests on adequate and independent state-law grounds. This argument is a familiar one. We rejected it in Kansas v. Marsh, 548 U. S. 163, 169 (2006) . Like the defendant in that case, Gleason urges that the decision below rests only on a rule of Kansas law announced in State v. Kleypas, 272 Kan. 894, 40 P. 3d 139 (2001) (per curiam)—a rule later reiterated in State v. Scott, 286 Kan. 54, 183 P. 3d 801 (2008) ( per curiam). As we stated in Marsh, “Kleypas, itself, rested on federal law.” 548 U. S., at 169. So too does the relevant passage of Scott, which rested on Kleypas’s discussion of the constitutional rule that jurors need not agree on mitigating circumstances. See Scott, supra, at 106–107, 183 P. 3d, at 837–838. The Kansas Supreme Court’s opinion in this case acknowledged as much, saying that “statements from Kleypas implicate the broader Eighth Amendment principle prohibiting barriers that preclude a sentencer’s consideration of all relevant mitigating evidence.” 299 Kan., at 1195, 329 P. 3d, at 1147. The Kansas Supreme Court’s opinion leaves no room for doubt that it was relying on the Federal Constitution. It stated that the instruction it required “protects a capital defendant’s Eighth Amendment right to individualized sentencing,” that the absence of the instruction “implicat[ed] Gleason’s right to individualized sentencing under the Eighth Amendment,” and that vacatur of Gleason’s death sentence was the “[c]onsequen[ce]” of Eighth Amendment error. Id., at 1196–1197, 329 P. 3d, at 1147–1148 (emphasis added). For this reason, the criticism leveled by the dissent is misdirected. It generally would have been “none of our business” had the Kansas Supreme Court vacated Gleason’s and the Carrs’ death sentences on state-law grounds. Marsh, 548 U. S., at 184 (Scalia, J., concurring). But it decidedly did not. And when the Kansas Supreme Court time and again invalidates death sentences because it says the Federal Constitution requires it, “review by this Court, far from undermining state autonomy, is the only possible way to vindicate it.” Ibid. “When we correct a state court’s federal errors, we return power to the State, and to its people.” Ibid. The state courts may experiment all they want with their own constitutions, and often do in the wake of this Court’s decisions. See Sutton, San Antonio Independent School District v. Rodriguez And Its Aftermath, 94 Va. L. Rev. 1963, 1971–1977 (2008). But what a state court cannot do is experiment with our Federal Constitution and expect to elude this Court’s review so long as victory goes to the criminal defendant. “Turning a blind eye” in such cases “would change the uniform ‘law of the land’ into a crazy quilt.” Marsh, supra, at 185. And it would enable state courts to blame the unpopular death-sentence reprieve of the most horrible criminals upon the Federal Constitution when it is in fact their own doing. B We turn, then, to the merits of the Kansas Supreme Court’s conclusion that the Eighth Amendment requires capital-sentencing courts in Kansas “to affirmatively inform the jury that mitigating circumstances need not be proven beyond a reasonable doubt.” 299 Kan., at 1197, 329 P. 3d, at 1148. Approaching the question in the abstract, and without reference to our capital-sentencing case law, we doubt whether it is even possible to apply a standard of proof to the mitigating-factor determination (the so-called “selection phase” of a capital-sentencing proceeding). It is possible to do so for the aggravating-factor determination (the so-called “eligibility phase”), because that is a purely factual determination. The facts justifying death set forth in the Kansas statute either did or did not exist—and one can require the finding that they did exist to be made beyond a reasonable doubt. Whether mitigation exists, however, is largely a judgment call (or perhaps a value call); what one juror might consider mitigating another might not. And of course the ultimate question whether mitigating circumstances outweigh aggravating circumstances is mostly a question of mercy—the quality of which, as we know, is not strained. It would mean nothing, we think, to tell the jury that the defendants must deserve mercy beyond a reasonable doubt; or must more-likely-than-not deserve it. It would be possible, of course, to instruct the jury that the facts establishing mitigating circumstances need only be proved by a preponderance, leaving the judgment whether those facts are indeed mitigating, and whether they outweigh the aggravators, to the jury’s discretion without a standard of proof. If we were to hold that the Constitution requires the mitigating-factor determination to be divided into its factual component and its judgmental component, and the former to be accorded a burden-of-proof instruction, we doubt whether that would produce anything but jury confusion. In the last analysis, jurors will accord mercy if they deem it appropriate, and withhold mercy if they do not, which is what our case law is designed to achieve. In any event, our case law does not require capital sentencing courts “to affirmatively inform the jury that mitigating circumstances need not be proved beyond a reasonable doubt.” Ibid. In Buchanan v. Angelone, 522 U. S. 269 (1998) , we upheld a death sentence even though the trial court “failed to provide the jury with express guidance on the concept of mitigation.” Id., at 275. Likewise in Weeks v. Angelone, 528 U. S. 225 (2000) , we reaffirmed that the Court has “never held that the State must structure in a particular way the manner in which juries consider mitigating evidence” and rejected the contention that it was constitutionally deficient to instruct jurors to “ ‘consider a mitigating circumstance if you find there is evidence to support it,’ ” without additional guidance. Id., at 232–233. Equally unavailing is the contention that even if an instruction that mitigating evidence need not be “proven beyond a reasonable doubt” is not always required, it was constitutionally necessary in these cases to avoid confusion. Ambiguity in capital-sentencing instructions gives rise to constitutional error only if “there is a reasonable likelihood that the jury has applied the challenged instruction in a way that prevents the consideration of constitutionally relevant evidence.” Boyde v. California, 494 U. S. 370, 380 (1990) (emphasis added). The alleged confusion stemming from the jury instructions used at the defendants’ sentencings does not clear that bar. A meager “possibility” of confusion is not enough. Ibid. As an initial matter, the defendants’ argument rests on the assumption that it would be unconstitutional to require the defense to prove mitigating circumstances beyond a reasonable doubt. Assuming without deciding that that is the case, the record belies the defendants’ contention that the instructions caused jurors to apply that standard of proof. The defendants focus upon the following instruction: “The State has the burden to prove beyond a reasonable doubt that there are one or more aggravating circumstances and that they are not outweighed by any mitigating circumstances found to exist.” App. to Pet. for Cert. in No. 14–452, p. 133 (Instr. 8).[3] The juxtaposition of aggravating and mitigating circumstances, so goes the argument, caused the jury to speculate that mitigating circumstances must also be proved beyond a reasonable doubt. 299 Kan., at 1197, 329 P. 3d, at 1148. It seems to us quite the opposite. The instruction makes clear that both the existence of aggravating circumstances and the conclusion that they outweigh mitigating circumstances must be proved beyond a reasonable doubt; mitigating circumstances themselves, on the other hand, must merely be “found to exist.” That same description, mitigating circumstances “found to exist,” is contained in three other instructions, App. to Pet. for Cert. in No. 14–452, at 133 (Instrs. 7, 9, and 10) (emphasis added)—unsurprisingly, since it recites the Kansas statute, see Kan. Stat. Ann. §21–4624(e) (1995). “Found to exist” certainly does not suggest proof beyond a reasonable doubt. The instructions as a whole distinguish clearly between aggravating and mitigating circumstances: “The State has the burden to prove beyond a reasonable doubt that there are one or more aggravating circumstances . . . ,” and the jury must decide unanimously that the State met that burden. App. to Pet. for Cert. in No. 14–452, at 133 (Instrs. 8 and 10) (emphasis added). “Mitigating circumstances,” on the other hand, “do not need to be found by all members of the jury” to “be considered by an individual juror in arriving at his or her sentencing decision.” Id., at 131 (Instr. 7). Not once do the instructions say that defense counsel bears the burden of proving the facts constituting a mitigating circumstance beyond a reasonable doubt—nor would that make much sense, since one of the mitigating circumstances is (curiously) “mercy,” which simply is not a fac-tual determination. We reject the Kansas Supreme Court’s decision that jurors were “left to speculate as to the correct burden of proof for mitigating circumstances.” 299 Kan., at 1197, 329 P. 3d, at 1148. For the reasons we have described, no juror would reasonably have speculated that mitigating circumstances must be proved by any particular standard, let alone beyond a reasonable doubt. The reality is that jurors do not “pars[e] instructions for subtle shades of meaning in the same way that lawyers might.” Boyde, supra, at 381. The instructions repeatedly told the jurors to consider any mitigating factor, meaning any aspect of the defendants’ background or the circumstances of their offense. Jurors would not have misunderstood these instructions to prevent their consideration of constitutionally relevant evidence. III We turn next to the contention that a joint capital-sentencing proceeding in the Carrs’ cases violated the defendants’ Eighth Amendment right to an “individualized sentencing determination.” 300 Kan., at 276, 331 P. 3d, at 717. The Kansas Supreme Court agreed with the defendants that, because of the joint sentencing proceeding, one defendant’s mitigating evidence put a thumb on death’s scale for the other, in violation of the other’s Eighth Amendment rights. Ibid. It accepted Reginald’s contention that he was prejudiced by his brother’s portrayal of him as the corrupting older brother. And it agreed that Reginald was prejudiced by his brother’s cross-examination of their sister, who equivocated about whether Reginald admitted to her that he was the shooter. (Reginald has all but abandoned that implausible theory of prejudice before this Court and contends only that the State “likely would not have introduced any such testimony” had he been sentenced alone. Brief for Respondent in No. 14–450, p. 34, n. 3.) Jonathan asserted that he was prejudiced by evidence associating him with his dangerous older brother, which caused the jury to perceive him as an incurable sociopath.[4] Both speculate that the evidence assertedly prejudicial to them would have been inadmissible in severed proceedings under Kansas law. The Kansas Supreme Court also launched a broader attack on the joint proceedings, contending that the joinder rendered it impossible for the jury to consider the Carrs’ relative moral culpability and to determine individually whether they were entitled to “mercy.” 300 Kan., at 278, 331 P. 3d, at 718–719. Whatever the merits of defendants’ procedural objections, we will not shoehorn them into the Eighth Amendment’s prohibition of “cruel and unusual punishments.” As the United States as amicus curiae intimates, the Eighth Amendment is inapposite when each defendant’s claim is, at bottom, that the jury considered evidence that would not have been admitted in a severed proceeding, and that the joint trial clouded the jury’s consideration of mitigating evidence like “mercy.” Brief for United States 24, n. 8. As we held in Romano v. Oklahoma, 512 U. S. 1 (1994) , it is not the role of the Eighth Amendment to establish a special “federal code of evidence” governing “the admissibility of evidence at capital sentencing proceedings.” Id., at 11–12. Rather, it is the Due Process Clause that wards off the introduction of “unduly prejudicial” evidence that would “rende[r] the trial fundamentally unfair.” Payne v. Tennessee, 501 U. S. 808, 825 (1991) ; see also Brown v. Sanders, 546 U. S. 212 –221 (2006). The test prescribed by Romano for a constitutional violation attributable to evidence improperly admitted at a capital-sentencing proceeding is whether the evidence “so infected the sentencing proceeding with unfairness as to render the jury’s imposition of the death penalty a denial of due process.” 512 U. S., at 12. The mere admission of evidence that might not otherwise have been admitted in a severed proceeding does not demand the automatic vacatur of a death sentence. In light of all the evidence presented at the guilt and penalty phases relevant to the jury’s sentencing determination, the contention that the admission of mitigating evidence by one brother could have “so infected” the jury’s consideration of the other’s sentence as to amount to a denial of due process is beyond the pale. To begin with, the court instructed the jury that it “must give separate consideration to each defendant,” that each was “entitled to have his sentence decided on the evidence and law which is applicable to him,” and that any evidence in the penalty phase “limited to only one defendant should not be considered by you as to the other defendant.” App. to Pet. for Cert. in No. 14–450, at 501 (Instr. 3). The court gave defendant-specific instructions for aggravating and mitigating circumstances. Id., at 502–508 (Instrs. 5, 6, 7, and 8). And the court instructed the jury to consider the “individual” or “particular defendant” by using four separate verdict forms for each defendant, one for each murdered occupant of the Birchwood house. Id., at 509 (Instr. 10); App. in No. 14–449 etc., at 461–492. We presume the jury followed these instructions and considered each defendant separately when deciding to impose a sentence of death for each of the brutal murders. Romano, supra, at 13. The contrary conclusion of the Kansas Supreme Court—that the presumption that jurors followed these instructions was “defeated by logic,” 300 Kan., at 280, 331 P. 3d, at 719—is untenable. The Carrs implausibly liken the prejudice resulting from the joint sentencing proceeding to the prejudice infecting the joint trial in Bruton v. United States, 391 U. S. 123 (1968) , where the prosecution admitted hearsay evidence of a codefendant’s confession implicating the defendant. That particular violation of the defendant’s confrontation rights, incriminating evidence of the most persuasive sort, ineradicable, as a practical matter, from the jury’s mind, justified what we have described as a narrow departure from the presumption that jurors follow their instructions, Richardson v. Marsh, 481 U. S. 200, 207 (1987) . We have declined to extend that exception, id., at 211, and have continued to apply the presumption to instructions regarding mitigating evidence in capital-sentencing proceedings, see, e.g., Weeks, 528 U. S., at 234. There is no reason to think the jury could not follow its instruction to consider the defendants separately in this case. Joint proceedings are not only permissible but are often preferable when the joined defendants’ criminal conduct arises out of a single chain of events. Joint trial may enable a jury “to arrive more reliably at its conclusions regarding the guilt or innocence of a particular defendant and to assign fairly the respective responsibilities of each defendant in the sentencing.” Buchanan v. Kentucky, 483 U. S. 402, 418 (1987) . That the codefendants might have “antagonistic” theories of mitigation, Zafiro v. United States, 506 U. S. 534, 538 (1993) , does not suffice to overcome Kansas’s “interest in promoting the reliability and consistency of its judicial process,” Buchanan, supra, at 418. Limiting instructions, like those used in the Carrs’ sentencing proceeding, “often will suffice to cure any risk of prejudice.” Zafiro, supra, at 539 (citing Richardson, supra, at 211). To forbid joinder in capital-sentencing proceedings would, perversely, increase the odds of “wanto[n] and freakis[h]” imposition of death sentences. Gregg v. Georgia, 428 U. S. 153 –207 (1976) (joint opinion of Stewart, Powell, and Stevens, JJ.). Better that two defendants who have together committed the same crimes be placed side-by-side to have their fates determined by a single jury. It is improper to vacate a death sentence based on pure “speculation” of fundamental unfairness, “rather than reasoned judgment,” Romano, supra, at 13–14. Only the most extravagant speculation would lead to the conclusion that the supposedly prejudicial evidence rendered the Carr brothers’ joint sentencing proceeding fundamentally unfair. It is beyond reason to think that the jury’s death verdicts were caused by the identification of Reginald as the “corrupter” or of Jonathan as the “corrupted,” the jury’s viewing of Reginald’s handcuffs, or the sister’s retracted statement that Reginald fired the final shots. None of that mattered. What these defendants did—acts of almost inconceivable cruelty and depravity—was described in excruciating detail by Holly, who relived with the jury, for two days, the Wichita Massacre. The joint sentencing proceedings did not render the sentencing proceedings fundamentally unfair. IV When we granted the State’s petition for a writ of certiorari for the Carrs’ cases, we declined to review whether the Confrontation Clause, U. S. Const., Amdt. 6, requires that defendants be allowed to cross-examine witnesses whose statements are recorded in police reports referred to by the State in penalty-phase proceedings. The Kansas Supreme Court did not make the admission of those statements a basis for its vacating of the death sentences, but merely “caution[ed]” that in the resentencing proceedings these out-of-court testimonial statements should be omitted, 300 Kan., at 288, 331 P. 3d, at 724. We are confident that cross-examination regarding these police reports would not have had the slightest effect upon the sen-tences. See Delaware v. Van Arsdall, 475 U. S. 673, 684 (1986) . * * * The judgments of the Supreme Court of Kansas are reversed, and these cases are remanded for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 The facts for this portion of the opinion come from the Kansas Supreme Court, 299 Kan. 1127, 1134–1147, 329 P. 3d 1102, 1113–1121 (2014), and the parties’ briefs. 2 The facts for this portion of the opinion come from the Kansas Supreme Court, 300 Kan. 1, 18–38, 331 P. 3d 544, 575–586 (2014), and witness testimony. See 21–A Tr. 59–75 (Oct. 7, 2002), 22–B Tr. 39–124 (Oct. 8, 2002), 23–A Tr. 4–118 (Oct. 9, 2002), 23–B Tr. 5–133 (Oct. 9, 2002), and 24–A Tr. 4–93 (Oct. 10, 2002). 3 The relevant penalty-phase instructions from the Carrs’ sentencing proceedings are materially indistinguishable. See App. to Pet. for Cert. in No. 14–450, pp. 501–510. 4 Jonathan also alleges that he was prejudiced by the jury’s witnessing his brother’s handcuffs, which his brother requested remain visible before the penalty phase commenced. That allegation is mystifying. That his brother’s handcuffs were visible (while his own restraints were not) more likely caused the jury to see Jonathan as the less dangerous of the two.

579.US.2015_14-916

The Veterans Benefits, Health Care, and Information Technology Act of 2006 requires the Secretary of Veterans Affairs to set annual goals for contracting with service-disabled and other veteran-owned small businesses. 38 U. S. C. §8127(a). To help reach those goals, a separate set-aside provision known as the “Rule of Two” provides that a contracting officer “shall award contracts” by restricting competition to veteran-owned small businesses if the officer reasonably expects that at least two such businesses will submit offers and that “the award can be made at a fair and reasonable price that offers best value to the United States.” §8127(d). Two exceptions provide that the contracting officer “may” use noncompetitive and sole-source contracts for contracts below specific dollar amounts. §§8127(b), (c). In 2012, the Department procured an Emergency Notification Service for four medical centers for a one-year period, with an option to extend the agreement for two more, from a non-veteran-owned business. The Department did so through the Federal Supply Schedule (FSS), a streamlined method that allows Government agencies to acquire particular goods and services under prenegotiated terms. After the initial year, the Department exercised its option for an additional year, and the agreement ended in 2013. Petitioner Kingdomware Technologies, Inc., a service-disabled veteran-owned small business, filed a bid protest with the Government Accountability Office (GAO), alleging that the Department procured multiple contracts through the FSS without employing the Rule of Two. The GAO determined that the Department’s actions were unlawful, but when the Department declined to follow the GAO’s nonbinding recommendation, Kingdomware filed suit, seeking declaratory and injunctive relief. The Court of Federal Claims granted summary judgment to the Government, and the Federal Circuit affirmed, holding that the Department was only required to apply the Rule of Two when necessary to satisfy its annual goals. Held: 1. This Court has jurisdiction to reach the merits of this case. For a federal court to have Article III jurisdiction “an actual controversy must exist . . . through all stages of the litigation.” Already, LLC v. Nike, Inc., 568 U. S. ___, ___. Here, no court is capable of granting petitioner relief initially sought in the complaint because the short-term FSS contracts have been completed by other contractors. However, the controversy is “ ‘capable of repetition, yet evading review.’ ” Spencer v. Kemna, 523 U. S. 1 . The procurements were fully performed in less than two years after they were awarded, and it is reasonable to expect that the Government will refuse to apply the Rule of Two in a future bid by Kingdomware. Pp. 6–8. 2. Section 8127(d)’s contracting procedures are mandatory and apply to all of the Department’s contracting determinations. Pp. 8–12. (a) Section 8127(d)’s text unambiguously requires the Department to use the Rule of Two before contracting under the competitive procedures. The word “shall” usually connotes a requirement, unlike the word “may,” which implies discretion. Compare Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U. S. 26 , with United States v. Rodgers, 461 U. S. 677 . The use of the word “may” in §§8127(b) and (c) confirms this reading; for when a statute distinguishes between “may” and “shall,” the latter generally imposes a mandatory duty. Pp. 8–9. (b) Alternative readings of §8127(d) are unpersuasive. First, §8127(d)’s prefatory clause, which declares that the Rule of Two is designed “for the purposes of” meeting §8127(a)’s annual contracting goals, has no bearing on whether §8127(d)’s requirement is mandatory or discretionary. The prefatory clause’s announcement of an objective does not change the operative clause’s plain meaning. See Yazoo & Mississippi Valley R. Co. v. Thomas, 132 U. S. 174 . Second, an FSS order is a “contract” within the ordinary meaning of that term; thus, FSS orders do not fall outside §8127(d), which applies when the Department “award[s] contracts.” Third, to say that the Rule of Two will hamper mundane Government purchases misapprehends current FSS practices, which have expanded well beyond simple procurement to, as in this case, contracts concerning complex information technology services over a multiyear period. Finally, because the mandate §8127(d) imposes is unambiguous, this Court declines the invitation to defer to the Department’s declaration that §8127 procedures are inapplicable to FSS orders. See Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 –843. Pp. 9–12. 754 F. 3d 923, reversed and remanded. Thomas, J., delivered the opinion for a unanimous Court.

Petitioner Kingdomware Technologies, Inc., a veteran-owned small business, unsuccessfully vied for a federal contract from the Department of Veterans Affairs to provide emergency-notification services. Kingdomware sued, arguing that the Department violated a federal law providing that it “shall award” contracts to veteran-owned small businesses when there is a “reasonable expectation” that two or more such businesses will bid for the contract at “a fair and reasonable price that offers best value to the United States.” 38 U. S. C. §8127(d). This provision is known as the Rule of Two. In this case, we consider whether the Department must use the Rule of Two every time it awards contracts or whether it must use the Rule of Two only to the extent necessary to meet annual minimum goals for contracting with veteran-owned small businesses. We conclude that the Department must use the Rule of Two when awarding contracts, even when the Department will otherwise meet its annual minimum contracting goals. I This case concerns the interplay between several federal statutes governing federal procurement. A In an effort to encourage small businesses, Congress has mandated that federal agencies restrict competition for some federal contracts. The Small Business Act thus requires many federal agencies, including the Department of Veterans Affairs, to set aside contracts to be awarded to small businesses. The Act requires each agency to set “an annual goal that presents, for that agency, the maximum practicable opportunity” for contracting with small businesses, including those “small business concerns owned and controlled by service-disabled veterans.” 15 U. S. C. §644(g)(1)(B). And federal regulations set forth procedures for most agencies to “set aside” contracts for small businesses. See, e.g., 48 CFR §19.502–2(b) (2015). In 1999, Congress expanded small-business opportunities for veterans by passing the Veterans Entrepreneurship and Small Business Development Act, 113Stat. 233. That Act established a 3% governmentwide contracting goal for contracting with service-disabled veteran-owned small businesses. 15 U. S. C. §644(g)(1)(A)(ii). When the Federal Government continually fell behind in achieving these goals, Congress tried to correct the situation. Relevant here, Congress enacted the Veterans Benefits, Health Care, and Information Technology Act of 2006, §§502, 503, 120Stat. 3431–3436 (codified, as amended, at 38 U. S. C. §§8127, 8128). That Act requires the Secretary of Veterans Affairs to set more specific annual goals that encourage contracting with veteran-owned and service-disabled veteran-owned small businesses. §8127(a). The Act’s “Rule of Two,” at issue here, provides that the Department “shall award” contracts by restricting competition for the contract to service-disabled or other veteran-owned small businesses. To restrict competition under the Act, the contracting officer must reasonably expect that at least two of these businesses will submit offers and that “the award can be made at a fair and reasonable price that offers best value to the United States.” §8127(d).[1] Congress provided two exceptions to the Rule. Under those exceptions, the Department may use noncompetitive and sole-source contracts when the contracts are below specific dollar amounts. Under §8127(b), a contracting officer “may use procedures other than competitive procedures” to award contracts to veteran-owned small businesses when the goods or services that are the subject of such contracts are worth less than the simplified acquisition threshold. 38 U. S. C. §8127(b); 41 U. S. C. §134 (establishing a “ ‘simplified acquisition threshold’ ” of $100,000); see also §1908 (authorizing adjustments for inflation); 75 Fed. Reg. 53130 (codified at 48 CFR §2.101 (2010)) (raising the amount to $150,000). And under 38 U. S. C. §8127(c), a contracting officer “may award a contract to a [veteran-owned small business] using procedures other than competitive procedures” if the contract is worth more than the simplified acquisition threshold but less than $5 million, the contracting officer determines that the business is “a responsible source with respect to performance of such contract opportunity,” and the award can be made at “a fair and reasonable price.” 38 U. S. C. §8127(c). In finalizing its regulations meant to implement the Act, the Department stated in a preamble that §8127’s procedures “do not apply to [Federal Supply Schedule] task or delivery orders.” VA Acquisition Regulation, 74 Fed. Reg. 64624 (2009). The Federal Supply Schedule (FSS) generally is a streamlined method for Government agencies to acquire certain supplies and services in bulk, such as office supplies or food equipment. 48 CFR §8.402(a) (2015). Instead of the normal bidding process for each individual order, FSS contracts are ordinarily pre-negotiated between outside vendors and the General Services Administration, which negotiates on behalf of various Government agencies. See §8.402(b); Sharp Electronics Corp. v. McHugh, 707 F. 3d 1367, 1369 (CA Fed 2013). Under FSS contracts, businesses agree to provide “[i]ndefinite delivery” of particular goods or services “at stated prices for given periods of time.” §8.402(a). Agencies receive a list of goods and services available through the FSS. Because the terms of purchasing these goods and services have already been negotiated, contracting officers can acquire these items and services simply by issuing purchase orders. B Kingdomware Technologies, Inc., is a service-disabled veteran-owned small business. Around January 2012, the Department decided to procure an Emergency Notification Service for four medical centers.[2] In an emergency, this service sends important information to Department personnel. The Department sent a request for a price quotation to a non-veteran-owned company through the FSS system. That company responded with a favorable price, which the Department accepted around February 22, 2012. The agreement was for one year, with an option to extend the agreement for two more. The Department exercised the one option to extend the time, and performance was completed in May 2013. Decl. of Corydon Ford Heard III ¶8. Kingdomware challenged the Department’s decision to award the contract to a non-veteran-owned company by filing a bid protest with the Government Accountability Office (GAO). See 31 U. S. C. §3552(a). Kingdomware alleged that the Department procured multiple contracts through the FSS without restricting competition using the Rule of Two, as required by §8127. Kingdomware contended that the Department could not award the contracts at issue here without first checking to see whether at least two veteran-owned small businesses could perform the work at a fair and reasonable price. The GAO issued a nonbinding determination that the Department’s failure to employ the Rule of Two was unlawful and recommended that the Department conduct market research to determine whether there were two veteran-owned businesses that could fulfill the procurement. The Department dis-agreed with the recommendation. Petitioner then filed suit in the Court of Federal Claims and sought declaratory and injunctive relief.[3] The Court of Federal Claims granted summary judgment to the Department. 107 Fed. Cl. 226 (2012). A divided panel of the Federal Circuit affirmed. 754 F. 3d 923 (2014). In the majority’s view, §8127 did not require the Department to use the Rule of Two in all contracting. Id., at 933–934. Instead, the court concluded, mandatory application of the Rule of Two was limited to contracts necessary to fulfill its statutory purpose—to provide a means of satisfying the Department’s annual contracting goals described in §8127(a). Id., at 934. Thus, so long as those goals were satisfied, the Court of Appeals concluded, the Department need not apply the Rule of Two any further. Ibid. Judge Reyna dissented, arguing that §8127 employs mandatory language that “could not be clearer” in requiring the Department to apply the Rule of Two in every instance of contracting. Id., at 935. We granted certiorari to decide whether §8127(d) requires the Department to apply the Rule of Two in all contracting, or whether the statute gives the Department some discretion in applying the rule. 576 U. S. ___ (2015). II Before we reach the merits, we must assess our jurisdiction. Article III of the Constitution limits federal courts to deciding “Cases” and “Controversies,” and “an actual controversy must exist not only at the time the complaint is filed, but through all stages of the litigation.” Already, LLC v. Nike, Inc., 568 U. S. ___, ___ (2013) (slip op., at 3–4) (internal quotation marks omitted). Here, no live controversy in the ordinary sense remains because no court is now capable of granting the relief petitioner seeks. When Kingdomware filed this suit four years ago, it sought a permanent injunction and declara-tory relief with respect to a particular procurement. The services at issue in that procurement were completed in May 2013. And the two earlier procurements, which Kingdomware had also protested, were complete in September 2012. See Decl. of Corydon Ford Heard III ¶¶6–8. As a result, no court can enjoin further performance of those services or solicit new bids for the performance of those services. And declaratory relief would have no effect here with respect to the present procurements because the services have already been rendered. Although a case would generally be moot in such circumstances, this Court’s precedents recognize an exception to the mootness doctrine for a controversy that is “ ‘capable of repetition, yet evading review.’ ” Spencer v. Kemna, 523 U. S. 1, 17 (1998) . That exception applies “only in exceptional situations,” where (1) “the challenged action [is] in its duration too short to be fully litigated prior to cessation or expiration,” and (2) “there [is] a reasonable expectation that the same complaining party [will] be subject to the same action again.” Ibid. (internal quotation marks omitted; brackets in original). That exception applies to these short-term contracts. First, the procurements were fully performed in less than two years after they were awarded. We have previously held that a period of two years is too short to complete judicial review of the lawfulness of the procurement. See Southern Pacific Terminal Co. v. ICC, 219 U. S. 498 –516 (1911). Second, it is reasonable to expect that the Department will refuse to apply the Rule of Two in a future procurement for the kind of services provided by Kingdomware. If Kingdomware’s interpretation of §8127(d) is correct, then the Department must use restricted competition rather than procure on the open market. And Kingdomware, which has been awarded many previous contracts, has shown a reasonable likelihood that it would be awarded a future contract if its interpretation of §8127(d) prevails. See Decl. of Corydon Ford Heard III ¶¶11–15 (explaining that the company continues to bid on similar contracts). Thus, we have jurisdiction because the same legal issue in this case is likely to recur in future controversies between the same parties in circumstances where the period of contract performance is too short to allow full judicial review before performance is complete. Our interpretation of §8127(d)’s requirements in this case will govern the Department’s future contracting. III On the merits, we hold that §8127 is mandatory, not discretionary. Its text requires the Department to apply the Rule of Two to all contracting determinations and to award contracts to veteran-owned small businesses. The Act does not allow the Department to evade the Rule of Two on the ground that it has already met its contracting goals or on the ground that the Department has placed an order through the FSS. A In statutory construction, we begin “with the language of the statute.” Barnhart v. Sigmon Coal Co., 534 U. S. 438, 450 (2002) . If the statutory language is unambiguous and “the statutory scheme is coherent and consistent”—as is the case here—“[t]he inquiry ceases.” Ibid. We hold that §8127(d) unambiguously requires the Department to use the Rule of Two before contracting under the competitive procedures. Section 8127(d) requires that “a contracting officer of the Department shall award contracts” to veteran-owned small businesses using restricted competition whenever the Rule of Two is satisfied, “[e]xcept as provided in subsections (b) and (c).” (Emphasis added.) Subsections (b) and (c) provide, in turn, that the Department “may” use noncompetitive procedures and sole-source contracts for lower value acquisitions. §§8127(b), (c). Except when the Department uses the noncompetitive and sole-source contracting procedures in subsections (b) and (c), §8127(d) requires the Department to use the Rule of Two before awarding a contract to another supplier. The text also has no exceptions for orders from the FSS system. Congress’ use of the word “shall” demonstrates that §8127(d) mandates the use of the Rule of Two in all contracting before using competitive procedures. Unlike the word “may,” which implies discretion, the word “shall” usually connotes a requirement. Compare Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U. S. 26, 35 (1998) (recognizing that “shall” is “mandatory” and “normally creates an obligation impervious to judicial discretion”), with United States v. Rodgers, 461 U. S. 677, 706 (1983) (explaining that “[t]he word ‘may,’ when used in a statute, usually implies some degree of discretion”). Accordingly, the Department shall (or must) prefer veteran-owned small businesses when the Rule of Two is satisfied. The surrounding subsections of §8127 confirm that Congress used the word “shall” in §8127(d) as a command. Like §8127(d), both §8127(b) and §8127(c) provide special procedures “[f]or purposes of meeting the goals under [§8127(a)].” §§8127(b), (c). But, in contrast to §8127(d), those latter two provisions state that “a contracting officer of the Department may use” (or, for §8127(c), “may award”) such contracts. §§8127(b), (c) (emphasis added). When a statute distinguishes between “may” and “shall,” it is generally clear that “shall” imposes a mandatory duty. See United States ex rel. Siegel v. Thoman, 156 U. S. 353 –360 (1895). We see no reason to depart from the usual inference here. We therefore hold that, before contracting with a non-veteran owned business, the Department must first apply the Rule of Two.[4] B The Federal Circuit and the Department offered several reasons for their alternative reading of §8127(d) as a discretionary provision that the Department can disregard for at least some contracting decisions. We disagree with them. To hold that §8127(d) is discretionary, the Federal Circuit relied on §8127(d)’s prefatory clause. 754 F. 3d, at 933. That clause declares that the Rule of Two is designed “for the purposes of” meeting the annual contracting goals that the Department is required to set under §8127(a). The Department originally made a similar argument before changing arguments in its briefing on the merits. Compare Brief in Opposition 13–15 with Brief for United States 24–25. But the prefatory clause has no bearing on whether §8127(d)’s requirement is mandatory or discretionary. The clause announces an objective that Congress hoped that the Department would achieve and charges the Secretary with setting annual benchmarks, but it does not change the plain meaning of the operative clause, §8127(d). See Yazoo & Mississippi Valley R. Co. v. Thomas, 132 U. S. 174, 188 (1889) (explaining that prefatory clauses or preambles cannot change the scope of the operative clause). The Federal Circuit’s interpretation also would produce an anomaly. If the Federal Circuit’s understanding of §8127(d)’s prefatory clause were correct, then §§8127(b) and (c), which also contain “[f]or purposes of meeting the goals” clauses, would cease to apply once the Department meets the Secretary’s goal, and the Department would be required to return to competitive bidding. If we interpreted the “purposes” clause of §8127(d) to mean that its mandate no longer applies if the goals are met, then the identical “purposes” clauses of §§8127(b) and (c) would also render those clauses’ permissive mandates inapplic-able. This would require the Department, once the goals are met, to award bids using the default contracting procedures rather than to use the noncompetitive and single-source provisions in §§8127(b) and (c). Second, the Department argues that the mandatory provision does not apply to “orders” under “pre-existing FSS contracts.” Brief for United States 25. The Department failed to raise this argument in the courts below, and we normally decline to entertain such forfeited arguments. See OBB Personenverkehr AG v. Sachs, 577 U. S. ___, ___ (2015) (slip op., at 10). But the Department’s forfeited argument fails in any event. Section 8127(d) applies when the Department “award[s] contracts.” When the Department places an FSS order, that order creates contractual obligations for each party and is a “contract” within the ordinary meaning of that term. See, e.g., Black’s Law Dictionary 389 (10th ed. 2014) (“[a]n agreement between two or more parties creating obligations that are enforce-able or otherwise recognizable at law”). It also creates a “contract” as defined by federal regulations, namely, a “mutually binding legal relationship obligating the seller to furnish the supplies or services . . . and the buyer to pay for them,” including “all types of commitments that obligate the Government to an expenditure of appropriated funds and” (as a general matter) “are in writing.” 48 CFR §2.101 (2015). An FSS order creates mutually binding obligations: for the contractor, to supply certain goods or services, and for the Government, to pay. The placement of the order creates a new contract; the underlying FSS contract gives the Government the option to buy, but it does not require the Government to make a purchase or expend funds. Further confirming that FSS orders are contracts, the Government is not completely bound by the FSS contract’s terms; to the contrary, when placing orders, agencies may sometimes seek different terms than are listed in the FSS. See §8.405–4 (permitting agencies to negotiate some new terms, such as requesting “a price reduction,” when ordering from the FSS). Third, the Department contends that our interpretation fails to appreciate the distinction between FSS orders and contracts. The Department maintains that FSS orders are only for simplified acquisitions, and that using the Rule of Two for these purchases will hamper mundane purchases like “griddles or food slicers.” Brief for United States 21. But this argument understates current practices under the FSS. The Department has expanded use of the FSS well beyond simple procurement. See Brief for Iraq and Afghanistan Veterans of America as Amicus Curiae 14–16. This case proves the point: the contract at issue here concerned complex information technology services over a multiyear period. Moreover, the Department may con-tinue to purchase items that cost less than the simplified acquisition threshold (currently $150,000) through the FSS, if the Department procures them from a veteran-owned small business. See 38 U. S. C. §8127(b). Finally and relatedly, the Department asks us to defer to its interpretation that FSS “orders” are not “contracts.” See Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 –844 (1984) (establishing deference to an agency’s interpretation of an ambiguous statute). Even assuming, arguendo, that the preamble to the agency’s rulemaking could be owed Chevron deference, we do not defer to the agency when the statute is unambiguous. See id., at 842–843. For the reasons already given, the text of §8127(d) clearly imposes a mandatory duty. Thus, we decline the Department’s invitation to defer to its interpretation. * * * We hold that the Rule of Two contracting procedures in §8127(d) are not limited to those contracts necessary to fulfill the Secretary’s goals under §8127(a). We also hold that §8127(d) applies to orders placed under the FSS. The judgment of the Court of Appeals for the Federal Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 This provision reads in full: 2 We use “Department” when referring to the Government as a party in this litigation. 3 Petitioner’s complaint additionally stated claims for two other bid protests. To simplify the proceedings, the parties entered into a joint stipulation of facts concerning only the one bid protest described above. The details concerning the two other disputed bids are relevant only for mootness analysis since the work related to both bids has been performed. See Part II, infra. 4 We need not decide today precisely what sort of search for veteran-owned small businesses the Department must conduct to comply with the Rule of Two. We do not decide, for example, whether the Department may satisfy its obligations by searching for eligible veteran-owned small businesses within the FSS, or whether it must conduct a broader search for such businesses.

579.US.2015_15-375

In Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. ___, this Court held that petitioner Supap Kirtsaeng could invoke the Copyright Act’s “first-sale doctrine,” see 17 U. S. C. §109(a), as a defense to the copyright infringement claim filed by textbook publisher John Wiley & Sons, Inc. Having won his case, Kirtsaeng returned to the District Court to seek more than $2 million in attorney’s fees from Wiley under the Copyright Act’s fee-shifting provision. See §505. The District Court denied Kirtsaeng’s application because, it reasoned, imposing a fee award against a losing party that had taken reasonable positions during litigation (as Wiley had done) would not serve the Act’s purposes. Affirming, the Second Circuit held that the District Court was correct to place “substantial weight” on the reasonableness of Wiley’s position and that the District Court did not abuse its discretion in determining that the other factors did not outweigh the reasonableness finding. Held: 1. When deciding whether to award attorney’s fees under §505, a district court should give substantial weight to the objective reasonableness of the losing party’s position, while still taking into account all other circumstances relevant to granting fees. Pp. 3–11. (a) Section 505 states that a district court “may . . . award a reasonable attorney’s fee to the prevailing party.” Although the text “clearly connotes discretion” and eschews any “precise rule or formula,” Fogerty v. Fantasy, Inc., 510 U. S. 517 , the Court has placed two restrictions on that authority: First, a court may not “award[ ] attorney’s fees as a matter of course,” id., at 533; and second, a court may not treat prevailing plaintiffs and prevailing defendants differently, id., at 527. The Court also noted “several nonexclusive factors” for courts to consider, e.g., “frivolousness, motivation, objective unreasonableness[,] and the need in particular circumstances to advance considerations of compensation and deterrence,” id., at 534, n. 19, and left open the possibility of providing further guidance in the future, id., at 534–535. This Court agrees with both Kirtsaeng and Wiley that additional guidance respecting the application of §505 is proper so as to further channel district court discretion towards the purposes of the Copyright Act. In addressing other open-ended fee-shifting statutes, this Court has emphasized that “in a system of laws discretion is rarely without limits,” and it has “found” those limits by looking to “the large objectives of the relevant Act.” Flight Attendants v. Zipes, 491 U. S. 754 . In accord with such precedents, this Court must determine what approach to fee awards under §505 best advances the well-settled objectives of the Copyright Act, which are to “enrich[ ] the general public through access to creative works” by striking a balance between encouraging and rewarding authors’ creations and enabling others to build on that work. Fogerty, 510 U. S., at 527, 526. Fee awards should thus encourage the types of lawsuits that advance those aims. Pp. 3–6. (b) Wiley’s approach—to put substantial weight on the reasonableness of a losing party’s position—passes this test because it enhances the probability that creators and users (i.e., plaintiffs and defendants) will enjoy the substantive rights the Act provides. Parties with strong positions are encouraged to stand on their rights, given the likelihood that they will recover fees from the losing (i.e., unreasonable) party; those with weak ones are deterred by the likelihood of having to pay two sets of fees. By contrast, Kirtsaeng’s proposal—to give special consideration to whether a suit meaningfully clarified copyright law by resolving an important and close legal issue—would produce no sure benefits. Even accepting that litigation of close cases advances the public interest, fee-shifting will not necessarily, or even usually, encourage parties to litigate those cases to judgment. While fees increase the reward for a victory, they also enhance the penalty for a defeat—and the parties in hard cases cannot be confident if they will win or lose. Wiley’s approach is also more administrable. A district court that has ruled on the merits of a copyright case can easily assess whether the losing party advanced an unreasonable position. By contrast, a judge may not know whether a newly decided issue will have broad legal significance. Pp. 6–10. (c) Still, objective reasonableness can be only a substantial factor in assessing fee applications—not the controlling one. In deciding whether to fee-shift, district courts must take into account a range of considerations beyond the reasonableness of litigating positions. Pp. 10–11. 2. While the Second Circuit properly calls for district courts to give “substantial weight” to the reasonableness of a losing party’s litigating positions, its language at times suggests that a finding of reasonableness raises a presumption against granting fees, and that goes too far in cabining the district court’s analysis. Because the District Court thus may not have understood the full scope of its discretion, it should have the opportunity to reconsider Kirtsaeng’s fee application. On remand, the District Court should continue to give substantial weight to the reasonableness of Wiley’s position but also take into account all other relevant factors. Pp. 11–12. 605 Fed. Appx. 48, vacated and remanded. Kagan, J., delivered the opinion for a unanimous Court.

Section 505 of the Copyright Act provides that a district court “may . . . award a reasonable attorney’s fee to the prevailing party.” 17 U. S. C. §505. The question pre-sented here is whether a court, in exercising that author-ity, should give substantial weight to the objective reasonableness of the losing party’s position. The answer, as both decisions below held, is yes—the court should. But the court must also give due consideration to all other circumstances relevant to granting fees; and it retains discretion, in light of those factors, to make an award even when the losing party advanced a reasonable claim or defense. Because we are not certain that the lower courts here understood the full scope of that discretion, we return the case for further consideration of the prevailing party’s fee application. I Petitioner Supap Kirtsaeng, a citizen of Thailand, came to the United States 20 years ago to study math at Cornell University. He quickly figured out that respondent John Wiley & Sons, an academic publishing company, sold virtually identical English-language textbooks in the two countries—but for far less in Thailand than in the United States. Seeing a ripe opportunity for arbitrage, Kirtsaeng asked family and friends to buy the foreign editions in Thai bookstores and ship them to him in New York. He then resold the textbooks to American students, reimbursed his Thai suppliers, and pocketed a tidy profit. Wiley sued Kirtsaeng for copyright infringement, claiming that his activities violated its exclusive right to distribute the textbooks. See 17 U. S. C. §§106(3), 602(a)(1). Kirtsaeng invoked the “first-sale doctrine” as a defense. That doctrine typically enables the lawful owner of a book (or other work) to resell or otherwise dispose of it as he wishes. See §109(a). But Wiley contended that the first-sale doctrine did not apply when a book (like those Kirtsaeng sold) was manufactured abroad. At the time, courts were in conflict on that issue. Some thought, as Kirtsaeng did, that the first-sale doctrine permitted the resale of foreign-made books; others maintained, along with Wiley, that it did not. And this Court, in its first pass at the issue, divided 4 to 4. See Costco Wholesale Corp. v. Omega, S. A., 562 U. S. 40 (2010) ( per curiam). In this case, the District Court sided with Wiley; so too did a divided panel of the Court of Appeals for the Second Circuit. See 654 F. 3d 210, 214, 222 (2011). To settle the continuing conflict, this Court granted Kirtsaeng’s petition for certiorari and reversed the Second Circuit in a 6-to-3 decision, thus establishing that the first-sale doctrine allows the resale of foreign-made books, just as it does domestic ones. See Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. ___, ___ (2013) (slip op., at 3). Returning victorious to the District Court, Kirtsaeng invoked §505 to seek more than $2 million in attorney’s fees from Wiley. The court denied his motion. Relying on Second Circuit precedent, the court gave “substantial weight” to the “objective reasonableness” of Wiley’s infringement claim. See No. 08–cv–07834 (SDNY, Dec. 20, 2013), App. to Pet. for Cert. 18a, 2013 WL 6722887, *4. In explanation of that approach, the court stated that “the imposition of a fee award against a copyright holder with an objectively reasonable”—although unsuccessful—“lit-igation position will generally not promote the purposes of the Copyright Act.” Id., at 11a (quoting Matthew Bender & Co. v. West Publishing Co., 240 F. 3d 116, 122 (CA2 2001) (emphasis deleted)). Here, Wiley’s position was reasonable: After all, several Courts of Appeals and three Justices of the Supreme Court had agreed with it. See App. to Pet. for Cert. 12a. And according to the District Court, no other circumstance “overr[o]de” that objective reasonableness, so as to warrant fee-shifting. Id., at 22a. The Court of Appeals affirmed, concluding in a brief summary order that “the district court properly placed ‘substantial weight’ on the reasonableness of [Wiley’s] position” and committed no abuse of discretion in deciding that other “factors did not outweigh” the reasonableness finding. 605 Fed. Appx. 48, 49, 50 (CA2 2015). We granted certiorari, 577 U. S. ___ (2016), to resolve disagreement in the lower courts about how to address an application for attorney’s fees in a copyright case.[1] II Section 505 states that a district court “may . . . award a reasonable attorney’s fee to the prevailing party.” It thus authorizes fee-shifting, but without specifying standards that courts should adopt, or guideposts they should use, in determining when such awards are appropriate. In Fogerty v. Fantasy, Inc., 510 U. S. 517 (1994) , this Court recognized the broad leeway §505 gives to district courts—but also established several principles and criteria to guide their decisions. See id., at 519 (asking “what standards should inform” the exercise of the trial court’s authority). The statutory language, we stated, “clearly connotes discretion,” and eschews any “precise rule or formula” for awarding fees. Id., at 533, 534. Still, we established a pair of restrictions. First, a district court may not “award[ ] attorney’s fees as a matter of course”; rather, a court must make a more particularized, case-by-case assessment. Id., at 533. Second, a court may not treat prevailing plaintiffs and prevailing defendants any differently; defendants should be “encouraged to litigate [meritorious copyright defenses] to the same extent that plaintiffs are encouraged to litigate meritorious claims of infringement.” Id., at 527. In addition, we noted with approval “several nonexclusive factors” to inform a court’s fee-shifting decisions: “frivolousness, motivation, objective unreasonableness[,] and the need in particular circumstances to advance considerations of compensation and deterrence.” Id., at 534, n. 19. And we left open the possibility of providing further guidance in the future, in response to (and grounded on) lower courts’ evolving experience. See id., at 534–535; Martin v. Franklin Capital Corp., 546 U. S. 132 , n. (2005) (noting that Fogerty was not intended to be the end of the matter). The parties here, though sharing some common ground, now dispute what else we should say to district courts. Both Kirtsaeng and Wiley agree—as they must—that §505 grants courts wide latitude to award attorney’s fees based on the totality of circumstances in a case. See Brief for Petitioner 17; Brief for Respondent 35. Yet both reject the position, taken by some Courts of Appeals, see supra, at 3, n. 1, that Fogerty spelled out the only appropriate limits on judicial discretion—in other words, that each district court should otherwise proceed as it sees fit, assigning whatever weight to whatever factors it chooses. Rather, Kirtsaeng and Wiley both call, in almost identical language, for “[c]hanneling district court discretion towards the purposes of the Copyright Act.” Brief for Petitioner 16; see Brief for Respondent 21 (“[A]n appellate court [should] channel a district court’s discretion so that it . . . further[s] the goals of the Copyright Act”). (And indeed, as discussed later, both describe those purposes identically. See infra, at 6.) But at that point, the two part ways. Wiley argues that giving substantial weight to the reasonableness of a losing party’s position will best serve the Act’s objectives. See Brief for Respondent 24–35. By contrast, Kirtsaeng favors giving special consideration to whether a lawsuit resolved an important and close legal issue and thus “meaningfully clarifie[d]” copyright law. Brief for Petitioner 36; see id., at 41–44. We join both parties in seeing a need for some additional guidance respecting the application of §505. In addressing other open-ended fee-shifting statutes, this Court has emphasized that “in a system of laws discretion is rarely without limits.” Flight Attendants v. Zipes, 491 U. S. 754, 758 (1989) ; see Halo Electronics, Inc. v. Pulse Electronics, Inc., ante, at 8. Without governing standards or principles, such provisions threaten to condone judicial “whim” or predilection. Martin, 546 U. S., at 139; see also ibid. (“[A] motion to [a court’s] discretion is a motion, not to its inclination, but to its judgment; and its judgment is to be guided by sound legal principles” (quoting United States v. Burr, 25 F. Cas. 30, 35 (No. 14,692d) (CC Va. 1807) (Marshall, C. J.))). At the least, utterly freewheeling inquiries often deprive litigants of “the basic principle of justice that like cases should be decided alike,” Martin, 546 U. S., at 139—as when, for example, one judge thinks the parties’ “motivation[s]” determinative and another believes the need for “compensation” trumps all else, Fogerty, 510 U. S., at 534, n. 19. And so too, such unconstrained discretion prevents individuals from predicting how fee decisions will turn out, and thus from making properly informed judgments about whether to litigate. For those reasons, when applying fee-shifting laws with “no explicit limit or condition,” Halo, ante, at 8, we have nonetheless “found limits” in them—and we have done so, just as both parties urge, by looking to “the large objectives of the relevant Act,” Zipes, 491 U. S., at 759 (internal quotation marks omitted); see supra, at 5. In accord with such precedents, we must consider if either Wiley’s or Kirtsaeng’s proposal well advances the Copyright Act’s goals. Those objectives are well settled. As Fogerty explained, “copyright law ultimately serves the purpose of enriching the general public through access to creative works.” 510 U. S., at 527; see U. S. Const., Art. I, §8, cl. 8 (“To promote the Progress of Science and useful Arts”). The statute achieves that end by striking a balance between two subsidiary aims: encouraging and rewarding authors’ creations while also enabling others to build on that work. See Fogerty, 510 U. S., at 526. Accordingly, fee awards under §505 should encourage the types of lawsuits that promote those purposes. (That is why, for example, Fogerty insisted on treating prevailing plaintiffs and prevailing defendants alike—because the one could “further the policies of the Copyright Act every bit as much as” the other. 510 U. S., at 527.) On that much, both parties agree. Brief for Petitioner 37; Brief for Respondent 29–30. The contested issue is whether giving substantial weight to the objective (un)reasonableness of a losing party’s litigating position—or, alternatively, to a lawsuit’s role in settling significant and uncertain legal issues—will predictably encourage such useful copyright litigation. The objective-reasonableness approach that Wiley favors passes that test because it both encourages parties with strong legal positions to stand on their rights and deters those with weak ones from proceeding with litigation. When a litigant—whether plaintiff or defendant—is clearly correct, the likelihood that he will recover fees from the opposing (i.e., unreasonable) party gives him an incentive to litigate the case all the way to the end. The holder of a copyright that has obviously been infringed has good reason to bring and maintain a suit even if the damages at stake are small; and likewise, a person defending against a patently meritless copyright claim has every incentive to keep fighting, no matter that attorney’s fees in a pro-tracted suit might be as or more costly than a settlement. Conversely, when a person (again, whether plaintiff or defendant) has an unreasonable litigating position, the likelihood that he will have to pay two sets of fees discourages legal action. The copyright holder with no reasonable infringement claim has good reason not to bring suit in the first instance (knowing he cannot force a settlement and will have to proceed to judgment); and the infringer with no reasonable defense has every reason to give in quickly, before each side’s litigation costs mount. All of those results promote the Copyright Act’s purposes, by enhancing the probability that both creators and users (i.e., potential plaintiffs and defendants) will enjoy the substantive rights the statute provides. By contrast, Kirtsaeng’s proposal would not produce any sure benefits. We accept his premise that litigation of close cases can help ensure that “the boundaries of copyright law [are] demarcated as clearly as possible,” thus advancing the public interest in creative work. Brief for Petitioner 19 (quoting Fogerty, 510 U. S., at 527). But we cannot agree that fee-shifting will necessarily, or even usually, encourage parties to litigate those cases to judgment. Fee awards are a double-edged sword: They increase the reward for a victory—but also enhance the penalty for a defeat. And the hallmark of hard cases is that no party can be confident if he will win or lose. That means Kirtsaeng’s approach could just as easily discourage as encourage parties to pursue the kinds of suits that “meaningfully clarif[y]” copyright law. Brief for Petitioner 36. It would (by definition) raise the stakes of such suits; but whether those higher stakes would provide an incentive—or instead a disincentive—to litigate hinges on a party’s attitude toward risk. Is the person risk-preferring or risk-averse—a high-roller or a penny-ante type? Only the former would litigate more in Kirtsaeng’s world. See Posner, An Economic Approach to Legal Procedure and Judicial Administration, 2 J. Legal Studies 399, 428 (1973) (fees “make[ ] the expected value of litigation less for risk-averse litigants, which will encourage [them to] settle[ ]”). And Kirtsaeng offers no reason to think that serious gamblers predominate. See, e.g., Texas Industries, Inc. v. Radcliff Materials, Inc., 451 U. S. 630 , n. 8 (1981) (“Economists disagree over whether business decisionmakers[ ] are ‘risk averse’ ”); CIGNA Corp. v. Amara, 563 U. S. 421, 430 (2011) (“[M]ost individuals are risk averse”). So the value of his standard, unlike Wiley’s, is entirely speculative.[2] What is more, Wiley’s approach is more administrable than Kirtsaeng’s. A district court that has ruled on the merits of a copyright case can easily assess whether the losing party advanced an unreasonable claim or defense. That is closely related to what the court has already done: In deciding any case, a judge cannot help but consider the strength and weakness of each side’s arguments. By contrast, a judge may not know at the conclusion of a suit whether a newly decided issue will have, as Kirtsaeng thinks critical, broad legal significance. The precedent-setting, law-clarifying value of a decision may become apparent only in retrospect—sometimes, not until many years later. And so too a decision’s practical impact (to the extent Kirtsaeng would have courts separately consider that factor). District courts are not accustomed to evaluating in real time either the jurisprudential or the on-the-ground import of their rulings. Exactly how they would do so is uncertain (Kirtsaeng points to no other context in which courts undertake such an analysis), but we fear that the inquiry would implicate our oft-stated concern that an application for attorney’s fees “should not result in a second major litigation.” Zipes, 491 U. S., at 766 (quoting Hensley v. Eckerhart, 461 U. S. 424, 437 (1983) ). And we suspect that even at the end of that post-lawsuit lawsuit, the results would typically reflect little more than edu-cated guesses. Contrary to Kirtsaeng’s view, placing substantial weight on objective reasonableness also treats plaintiffs and defendants even-handedly, as Fogerty commands. No matter which side wins a case, the court must assess whether the other side’s position was (un)reasonable. And of course, both plaintiffs and defendants can (and sometimes do) make unreasonable arguments. Kirtsaeng claims that the reasonableness inquiry systematically favors plaintiffs because a losing defendant “will virtually always be found to have done something culpable.” Brief for Petitioner 29 (emphasis in original). But that conflates two different questions: whether a defendant in fact infringed a copyright and whether he made serious arguments in defense of his conduct. Courts every day see reasonable defenses that ultimately fail (just as they see reasonable claims that come to nothing); in this context, as in any other, they are capable of distinguishing between those defenses (or claims) and the objectively unreason-able variety. And if some court confuses the issue of liability with that of reasonableness, its fee award should be reversed for abuse of discretion.[3] All of that said, objective reasonableness can be only an important factor in assessing fee applications—not the controlling one. As we recognized in Fogerty, §505 con-fers broad discretion on district courts and, in deciding whether to fee-shift, they must take into account a range of considerations beyond the reasonableness of litigating positions. See supra, at 4. That means in any given case a court may award fees even though the losing party offered reasonable arguments (or, conversely, deny fees even though the losing party made unreasonable ones). For example, a court may order fee-shifting because of a party’s litigation misconduct, whatever the reasonableness of his claims or defenses. See, e.g., Viva Video, Inc. v. Cabrera, 9 Fed. Appx. 77, 80 (CA2 2001). Or a court may do so to deter repeated instances of copyright infringement or overaggressive assertions of copyright claims, again even if the losing position was reasonable in a particular case. See, e.g., Bridgeport Music, Inc. v. WB Music Corp., 520 F. 3d 588, 593–595 (CA6 2008) (awarding fees against a copyright holder who filed hundreds of suits on an overbroad legal theory, including in a subset of cases in which it was objectively reasonable). Although objective reasonableness carries significant weight, courts must view all the circumstances of a case on their own terms, in light of the Copyright Act’s essential goals. And on that score, Kirtsaeng has raised serious questions about how fee-shifting actually operates in the Second Circuit. To be sure, the Court of Appeals’ framing of the inquiry resembles our own: It calls for a district court to give “substantial weight” to the reasonableness of a losing party’s litigating positions while also considering other relevant circumstances. See 605 Fed. Appx., at 49–50; Matthew Bender, 240 F. 3d, at 122. But the Court of Appeals’ language at times suggests that a finding of reasonableness raises a presumption against granting fees, see ibid.; supra, at 2–3—and that goes too far in cabining how a district court must structure its analysis and what it may conclude from its review of relevant factors. Still more, district courts in the Second Circuit appear to have overly learned the Court of Appeals’ lesson, turning “substantial” into more nearly “dispositive” weight. As Kirtsaeng notes, hardly any decisions in that Circuit have granted fees when the losing party raised a reasonable argument (and none have denied fees when the losing party failed to do so). See Reply Brief 15. For these reasons, we vacate the decision below so that the District Court can take another look at Kirtsaeng’s fee application. In sending back the case for this purpose, we do not at all intimate that the District Court should reach a different conclusion. Rather, we merely ensure that the court will evaluate the motion consistent with the analysis we have set out—giving substantial weight to the reasonableness of Wiley’s litigating position, but also taking into account all other relevant factors. * * * The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Compare, e.g., Matthew Bender & Co. v. West Publishing Co., 240 F. 3d 116, 122 (CA2 2001) (giving substantial weight to objective reasonableness), with, e.g., Bond v. Blum, 317 F. 3d 385, 397–398 (CA4 2003) (endorsing a totality-of-the-circumstances approach, without according special significance to any factor), and with, e.g., Hogan Systems, Inc. v. Cybersource Int’l, Inc., 158 F. 3d 319, 325 (CA5 1998) (presuming that a prevailing party receives fees). 2 This case serves as a good illustration. Imagine you are Kirtsaeng at a key moment in his case—say, when deciding whether to petition this Court for certiorari. And suppose (as Kirtsaeng now wishes) that the prevailing party in a hard and important case—like this one—will probably get a fee award. Does that make you more likely to file, because you will recoup your own fees if you win? Or less likely to file, because you will foot Wiley’s bills if you lose? Here are some answers to choose from (recalling that you cannot confidently predict which way the Court will rule): (A) Six of one, half a dozen of the other. (B) Depends if I’m feeling lucky that day. (C) Less likely—this is getting scary; who knows how much money Wiley will spend on Supreme Court lawyers? (D) More likely—the higher the stakes, the greater the rush. Only if lots of people answer (D) will Kirtsaeng’s standard work in the way advertised. Maybe. But then again, maybe not. 3 Kirtsaeng also offers statistics meant to show that in practice, even if not in theory, the objective reasonableness inquiry unduly favors plaintiffs; but the Solicitor General as amicus curiae has cast significant doubt on that claim. According to Kirtsaeng, 86% of winning copyright holders, but only 45% of prevailing defendants, have received fee awards over the last 15 years in the Second Circuit (which, recall, gives substantial weight to objective reasonableness). See Reply Brief 17–18; supra, at 2–3. But first, the Solicitor General represents that the overall numbers are actually 77% and 53%, respectively. See Tr. of Oral Arg. 41. And second, the Solicitor General points out that all these percentages include default judgments, which almost invariably give rise to fee awards—but usually of a very small amount—because the defendant has not shown up to oppose either the suit or the fee application. When those cases are taken out, the statistics look fairly similar: 60% for plaintiffs versus 53% for defendants. See id., at 42. And of course, there may be good reasons why copyright plaintiffs and defendants do not make reasonable arguments in perfectly equal proportion.

577.US.2015_14-8358

Petitioner Avondale Lockhart pleaded guilty to possessing child pornography in violation of 18 U. S. C. §2252(a)(4). Because Lockhart had a prior state-court conviction for first-degree sexual abuse involving his adult girlfriend, his presentence report concluded that he was subject to the 10-year mandatory minimum sentence enhancement provided in §2252(b)(2), which is triggered by, inter alia, prior state convictions for crimes “relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.” Lockhart argued that the limiting phrase “involving a minor or ward” applied to all three state crimes, so his prior conviction did not trigger the enhancement. Disagreeing, the District Court applied the mandatory minimum. The Second Circuit affirmed. Held: Lockhart’s prior conviction is encompassed by §2252(b)(2). Pp. 2–15. (a) A natural reading of the text supports that conclusion. The “rule of the last antecedent,” a canon of statutory interpretation stating that “a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows,” Barnhart v. Thomas, 540 U. S. 20 , clarifies that the phrase “involving a minor or ward” modifies only the immediately preceding noun phrase “abusive sexual conduct” and that the phrases “aggravated sexual abuse” and “sexual abuse” are not so restricted. The rule “can . . . be overcome by other indicia of meaning,” ibid., but §2252(b)(2)’s context reinforces its application in this case. Pp. 2–5. (b) Section 2252(b)(2)’s enhancement can also be triggered by, inter alia, a prior federal sexual abuse offense enumerated in Chapter 109A of the Federal Criminal Code. Interpreting §2252(b)(2) using the “rule of the last antecedent,” the headings in Chapter 109A mirror precisely the order, precisely the divisions, and nearly precisely the words used to describe the state sexual-abuse predicates. Applying the modifier “involving a minor or ward” to all three items in §2252(b)(2)’s list, by contrast, would require this Court to interpret the state predicates in a way that departs from the federal template. If Congress had intended that result, it is doubtful that Congress would have followed so closely the structure and language of Chapter 109A. Pp. 5–7. (c) Lockhart’s counterarguments are rejected. Pp. 7–14. (1) Porto Rico Railway, Light & Power Co. v. Mor, 253 U. S. 345 , United States v. Bass, 404 U. S. 336 , and Jama v. Immigration and Customs Enforcement, 543 U. S. 335 , do not require this Court to apply Lockhart’s countervailing series-qualifier principle. In those cases, the Court simply observed that the last-antecedent rule may be overcome by contextual indicia of meaning. Lockhart’s attempts to identify such indicia are unavailing. He claims that the state predicates are so similar that a limiting phrase could apply equally to all three. But by transforming a list of separate predicates into a set of near-synonyms, Lockhart’s reading results in too much redundancy and risks running headlong into the rule against superfluity. Pp. 7–10. (2) Lockhart contends that the existence of other disparities between §2252(b)(2)’s state and federal sexual-abuse predicates indicate that parity was not Congress’ concern. However, this Court’s construction relies on contextual cues particular to the sexual-abuse predicates, not on a general assumption that Congress sought full parity between all state and federal predicates. Pp. 10–11. (3) The provision’s legislative history “hardly speaks with [a] clarity of purpose,” Universal Camera Corp. v. NLRB, 340 U. S. 474 , and does nothing to explain why Congress would have wanted to structure §2252(b)(2) to treat state and federal predicates differently. Pp. 11–14. (4) Finally, Lockhart suggests the rule of lenity is triggered here, where applying his series-qualifier principle would lead to an alternative construction of §2252(b)(2). The rule of lenity is used to resolve ambiguity only when the ordinary canons have revealed no satisfactory construction. Here, however, the rule of the last antecedent is well supported by context, and Lockhart’s alternative is not. P. 14. 749 F. 3d 148, affirmed. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, and Alito, JJ., joined. Kagan, J., filed a dissenting opinion, in which Breyer, J., joined.

Defendants convicted of possessing child pornography in violation of 18 U. S. C. §2252(a)(4) are subject to a 10-year mandatory minimum sentence and an increased maximum sentence if they have “a prior conviction . . . under the laws of any State relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.” §2252(b)(2). The question before us is whether the phrase “involving a minor or ward” modifies all items in the list of predicate crimes (“aggravated sexual abuse,” “sexual abuse,” and “abusive sexual conduct”) or only the one item that immediately precedes it (“abusive sexual conduct”). Below, the Court of Appeals for the Second Circuit joined several other Courts of Appeals in holding that it modifies only “abusive sexual conduct.” The Eighth Circuit has reached the contrary result. We granted certiorari to resolve that split. 575 U. S. ___ (2015). We affirm the Second Circuit’s holding that the phrase “involving a minor or ward” in §2252(b)(2) modifies only “abusive sexual conduct.” I In April 2000, Avondale Lockhart was convicted of sexual abuse in the first degree under N. Y. Penal Law Ann. §130.65(1) (West Cum. Supp. 2015). The crime involved his then-53-year-old girlfriend. Presentence Investigation Report (PSR), in No. 11–CR–231–01, p. 13, ¶¶47–48. Eleven years later, Lockhart was indicted in the Eastern District of New York for attempting to receive child pornography in violation of 18 U. S. C. §2252(a)(2) and for possessing child pornography in violation of §2252(a)(4)(b). Lockhart pleaded guilty to the possession offense and the Government dismissed the receipt offense. Lockhart’s presentence report calculated a guidelines range of 78 to 97 months for the possession offense. But the report also concluded that Lockhart was subject to §2252(b)(2)’s mandatory minimum because his prior New York abuse conviction related “to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward.” PSR ¶¶87–88. Lockhart objected, arguing that the statutory phrase “involving a minor or ward” applies to all three listed crimes: “aggravated sexual abuse,” “sexual abuse,” and “abusive sexual conduct.” He therefore contended that his prior conviction for sexual abuse involving an adult fell outside the enhancement’s ambit. The District Court rejected Lockhart’s argument and applied the mandatory minimum. The Second Circuit affirmed his sentence. 749 F. 3d 148 (CA2 2014). II Section 2252(b)(2) reads in full: “Whoever violates, or attempts or conspires to violate [ 18 U. S. C. §2252(a)(4)] shall be fined under this title or imprisoned not more than 10 years, or both, but . . . if such person has a prior conviction under this chapter, chapter 71, chapter 109A, or chapter 117, or under section 920 of title 10 (article 120 of the Uniform Code of Military Justice), or under the laws of any State relating to aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward, or the production, possession, receipt, mailing, sale, distribution, shipment, or transportation of child pornography, such person shall be fined under this title and imprisoned for not less than 10 years nor more than 20 years.” This case concerns that provision’s list of state sexual-abuse offenses. The issue before us is whether the limiting phrase that appears at the end of that list—“involving a minor or ward”—applies to all three predicate crimes preceding it in the list or only the final predicate crime. We hold that “involving a minor or ward” modifies only “abusive sexual conduct,” the antecedent immediately preceding it. Although §2252(b)(2)’s list of state predicates is awkwardly phrased (to put it charitably), the provision’s text and context together reveal a straightforward reading. A timeworn textual canon is confirmed by the structure and internal logic of the statutory scheme. A Consider the text. When this Court has interpreted statutes that include a list of terms or phrases followed by a limiting clause, we have typically applied an interpretive strategy called the “rule of the last antecedent.” See Barnhart v. Thomas, 540 U. S. 20, 26 (2003) . The rule provides that “a limiting clause or phrase . . . should ordinarily be read as modifying only the noun or phrase that it immediately follows.” Ibid.; see also Black’s Law Dictionary 1532–1533 (10th ed. 2014) (“[Q]ualifying words or phrases modify the words or phrases immediately preceding them and not words or phrases more remote, unless the extension is necessary from the context or the spirit of the entire writing”); A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 144 (2012). This Court has applied the rule from our earliest decisions to our more recent. See, e.g., Sims Lessee v. Irvine, 3 Dall. 425, 444, n. (1799); FTC v. Mandel Brothers, Inc., 359 U. S. 385, 389, n. 4 (1959) ; Barnhart, 540 U. S., at 26. The rule reflects the basic intuition that when a modifier appears at the end of a list, it is easier to apply that modifier only to the item directly before it. That is particularly true where it takes more than a little mental energy to process the individual entries in the list, making it a heavy lift to carry the modifier across them all. For example, imagine you are the general manager of the Yankees and you are rounding out your 2016 roster. You tell your scouts to find a defensive catcher, a quick-footed shortstop, or a pitcher from last year’s World Champion Kansas City Royals. It would be natural for your scouts to confine their search for a pitcher to last year’s championship team, but to look more broadly for catchers and shortstops. Applied here, the last antecedent principle suggests that the phrase “involving a minor or ward” modifies only the phrase that it immediately follows: “abusive sexual conduct.” As a corollary, it also suggests that the phrases “aggravated sexual abuse” and “sexual abuse” are not so constrained. Of course, as with any canon of statutory interpretation, the rule of the last antecedent “is not an absolute and can assuredly be overcome by other indicia of meaning.” Barnhart, 540 U. S., at 26; see also Davis v. Michigan Dept. of Treasury, 489 U. S. 803, 809 (1989) (“It is a fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme”). For instance, take “ ‘the laws, the treaties, and the constitution of the United States.’ ” Post, at 7, n. 1 (Kagan, J., dissenting). A reader intuitively applies “of the United States” to “the laws,” “the treaties” and “the constitution” because (among other things) laws, treaties, and the constitution are often cited together, because readers are used to seeing “of the United States” modify each of them, and because the listed items are simple and parallel without unexpected internal modifiers or structure. Section 2252(b)(2), by contrast, does not contain items that readers are used to seeing listed together or a concluding modifier that readers are accustomed to applying to each of them. And the varied syntax of each item in the list makes it hard for the reader to carry the final modifying clause across all three. More importantly, here the interpretation urged by the rule of the last antecedent is not overcome by other indicia of meaning. To the contrary, §2252(b)(2)’s context fortifies the meaning that principle commands. B Our inquiry into §2252(b)(2)’s context begins with the internal logic of that provision. Section 2252(b)(2) establishes sentencing minimums and maximums for three categories of offenders. The first third of the section imposes a 10-year maximum sentence on offenders with no prior convictions. The second third imposes a 10-year minimum and 20-year maximum on offenders who have previously violated a federal offense listed within various chapters of the Federal Criminal Code. And the last third imposes the same minimum and maximum on offenders who have previously committed state “sexual abuse,aggravated sexual abuse, or abusive sexual conduct involving a minor or ward” as well as a number of state crimes related to the possession and distribution of child pornography. Among the chapters of the Federal Criminal Code that can trigger §2252(b)(2)’s recidivist enhancement are crimes “under . . . chapter 109A.” Chapter 109A criminal- izes a range of sexual-abuse offenses involving adults or minors and wards.[1] And it places those federal sexual-abuse crimes under headings that use language nearly identical to the language §2252(b)(2) uses to enumerate the three categories of state sexual-abuse predicates. The first section in Chapter 109A is titled “Aggravated sexual abuse.” 18 U. S. C. §2241. The second is titled “Sexual abuse.” §2242. And the third is titled “Sexual abuse of a minor or ward.” §2243. Applying the rule of the last antecedent, those sections mirror precisely the order, precisely the divisions, and nearly precisely the words used to describe the three state sexual-abuse predicate crimes in §2252(b)(2): “aggravated sexual abuse,” “sexual abuse,” and “abusive sexual conduct involving a minor or ward.” This similarity appears to be more than a coincidence. We cannot state with certainty that Congress used Chapter 109A as a template for the list of state predicates set out in §2252(b)(2), but we cannot ignore the parallel, particularly because the headings in Chapter 109A were in place when Congress amended the statute to add §2252(b)(2)’s state sexual-abuse predicates.[2] If Congress had intended to limit each of the state predicates to conduct “involving a minor or ward,” we doubt it would have followed, or thought it needed to follow, so closely the structure and language of Chapter 109A.[3] The conclusion that Congress followed the federal template is supported by the fact that Congress did nothing to indicate that offenders with prior federal sexual-abuse convictions are more culpable, harmful, or worthy of enhanced punishment than offenders with nearly identical state priors. We therefore see no reason to interpret §2252(b)(2) so that “[s]exual abuse” that occurs in the Second Circuit courthouse triggers the sentence enhancement, but “sexual abuse” that occurs next door in the Manhattan munici-pal building does not. III A Lockhart argues, to the contrary, that the phrase “involving a minor or ward” should be interpreted to modify all three state sexual-abuse predicates. He first contends, as does our dissenting colleague, that the so-called series-qualifier principle supports his reading. This principle, Lockhart says, requires a modifier to apply to all items in a series when such an application would represent a natural construction. Brief for Petitioner 12; post, at 4. This Court has long acknowledged that structural or contextual evidence may “rebut the last antecedent inference.” Jama v. Immigration and Customs Enforcement, 543 U. S. 335 , n. 4 (2005). For instance, in Porto Rico Railway, Light & Power Co. v. Mor, 253 U. S. 345 (1920) , on which Lockhart relies, this Court declined to apply the rule of the last antecedent where “[n]o reason appears why” a modifying clause is not “applicable as much to the first and other words as to the last” and where “special reasons exist for so construing the clause in question.” Id., at 348. In United States v. Bass, 404 U. S. 336 (1971) , this Court declined to apply the rule of the last antecedent where “there is no reason consistent with any discernable purpose of the statute to apply” the limiting phrase to the last antecedent alone. Id., at 341. Likewise, in Jama, the Court suggested that the rule would not be appropriate where the “modifying clause appear[s] . . . at the end of a single, integrated list.” 543 U. S., at 344, n. 4. And, most recently, in Paroline v. United States, 572 U. S. ___ (2014), the Court noted that the rule need not be applied “in a mechanical way where it would require accepting ‘unlikely premises.’ ” Id., at ___ (slip op., at 9). But in none of those cases did the Court describe, much less apply, a countervailing grammatical mandate that could bear the weight that either Lockhart or the dissent places on the series qualifier principle. Instead, the Court simply observed that sometimes context weighs against the application of the rule of the last antecedent. Barnhart, 540 U. S., at 26. Whether a modifier is “applicable as much to the first . . . as to the last” words in a list, whether a set of items form a “single, integrated list,” and whether the application of the rule would require acceptance of an “unlikely premise” are fundamentally contextual questions. Lockhart attempts to identify contextual indicia that he says rebut the rule of the last antecedent, but those indicia hurt rather than help his prospects. He points out that the final two state predicates, “sexual abuse” and “abusive sexual conduct,” are “nearly synonymous as a matter of everyday speech.” Brief for Petitioner 17. And, of course, anyone who commits “aggravated sexual abuse” has also necessarily committed “sexual abuse.” So, he posits, the items in the list are sufficiently similar that a limiting phrase could apply equally to all three of them. But Lockhart’s effort to demonstrate some similarity among the items in the list of state predicates reveals far too much similarity. The three state predicate crimes are not just related on Lockhart’s reading; they are hopelessly redundant. Any conduct that would qualify as “aggravated sexual abuse . . . involving a minor or ward” or “sexual abuse . . . involving a minor or ward” would also qualify as “abusive sexual conduct involving a minor or ward.” We take no position today on the meaning of the terms “aggravated sexual abuse,” “sexual abuse,” and “abusive sexual conduct,” including their similarities and differences. But it is clear that applying the limiting phrase to all three items would risk running headlong into the rule against superfluity by transforming a list of separate predicates into a set of synonyms describing the same predicate. See Bailey v. United States, 516 U. S. 137, 146 (1995) (“We assume that Congress used two terms because it intended each term to have a particular, nonsuperfluous meaning”). Applying the limiting phrase “involving a minor or ward” more sparingly, by contrast, preserves some distinction between the categories of state predicates by limiting only the third category to conduct “involving a minor or ward.” We recognize that this interpretation does not eliminate all superfluity between “aggravated sexual abuse” and “sexual abuse.” See United States v. Atlantic Research Corp., 551 U. S. 128, 137 (2007) (“[O]ur hesitancy to construe statutes to render language superfluousdoes not require us to avoid surplusage at all costs. It is appropriate to tolerate a degree of surplusage”). But there is a ready explanation for the redundancy that remains: It follows the categories in Chapter 109A’s federal template. See supra, at 6. We see no similar explanation for Lockhart’s complete collapse of the list. The dissent offers a suggestion rooted in its impressions about how people ordinarily speak and write. Post, at 1–4. The problem is that, as even the dissent acknowledges, §2252(b)(2)’s list of state predicates is hardly intuitive. No one would mistake its odd repetition and inelegant phrasing for a reflection of the accumulated wisdom of everyday speech patterns. It would be as if a friend asked you to get her tart lemons, sour lemons, or sour fruit from Mexico. If you brought back lemons from California, but your friend insisted that she was using customary speech and obvi-ously asked for Mexican fruit only, you would be forgiven for disagreeing on both counts. Faced with §2252(b)(2)’s inartful drafting, then, do we interpret the provision by viewing it as a clear, commonsense list best construed as if conversational English? Or do we look around to see if there might be some provenance to its peculiarity? With Chapter 109A so readily at hand, we are unpersuaded by our dissenting colleague’s invocation of basic examples from day-to-day life. Whatever the validity of the dissent’s broader point, this simply is not a case in which colloquial practice is of much use. Section 2252(b)(2)’s list is hardly the way an average person, or even an average lawyer, would set about to describe the relevant conduct if they had started from scratch. B Lockhart next takes aim at our construction of §2252(b)(2) to avoid disparity between the state and federal sexual-abuse predicates. He contends that other dispar-ities between state and federal predicates in §2252(b)(2) indicate that parity was not Congress’ concern. For example, §2252(b)(2) imposes the recidivist enhancement on offenders with prior federal convictions under Chapter 71 of Title 18, which governs obscenity. See §§1461–1470. Yet §2252(b)(2) does not impose a similar enhancementfor offenses under state obscenity laws. Similarly, §2252(b)(2)’s neighbor provision, §2252(b)(1), creates a mandatory minimum for sex trafficking involving children, but not sex trafficking involving adults. However, our construction of §2252(b)(2)’s sexual-abuse predicates does not rely on a general assumption that Congress sought full parity between all of the federal and state predicates in §2252(b)(2). It relies instead on contextual cues particular to the sexual-abuse predicates. To enumerate the state sexual-abuse predicates, Congress used language similar to that in Chapter 109A of the Federal Criminal Code, which describes crimes involving both adults and children. See supra, at 6. We therefore assume that the same language used to describe the state sexual-abuse predicates also describes conduct involving both adults and children. C Lockhart, joined by the dissent, see post, at 9–11, next says that the provision’s legislative history supports the view that Congress deliberately structured §2252(b)(2) to treat state and federal predicates differently. They rely on two sources. The first is a reference in a Report from the Senate Judiciary Committee on the Child Pornography Prevention Act of 1996, 110Stat. 3009–26. That Act was the first to add the language at issue here—“aggravated sexual abuse, sexual abuse, or abusive sexual conduct involving a minor or ward”—to the U. S. Code. (It was initially added to §2252(b)(1), then added two years later to §2252(b)(2)). The Report noted that the enhancement applies to persons with prior convictions “under any State child abuse law or law relating to the production, receipt or distribution of child pornography.” See S. Rep. No. 104–358, p. 9 (1996). But that reference incompletely describes the state pornography production and distribution predicates, which cover not only “production, receipt, or distributing of child pornography,” as the Report indicates, but also “production, possession, receipt, mailing, sale, distribution, shipment, or transportation of child pornography,” §2252(b)(2). For the reasons discussed, we have no trouble concluding that the Report also incompletely describes the state sexual-abuse predicates. Lockhart and the dissent also rely on a letter sent from the Department of Justice (DOJ) to the House of Representative’s Committee on the Judiciary commenting on the proposed “Child Protection and Sexual Predator Punishment Act of 1998.” H. R. Rep. No. 105–557, pp. 26–34 (1998). In the letter, DOJ provides commentary on the then-present state of §§2252(b)(1) and 2252(b)(2), noting that although there is a “5-year mandatory minimum sentence for individuals charged with receipt or distribution of child pornography and who have prior state convictions for child molestation” pursuant to §2252(b)(1), there is “no enhanced provision for those individuals charged with possession of child pornography who have prior convictions for child abuse” pursuant to §2252(b)(2). Id., at 31. That letter, they say, demonstrates that DOJ understood the language at issue here to impose a sentencing enhancement only for prior state convictions involving children. We doubt that DOJ was trying to describe the full reach of the language in §2252(b)(1), as the dissent suggests. To the contrary, there are several clues that the letter was relaying on just one of the provision’s many salient features. For instance, the letter’s references to “child molestation” and “child abuse” do not encompass a large number of state crimes that are unambiguously covered by “abusive sexual conduct involving a minor or ward”—namely, crimes involving “wards.” Wards can be minors, but they can also be adults. See, e.g., §2243(b) (defining “wards” as persons who are “in official detention” and “under . . . custodial, supervisory, or disciplinary authority”). More-over, we doubt that DOJ intended to express a belief that the potentially broad scope of serious crimes encompassed by “aggravated sexual abuse, sexual abuse, and abusive sexual conduct” reaches no further than state crimes that would traditionally be characterized as “child molestation” or “child abuse.” Thus, Congress’ amendment to the provision did give “DOJ just what it wanted,” post, at 10. But the amendment also did more than that. We therefore think it unnecessary to restrict our interpretation of the provision to the parts of it that DOJ chose to highlight in its letter. Just as importantly, the terse descriptions of the provision in the Senate Report and DOJ letter do nothing to explain why Congress would have wanted to apply the mandatory minimum to individuals convicted in federal court of sex-ual abuse or aggravated sexual abuse involving an adult, but not to individuals convicted in state court of the same. The legislative history, in short, “hardly speaks with [a] clarity of purpose” through which we can discern Congress’ statutory objective. Universal Camera Corp. v. NLRB, 340 U. S. 474, 483 (1951) . The best explanation Lockhart can muster is a basic administrability concern: Congress “knew what conduct it was capturing under federal law and could be confident that all covered federal offenses were proper predicates. But Congress did not have the same familiarity with the varied and mutable sexual-abuse laws of all fifty states.” Brief for Petitioner 27. Perhaps Congress worried that state laws punishing relatively minor offenses like public lewdness or indecent exposure involving an adult would be swept into §2252(b)(2). Id., at 28. But the risk Lockhart identifies is minimal. Whether the terms in §2252(b)(2) are given their “generic” meaning, see Descamps v. United States, 570 U. S. ___ (2013); Taylor v. United States, 495 U. S. 575 (1990) , or are defined in light of their federal counterparts—which we do not decide—they are unlikely to sweep in the bizarre or unexpected state offenses that worry Lockhart. D Finally, Lockhart asks us to apply the rule of lenity. We have used the lenity principle to resolve ambiguity in favor of the defendant only “at the end of the process of construing what Congress has expressed” when the ordinary canons of statutory construction have revealed no satisfactory construction. Callanan v. United States, 364 U. S. 587, 596 (1961) . That is not the case here. To be sure, Lockhart contends that if we applied a different principle of statutory construction—namely, his “series-qualifier principle”—we would arrive at an alternative construction of §2252(b)(2). But the arguable availability of multiple, divergent principles of statutory construction cannot automatically trigger the rule of lenity. Cf. Llewellyn, Remarks on the Theory of Appellate Decision and the Rules or Canons About How Statutes Are To Be Construed, 3 Vand. L. Rev. 395, 401 (1950) (“[T]here are two opposing canons on almost every point”). Here, the rule of the last antecedent is well supported by context and Lockhart’s alternative is not. We will not apply the rule of lenity to override a sensible grammatical principle buttressed by the statute’s text and structure. * * * We conclude that the text and structure of §2252(b)(2) confirm that the provision applies to prior state convictions for “sexual abuse” and “aggravated sexual abuse,” whether or not the convictions involved a minor or ward. We therefore hold that Lockhart’s prior conviction for sexual abuse of an adult is encompassed by §2252(b)(2). The judgment of the Court of Appeals, accordingly, is affirmed. So ordered.Notes 1 For example, §2241(a) of Chapter 109A prohibits forced sexual acts against “another person”—not just a person under a certain age. Section 2241(c) specially criminalizes sexual acts “with another person who has not attained the age of 12 years,” and §2243(b) does the same for sexual acts with wards who are “in official detention” or “under the custodial, supervisory, or disciplinary authority of the person so engaging.” 2 See 18 U. S. C. §2241 (1994 ed.) (“Aggravated sexual abuse”); §2242 (“Sexual abuse”); §2243 (“Sexual abuse of a minor or ward”). 3 The dissent points out that §2252(b)(2) (2012 ed.) did not also borrow from the heading of the fourth section in Chapter 109A (or, we note, from the fifth, sixth, seventh, or eighth sections) in defining its categories of state sexual-abuse predicates. Post, at 14-15 (Kagan, J. dissenting). But the significance of the similarity between the three state predicates in §2252(b)(2) and the wording, structure, and order of the first three sections of Chapter 109A is not diminished by the fact that Congress stopped there (especially when the remaining sections largely set out derivations from, definitions of, and penalties for the first three). See, e.g., §2244 (listing offenses derived from §§2241, 2242, and 2243); §2245 (creating an enhancement for offenses under Chapter 109A resulting in death); §2246 (listing definitions).

578.US.2015_14-419

A federal statute provides that a court may freeze before trial certain assets belonging to a defendant accused of violations of federal health care or banking laws. Those assets include (1) property “obtained as a result of” the crime, (2) property “traceable” to the crime, and (3), as relevant here, other “property of equivalent value.” 18 U. S. C. §1345(a)(2). The Government has charged petitioner Luis with fraudulently obtaining nearly $45 million through crimes related to health care. In order to preserve the $2 million remaining in Luis’ possession for payment of restitution and other criminal penalties, the Government secured a pretrial order prohibiting Luis from dissipating her assets, including assets unrelated to her alleged crimes. Though the District Court recognized that the order might prevent Luis from obtaining counsel of her choice, it held that the Sixth Amendment did not give her the right to use her own untainted funds for that purpose. The Eleventh Circuit affirmed. Held: The judgment is vacated, and the case is remanded. 564 Fed. Appx. 493, vacated and remanded. Justice Breyer, joined by The Chief Justice, Justice Ginsburg, and Justice Sotomayor, concluded that the pretrial restraint of legitimate, untainted assets needed to retain counsel of choice violates the Sixth Amendment. The nature and importance of the constitutional right taken together with the nature of the assets lead to this conclusion. Pp. 3–16. (a) The Sixth Amendment right to counsel grants a defendant “a fair opportunity to secure counsel of his own choice,” Powell v. Alabama, 287 U. S. 45 , that he “can afford to hire,” Caplin & Drysdale, Chartered v. United States, 491 U. S. 617 . This Court has consistently referred to the right to counsel of choice as “fundamental.” Pp. 3–5. (b) While the Government does not deny Luis’ fundamental right to be represented by a qualified attorney whom she chooses and can afford to hire, it would nonetheless undermine the value of that right by taking from Luis the ability to use funds she needs to pay for her chosen attorney. The Government attempts to justify this consequence by pointing out that there are important interests on the other side of the legal equation. It wishes to guarantee that funds will be available later to help pay for statutory penalties and restitution, for example. The Government further argues that two previous cases from this Court, Caplin & Drysdale, supra, at 619, and United States v. Monsanto, 491 U. S. 600 , support the issuance of a restraining order in this case. However, the nature of the assets at issue here differs from the assets at issue in those earlier cases. And that distinction makes a difference. Pp. 5–16. (1) Here, the property is untainted, i.e., it belongs to Luis. As described in Caplin & Drysdale and Monsanto, the Government may well be able to freeze before trial “tainted” assets—e.g., loot, contraband, or property otherwise associated with the planning, implementing, or concealing of a crime. As a matter of property law, the defendant’s ownership interest in such property is imperfect. For example, a different federal statute provides that title to property used to commit a crime (or otherwise “traceable” to a crime) passes to the Government at the instant the crime is planned or committed. See 21 U. S. C. §853(c). But here, the Government seeks to impose restrictions upon Luis’ untainted property without any showing of any equivalent governmental interest in that property. Pp. 5–10. (2) This distinction does not by itself answer the constitutional question because the law of property may allow a person without a present interest in a piece of property to impose restrictions upon a current owner, say, to prevent waste. However, insofar as innocent funds are needed to obtain counsel of choice, the Sixth Amendment prohibits the court order sought here. Three basic considerations lead to this conclusion. First, the nature of the competing interests argues against this kind of court order. On the one side is a fundamental Sixth Amendment right to assistance of counsel. On the other side is the Government’s interest in securing its punishment of choice, as well as the victim’s interest in securing restitution. These latter interests are important, but—compared to the right to counsel—they seem to lie somewhat further from the heart of a fair, effective criminal justice system. Second, relevant, common-law legal tradition offers virtually no significant support for the Government’s position and in fact argues to the contrary. Indeed, there appears to be no decision of this Court authorizing unfettered, pretrial forfeiture of the defendant’s own “innocent” property. Third, as a practical matter, accepting the Government’s position could erode the right to counsel considerably. It would, in fact, unleash a principle of constitutional law with no obvious stopping place, as Congress could write more statutes authorizing restraints in other cases involving illegal behavior that come with steep financial consequences. These defendants, often rendered indigent, would fall back upon publicly paid counsel, including overworked and underpaid public defenders. The upshot is a substantial risk that accepting the Government’s views would render less effective the basic right the Sixth Amendment seeks to protect. Pp. 11–15. (3) The constitutional line between a criminal defendant’s tainted funds and innocent funds needed to pay for counsel should prove workable. Money may be fungible, but courts, which use tracing rules in cases of, e.g., fraud and pension rights, have experience separating tainted assets from untainted assets, just as they have experience determining how much money is needed to cover the costs of a lawyer. Pp. 15–16. Justice Thomas concluded that the rule that a pretrial freeze of untainted assets violates a defendant’s Sixth Amendment right to counsel of choice rests strictly on the Sixth Amendment’s text and common-law backdrop. Pp. 1–12. (a) The Sixth Amendment abolished the common-law rule that generally prohibited representation in felony cases. “The right to select counsel of one’s choice” is thus “the root meaning” of the Sixth Amendment right to counsel. United States v. Gonzalez-Lopez, 548 U. S. 140 –148. Constitutional rights protect the necessary prerequisites for their exercise. As a result, the Sixth Amendment denies the Government unchecked power to freeze a defendant’s assets before trial simply to secure potential forfeiture upon conviction. Unless the right to counsel protects the right to use lawfully owned property to pay for an attorney, the right to counsel—originally understood to protect only the right to hire counsel of choice—would be meaningless. Without pretrial protection for at least some of a defendant’s assets, the Government could nullify the right to counsel of choice, eviscerating the Sixth Amendment’s original meaning and purpose. The modern, judicially created right to government-appointed counsel does not obviate these concerns. Pp. 1–5. (b) History confirms this textual understanding. The common-law forfeiture tradition provides an administrable rule for the Sixth Amendment’s protection: A criminal defendant’s untainted assets are protected from government interference before trial and judgment, but his tainted assets may be seized before trial as contraband or through a separate in rem proceeding. Reading the Sixth Amendment to track the historical line between tainted and untainted assets avoids case-by-case adjudication and ensures that the original meaning of the right to counsel does real work. Here, the incursion of the pretrial asset freeze into untainted assets, for which there is no historical tradition, violates the Sixth Amendment. Pp. 5–9. (c) This conclusion leaves no room for an atextual balancing analysis. Pp. 9–12. Breyer, J., announced the judgment of the Court and delivered an opinion, in which Roberts, C. J., and Ginsburg and Sotomayor, JJ., joined. Thomas, J., filed an opinion concurring in the judgment. Kennedy, J., filed a dissenting opinion, in which Alito, J., joined. Kagan, J., filed a dissenting opinion.

(referring to, e.g., Kaley v. United States, 571 U. S. ___ (2014); Caplin & Drysdale, Chartered v. United States, 491 U. S. 617, 631 (1989) ; United States v. Monsanto, 491 U. S. 600, 616 (1989) ). We granted Luis’ petition for certiorari. II The question presented is “[w]hether the pretrial restraint of a criminal defendant’s legitimate, untainted assets (those not traceable to a criminal offense) needed to retain counsel of choice violates the Fifth and Sixth Amendments.” Pet. for Cert. ii. We see no reasonable way to interpret the relevant statutes to avoid answering this constitutional question. Cf. Monsanto, supra, at 614. Hence, we answer it, and our answer is that the pretrial restraint of legitimate, untainted assets needed to retain counsel of choice violates the Sixth Amendment. The nature and importance of the constitutional right taken together with the nature of the assets lead us to thisconclusion. A No one doubts the fundamental character of a criminal defendant’s Sixth Amendment right to the “Assistance of Counsel.” In Gideon v. Wainwright, 372 U. S. 335 (1963) , the Court explained: “ ‘The right to be heard would be, in many cases, of little avail if it did not comprehend the right to be heard by counsel. Even the intelligent and educated layman has small and sometimes no skill in the science of law. If charged with crime, he is incapable, generally, of determining for himself whether the indictment is good or bad. He is unfamiliar with the rules of evidence. Left without the aid of counsel he may be put on trial without a proper charge, and convicted upon incompetent evidence, or evidence irrelevant to the issue or otherwise inadmissible. He lacks both the skill and knowledge adequately to prepare his defense, even though he have a perfect one. He requires the guiding hand of counsel at every step in the proceedings against him. Without it, though he be not guilty, he faces the danger of conviction because he does not know how to establish his innocence.’ ” Id., at 344–345 (quoting Powell v. Alabama, 287 U. S. 45 –69 (1932)). It is consequently not surprising: first, that this Court’s opinions often refer to the right to counsel as “fundamental,” id., at 68; see Grosjean v. American Press Co., 297 U. S. 233 –244 (1936) (similar); Johnson v. Zerbst, 304 U. S. 458 –463 (1938) (similar); second, that commentators describe the right as a “great engin[e] by which an innocent man can make the truth of his innocence visible,” Amar, Sixth Amendment First Principles, 84 Geo. L. J. 641, 643 (1996); see Herring v. New York, 422 U. S. 853, 862 (1975) ; third, that we have understood the right to require that the Government provide counsel for an indigent defendant accused of all but the least serious crimes, see Gideon, supra, at 344; and fourth, that we have considered the wrongful deprivation of the right to counsel a “structural” error that so “affec[ts] the framework within which the trial proceeds” that courts may not even ask whether the error harmed the defendant. United States v. Gonzalez-Lopez, 548 U. S. 140, 148 (2006) (internal quotation marks omitted); see id., at 150. Given the necessarily close working relationship between lawyer and client, the need for confidence, and the critical importance of trust, neither is it surprising that the Court has held that the Sixth Amendment grants a defendant “a fair opportunity to secure counsel of his own choice.” Powell, supra, at 53; see Gonzalez-Lopez, supra, at 150 (describing “these myriad aspects of representation”). This “fair opportunity” for the defendant to secure counsel of choice has limits. A defendant has no right, for example, to an attorney who is not a member of the bar, or who has a conflict of interest due to a relationship with an opposing party. See Wheat v. United States, 486 U. S. 153, 159 (1988) . And an indigent defendant, while entitled to adequate representation, has no right to have the Government pay for his preferred representational choice. See Caplin & Drysdale, 491 U. S., at 624. We nonetheless emphasize that the constitutional right at issue here is fundamental: “[T]he Sixth Amendment guarantees a defendant the right to be represented by an otherwise qualified attorney whom that defendant can afford to hire.” Ibid. B The Government cannot, and does not, deny Luis’ right to be represented by a qualified attorney whom she chooses and can afford. But the Government would underminethe value of that right by taking from Luis the ability to use the funds she needs to pay for her chosen attorney. The Government points out that, while freezing the funds may have this consequence, there are important interests on the other side of the legal equation: It wishes to guarantee that those funds will be available later to help pay for statutory penalties (including forfeiture of untainted assets) and restitution, should it secure convictions. And it points to two cases from this Court, Caplin & Drysdale, supra, at 619, and Monsanto, 491 U. S., at 615, which, in the Government’s view, hold that the Sixth Amendment does not pose an obstacle to its doing so here. In our view, however, the nature of the assets at issue here differs from the assets at issue in those earlier cases. And that distinction makes a difference. 1 The relevant difference consists of the fact that the property here is untainted; i.e., it belongs to the defendant, pure and simple. In this respect it differs from a robber’s loot, a drug seller’s cocaine, a burglar’s tools, or other property associated with the planning, implementing, or concealing of a crime. The Government may well be able to freeze, perhaps to seize, assets of the latter, “tainted” kind before trial. As a matter of property law the defendant’s ownership interest is imperfect. The robber’s loot belongs to the victim, not to the defendant. See Telegraph Co. v. Davenport, 97 U. S. 369, 372 (1878) (“The great principle that no one can be deprived of his property without his assent, except by the processes of the law, requires . . . that the property wrongfully transferred or stolen should be restored to its rightful owner”). The cocaine is contraband, long considered forfeitable to the Government wherever found. See, e.g., 21 U. S. C. §881(a) (“[Controlled substances] shall be subject to forfeiture to the United States and no property right shall exist in them”); Carroll v. United States, 267 U. S. 132, 159 (1925) (describing the seizure of “contraband forfeitable prop-erty”). And title to property used to commit a crime (orotherwise “traceable” to a crime) often passes to the Government at the instant the crime is planned or committed. See, e.g., §853(c) (providing that the Government’s ownership interest in such property relates back to the time of the crime). The property at issue here, however, is not loot, contraband, or otherwise “tainted.” It belongs to the defendant. That fact undermines the Government’s reliance upon precedent, for both Caplin & Drysdale and Monsanto relied critically upon the fact that the property at issue was “tainted,” and that title to the property therefore had passed from the defendant to the Government before the court issued its order freezing (or otherwise disposing of ) the assets. In Caplin & Drysdale, the Court considered a post-conviction forfeiture that took from a convicted defendant funds he would have used to pay his lawyer. The Court held that the forfeiture was constitutional. In doing so, however, it emphasized that the forfeiture statute at issue provided that “ ‘[a]ll right, title, and interest in property [constituting or derived from any proceeds obtained from the crime] vests in the United States upon the commission of the act giving rise to [the] forfeiture.’ ” 491 U. S., at 625, n. 4 (quoting §853(c)) (emphasis added). It added that the law had “long-recognized” as “lawful” the “practice of vesting title to any forfeitable asset[s] in the United State[s] at the time of the crim[e].” Id., at 627. It pointed out that the defendant did not “claim, as a general proposition, that the [vesting] provision is unconstitutional, or that Congress cannot, as a general matter, vest title to assets derived from the crime in the Government, as of the date of the criminal act in question.” Id., at 627–628. And, given the vesting language, the Court explained that the defendant “did not hold good title” to the property. Id., at 627. The Court therefore concluded that “[t]here is no constitutional principle that gives one person [namely, the defendant] the right to give another’s [namely, the Government’s] property to a third party,” namely, the lawyer. Id., at 628. In Monsanto, the Court considered a pretrial restraining order that prevented a not-yet-convicted defendant from using certain assets to pay for his lawyer. The defendant argued that, given this difference, Caplin & Drysdale’s conclusion should not apply. The Court noted, however, that the property at issue was forfeitable under the same statute that was at issue in Caplin & Drysdale. See Monsanto, supra, at 614. And, as in Caplin & Drysdale, the application of that statute to Monsanto’s case concerned only the pretrial restraint of assets that were traceable to the crime, see 491 U. S., at 602–603; thus, the statute passed title to those funds at the time the crime was committed (i.e., before the trial), see §853(c). The Court said that Caplin & Drysdale had already “weigh[ed] . . . th[e] very interests” at issue. Monsanto, supra, at 616. And it “rel[ied] on” its “conclusion” in Caplin & Drysdale to dispose of, and to reject, the defendant’s “similar constitutional claims.” 491 U. S., at 614. Justice Kennedy prefers to read Caplin & Drysdale and Monsanto broadly, as holding that “the Government, having established probable cause to believe that Luis’ substitute [i.e., innocent] assets will be forfeitable upon conviction, should be permitted to obtain a restraining order barring her from spending those funds prior to trial.” Post, at 6–7 (dissenting opinion). In other words, he believes that those cases stand for the proposition that property—whether tainted or untainted—is subject to pretrial restraint, so long as the property might someday be subject to forfeiture. But this reading asks too much of our precedents. For one thing, as discussed, Caplin & Drysdale and Monsanto involved the restraint only of tainted assets, and thus we had no occasion to opine in those cases about the constitutionality of pretrial restraints of other, untainted assets. For another thing, Justice Kennedy’s broad rule ignores the statutory background against which Caplin & Drysdale and Monsanto were decided. The Court in those cases referenced §853(c) more than a dozen times. And it acknowledged that whether property is “forfeitable” or subject to pretrial restraint under Congress’ scheme is a nuanced inquiry that very much depends on who has the superior interest in the property at issue. See Caplin & Drysdale, supra, at 626–628; Monsanto, 491 U. S., at 616. We see this in, for example, §853(e)(1), which explicitly authorizes restraining orders or injunctions against “property described in subsection (a) of this section” (i.e., tainted assets). We see this too in §853(e)(1)(B), which requires the Government—in certain circumstances—to give “notice to persons appearing to have an interest in the property and opportunity for hearing” before obtaining a restraining order against such property. We see this in §853(c), which allows “bona fide purchaser[s] for value” to keep property that would otherwise be subject to forfeiture. And we see this in §853(n)(6)(A), which exempts certain property from forfeiture when a third party can show a vested interest in the property that is “superior” to that of the Government. The distinction that we have discussed is thus an important one, not a technicality. It is the difference between what is yours and what is mine. In Caplin & Drysdale and Monsanto, the Government wanted to impose restrictions upon (or seize) property that the Government had probable cause to believe was the proceeds of, or traceable to, a crime. See Monsanto, supra, at 615. The relevant statute said that the Government took title to those tainted assets as of the time of the crime. See §853(c). And the defendants in those cases consequently had to concede that the disputed property was in an important sense the Government’s at the time the court imposed the restrictions. See Caplin & Drysdale, supra, at 619–620; Monsanto, supra, at 602–603. This is not to say that the Government “owned” the tainted property outright (in the sense that it could take possession of the property even before obtaining a conviction). See post, at 7–10 (Kennedy, J., dissenting). Rather, it is to say that the Government even before trial had a “substantial” interest in the tainted property sufficient to justify the property’s pretrial restraint. See Caplin & Drysdale, supra, at 627 (“[T]he property rights given the Government by virtue of [§853(c)’s relation-back provision] are more substantial than petitioner acknowledges”); United States v. Stowell, 133 U. S. 1, 19 (1890) (“As soon as [the possessor of the forfeitable asset committed the violation] . . . , the forfeiture . . . took effect, and (though needing judicial condemnation to perfect it) operated from that time as a statutory conveyance to the United States of all right, title and interest then remaining in the [possessor]; and was as valid and effectual, against all the world, as a recorded deed” (emphasis added)). If we analogize to bankruptcy law, the Government, by application of §853(c)’s relation-back provision, became something like a secured creditor with a lien on the defendant’s tainted assets superior to that of most any other party. See 4 Collier on Bankruptcy ¶506.03[1] (16th ed. 2015). For this reason, §853(c) has operated in our cases as a significant limitation on criminal defendants’ prop-erty rights in such assets—even before conviction. SeeMonsanto, supra, at 613 (“Permitting a defendant to use [tainted] assets for his private purposes that, under this [relation-back] provision, will become the property of the United States if a conviction occurs cannot be sanctioned”); cf. Grupo Mexicano de Desarrollo, S. A. v. Alliance Bond Fund, Inc., 527 U. S. 308, 326 (1999) (noting that the Court had previously authorized injunctions against the further dissipation of property where, among other things, “the creditor (the Government) asserted an equitable lien on the property”). Here, by contrast, the Government seeks to impose restrictions upon Luis’ untainted property without any showing of any equivalent governmental interest in that property. Again, if this were a bankruptcy case, the Government would be at most an unsecured creditor. Al-though such creditors someday might collect from a debtor’s general assets, they cannot be said to have any present claim to, or interest in, the debtor’s property. See id., at 330 (“[B]efore judgment . . . an unsecured creditor has no rights at law or in equity in the property of his debtor”); see also 5 Collier on Bankruptcy ¶541.05[1][b] (“[G]eneral unsecured creditor[s]” have “no specific property interest in the goods held or sold by the debtor”). The competing property interests in the tainted- and untainted-asset contexts therefore are not “exactly the same.” Post, at 2 (Kagan, J., dissenting). At least regarding her untainted assets, Luis can at this point reasonably claim that the property is still “mine,” free and clear. 2 This distinction between (1) what is primarily “mine” (the defendant’s) and (2) what is primarily “yours” (the Government’s) does not by itself answer the constitutional question posed, for the law of property sometimes allows a person without a present interest in a piece of property to impose restrictions upon a current owner, say, to prevent waste. A holder of a reversionary interest, for example, can prevent the owner of a life estate from wasting the property. See, e.g., Peterson v. Ferrell, 127 N. C. 169, 170, 37 S. E. 189, 190 (1900). Those who later may become beneficiaries of a trust are sometimes able to prevent the trustee from dissipating the trust’s assets. See, e.g., Kollock v. Webb, 113 Ga. 762, 769, 39 S. E. 339, 343 (1901). And holders of a contingent, future executory interest in property (an interest that might become possessory at some point down the road) can, in limited circumstances, enjoin the activities of the current owner. See, e.g., Dees v. Cheuvronts, 240 Ill. 486, 491, 88 N. E. 1011, 1012 (1909) (“[E]quity w[ill] interfere . . . only when it is made to appear that the contingency . . . is reasonably certain to happen, and the waste is . . . wanton and conscienceless”). The Government here seeks a somewhat analogous order, i.e., an order that will preserve Luis’ untainted assets so that they will be available to cover the costs of forfeiture and restitution if she is convicted, and if the court later determines that her tainted assets are insufficient or otherwise unavailable. The Government finds statutory authority for its request in language authorizing a court to enjoin a criminal defendant from, for example, disposing of innocent “property of equivalent value” to that of tainted property. 18 U. S. C. §1345(a)(2)(B)(i). But Luis needs some portion of those same funds to pay for the lawyer of her choice. Thus, the legal conflict arises. And, in our view, insofar as innocent (i.e., untainted) funds are needed to obtain counsel of choice, we believe that the Sixth Amendment prohibits the court order that the Government seeks. Three basic considerations lead us to this conclusion. First, the nature of the competing interests argues against this kind of court order. On the one side we find, as we have previously explained, supra, at 3–5, a Sixth Amendment right to assistance of counsel that is a fundamental constituent of due process of law, see Powell, 287 U. S.,at 68–69. And that right includes “the right to be represented by an otherwise qualified attorney whom that defendant can afford to hire.” Caplin & Drysdale, 491 U. S., at 624. The order at issue in this case would seriously undermine that constitutional right. On the other side we find interests that include the Government’s contingent interest in securing its punishment of choice (namely, criminal forfeiture) as well as the victims’ interest in securing restitution (notably, from funds belonging to the defendant, not the victims). While these interests are important, to deny the Government the order it requests will not inevitably undermine them, for, at least sometimes, the defendant may possess other assets—say, “tainted” property—that might be used for forfeitures and restitution. Cf. Gonzalez-Lopez, 548 U. S., at 148 (“Deprivation of the right” to counsel of the defendant’s choice “is ‘complete’ when the defendant is erroneously prevented from being represented by the lawyer he wants”). Nor do the interests in obtaining payment of a criminal forfeiture or restitution order enjoy constitutional protection. Rather, despite their importance, compared to the right to counsel of choice, these interests would seem to lie somewhat further from the heart of a fair, effective criminal justice system. Second, relevant legal tradition offers virtually no significant support for the Government’s position. Rather, tradition argues to the contrary. Describing the 18th-century English legal world (which recognized only a limited right to counsel), Blackstone wrote that “only” those “goods and chattels” that “a man has at the time of conviction shall be forfeited.” 4 W. Blackstone, Commentaries on the Laws of England 388 (1765) (emphasisadded); see 1 J. Chitty, Practical Treatise on the CriminalLaw 737 (1816) (“[T]he party indicted may sell any of [his property] . . . to assist him in preparing for his defense on the trial”). Describing the common law as understood in 19th-century America (which recognized a broader right to counsel), Justice Story wrote: “It is well known, that at the common law, in many cases of felonies, the party forfeited his goods and chattels to the crown. The forfeiture . . . was a part, or at least a consequence, of the judgment of conviction. It is plain from this statement, that no right to the goods and chattels of the felon could be acquired by the crown by the mere commission of the offense; but the right attached only by the conviction of the offender. . . . In the contemplation of the common law, the offender’s right was not divested until the conviction.” The Palmyra, 12 Wheat. 1, 14 (1827). See generally Powell, supra, at 60–61 (describing the scope of the right to counsel in 18th-century Britain and colonial America). As we have explained, supra, at 6–10, cases such as Caplin & Drysdale and Monsanto permit the Government to freeze a defendant’s assets pretrial, but the opinions in those cases highlight the fact that the property at issue was “tainted,” i.e., it did not belong entirely to the defendant. We have found no decision of this Court authorizing unfettered, pretrial forfeiture of the defendant’s own “innocent” property—property with no connection to the charged crime. Nor do we see any grounds for distinguishing the historic preference against preconviction forfeitures from the preconviction restraint at issue here. As far as Luis’ Sixth Amendment right to counsel of choice is concerned, a restraining order might as well be a forfeiture; that is, the restraint itself suffices to completely deny this constitutional right. See Gonzalez-Lopez, supra, at 148. Third, as a practical matter, to accept the Government’s position could well erode the right to counsel to a considerably greater extent than we have so far indicated. To permit the Government to freeze Luis’ untainted assets would unleash a principle of constitutional law that would have no obvious stopping place. The statutory provision before us authorizing the present restraining order refers only to “banking law violation[s]” and “Federal health care offense[s].” 18 U. S. C. §1345(a)(2). But, in the Government’s view, Congress could write more statutes authorizing pretrial restraints in cases involving other illegal behavior—after all, a broad range of such behavior can lead to postconviction forfeiture of untainted assets. See, e.g., §1963(m) (providing for forfeiture of innocent, substitute assets for any violation of the Racketeer Influenced and Corrupt Organizations Act). Moreover, the financial consequences of a criminal conviction are steep. Even beyond the forfeiture itself, criminal fines can be high, and restitution orders expensive. See, e.g., §1344 ($1 million fine for bank fraud); §3571 (mail and wire fraud fines of up to $250,000 for individuals and $500,000 for organizations); United States v. Gushlak, 728 F. 3d 184, 187, 203 (CA2 2013) ($17.5 million restitution award against an individual defendant in a fraud-on-the-market case); FTC v. Trudeau, 662 F. 3d 947, 949 (CA7 2011) ($37.6 million remedial sanction for fraud). How are defendants whose innocent assets are frozen in cases like these supposed to pay for a lawyer—particularly if they lack “tainted assets” because they are innocent, a class of defendants whom the right to counsel certainly seeks to protect? See Powell, 287 U. S., at 69; Amar, 84 Geo. L. J., at 643 (“[T]he Sixth Amendment is generally designed to elicit truth and protect innocence”). These defendants, rendered indigent, would fall back upon publicly paid counsel, including overworked and underpaid public defenders. As the Department of Justice explains, only 27 percent of county-based public defender offices have sufficient attorneys to meet nationally recommended caseload standards. Dept. of Justice, Bureau of Justice Statistics, D. Farole & L. Langton, Census of Public Defender Offices, 2007: County-based and Local Public Defender Offices, 2007, p. 10 (Sept. 2010). And as one amicus points out, “[m]any federal public defender organizations and lawyers appointed under the Criminal Justice Act serve numerous clients and have only limited resources.” Brief for New York Council of Defense Lawyers 11. The upshot is a substantial risk that accept-ing the Government’s views would—by increasing the government-paid-defender workload—render less effective the basic right the Sixth Amendment seeks to protect. 3 We add that the constitutional line we have drawn should prove workable. That line distinguishes between a criminal defendant’s (1) tainted funds and (2) innocent funds needed to pay for counsel. We concede, as Justice Kennedy points out, post, at 12–13, that money is fungible; and sometimes it will be difficult to say whether a particular bank account contains tainted or untainted funds. But the law has tracing rules that help courts implement the kind of distinction we require in this case. With the help of those rules, the victim of a robbery, for example, will likely obtain the car that the robber used stolen money to buy. See, e.g., 1 G. Palmer, Law of Restitution §2.14, p. 175 (1978) (“tracing” permits a claim against “an asset which is traceable to or the product of” tainted funds); 4 A. Scott, Law of Trusts §518, pp. 3309–3314 (1956) (describing the tracing rules governing commingled accounts). And those rules will likely also prevent Luis from benefiting from many of the money transfers and purchases Justice Kennedy describes. See post, at 12–13. Courts use tracing rules in cases involving fraud, pension rights, bankruptcy, trusts, etc. See, e.g., Montanile v. Board of Trustees of Nat. Elevator Industry Health Benefit Plan, 577 U. S. ___, ___–___ (2016) (slip op., at 8–9). They consequently have experience separating tainted assets from untainted assets, just as they have experience determining how much money is needed to cover the costs of a lawyer. See, e.g., 18 U. S. C. §1345(b) (“The court shall proceed as soon as practicable to the hearing and determination of [actions to freeze a defendant’s tainted or untainted assets]”); 28 U. S. C. §2412(d) (courts must determine reasonable attorneys’ fees under the Equal Access to Justice Act); see also Kaley, 571 U. S., at ___, and n. 3 (slip op., at 3, and n. 3) (“Since Monsanto, the lower courts have generally provided a hearing. . . . [to determine] whether probable cause exists to believe that the assets in dispute are traceable . . . to the crime charged in the indictment”). We therefore see little reason to worry, as Justice Kennedy seems to, that defendants will “be allowed to circumvent [the usual forfeiture rules] by using . . . funds to pay for a high, or even the highest, priced defense team [they] can find.” Post, at 7. * * * For the reasons stated, we conclude that the defendant in this case has a Sixth Amendment right to use her own “innocent” property to pay a reasonable fee for the assistance of counsel. On the assumptions made here, the District Court’s order prevents Luis from exercising that right. We consequently vacate the judgment of the Court of Appeals and remand the case for further proceedings. It is so ordered. APPENDIX Title 18 U. S. C. §1345 provides: “(a)(1) If a person is— “(A) violating or about to violate this chapter or section 287, 371 (insofar as such violation involves a conspiracy to defraud the United States or any agency thereof), or 1001 of this title; “(B) committing or about to commit a banking law violation (as defined in section 3322(d) of this title); or “(C) committing or about to commit a Federal health care offense; “the Attorney General may commence a civil action in any Federal court to enjoin such violation. “(2) If a person is alienating or disposing of property, or intends to alienate or dispose of property, obtained as a result of a banking law violation (as defined in section 3322(d) of this title) or a Federal health care offense or property which is traceable to such violation, the Attorney General may commence a civil action in any Federal court— “(A) to enjoin such alienation or disposition of property; or “(B) for a restraining order to— “(i) prohibit any person from withdrawing, transferring, removing, dissipating, or disposing of any such property or property of equivalent value; and “(ii) appoint a temporary receiver to administer such restraining order. “(3) A permanent or temporary injunction or restraining order shall be granted without bond. “(b) The court shall proceed as soon as practicable to the hearing and determination of such an action, and may, at any time before final determination, enter such a restraining order or prohibition, or take such other action, as is warranted to prevent a continuing and substantial injury to the United States or to any person or class of persons for whose protection the action is brought. A proceeding under this section is governed by the Federal Rules of Civil Procedure, except that, if an indictment has been returned against the respondent, discovery is governed by the Federal Rules of Criminal Procedure.”

578.US.2015_14-1096

Any alien convicted of an “aggravated felony” after entering the United States is deportable, ineligible for several forms of discretionary relief, and subject to expedited removal. 8 U. S. C. §§1227(a)(2)(A)(iii), (3). An “aggravated felony” is defined as any of numerous offenses listed in §1101(a)(43), each of which is typically identified either as an offense “described in” a specific federal statute or by a generic label (e.g., “murder”). Section 1101(a)(43)’s penultimate sentence states that each enumerated crime is an aggravated felony irrespective of whether it violates federal, state, or foreign law. Petitioner Jorge Luna Torres (Luna), a lawful permanent resident, pleaded guilty in a New York court to attempted third-degree arson. When immigration officials discovered his conviction, they initiated removal proceedings. The Immigration Judge determined that Luna’s arson conviction was for an “aggravated felony” and held that Luna was therefore ineligible for discretionary relief. The Board of Immigration Appeals affirmed. It found the federal and New York arson offenses to be identical except for the former’s requirement that the crime have a connection to interstate or foreign commerce. Because the federal statute’s commerce element serves only a jurisdictional function, the Board held, New York’s arson offense is “described in” the federal statute, 18 U. S. C. §844(i), for purposes of determining whether an alien has been convicted of an aggravated felony. The Second Circuit denied review. Held: A state offense counts as a §1101(a)(43) “aggravated felony” when it has every element of a listed federal crime except one requiring a connection to interstate or foreign commerce. Because Congress lacks general constitutional authority to punish crimes, most federal offenses include a jurisdictional element to tie the substantive crime to one of Congress’s enumerated powers. State legislatures are not similarly constrained, and so state crimes do not need such a jurisdictional hook. That discrepancy creates the issue here—whether a state offense lacking a jurisdictional element but otherwise mirroring a particular federal offense can be said to be “described” by that offense. Dictionary definitions of the word “described” do not clearly resolve this question one way or the other. Rather, two contextual considerations decide this case: §1101(a)(43)’s penultimate sentence and a well-established background principle that distinguishes between substantive and jurisdictional elements in criminal statutes. Pp. 4–21. (a) Section §1101(a)(43)’s penultimate sentence shows that Congress meant the term “aggravated felony” to capture serious crimes regardless of whether they are made illegal by the Federal Government, a State, or a foreign country. But Luna’s view would substantially undercut that function by excluding from the Act’s coverage all state and foreign versions of any enumerated federal offense containing an interstate commerce element. And it would do so in a particularly perverse fashion—excluding state and foreign convictions for many of §1101(a)(43)’s gravest crimes (e.g., most child pornography offenses), while reaching convictions for far less harmful offenses (e.g., operating an unlawful gambling business). Luna theorizes that such haphazard coverage might reflect Congress’s belief that crimes with an interstate connection are generally more serious than those without. But it is implausible that Congress viewed the presence of an interstate commerce element as separating serious from non-serious conduct. Luna’s theory misconceives the function of interstate commerce elements and runs counter to the penultimate sentence’s central message—that the state, federal, or foreign nature of a crime is irrelevant. And his claim that many serious crimes excluded for want of an interstate commerce element would nonetheless count as §1101(a)(43)(F) “crime[s] of violence” provides little comfort: That alternative would not include nearly all such offenses, nor even the worst ones. Pp. 7–14. (b) The settled practice of distinguishing between substantive and jurisdictional elements in federal criminal statutes also supports reading §1101(a)(43) to include state analogues that lack only an interstate commerce requirement. Congress uses substantive and jurisdictional elements for different reasons and does not expect them to receive identical treatment. See, e.g., United States v. Yermian, 468 U. S. 63 . And that is true where, as here, the judicial task is to compare federal and state offenses. See Lewis v. United States, 523 U. S. 155 . Pp. 14–19. 764 F. 3d 152, affirmed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, and Alito, JJ., joined. Sotomayor, J., filed a dissenting opinion, in which Thomas and Breyer, JJ., joined.

The Immigration and Nationality Act (INA or Act) imposes certain adverse immigration consequences on an alien convicted of an “aggravated felony.” The INA defines that term by listing various crimes, most of which are identified as offenses “described in” specified provisions of the federal criminal code. Immediately following that list, the Act provides that the referenced offenses are aggravated felonies irrespective of whether they are “in violation of Federal[,] State[,]” or foreign law. 108 Stat. 4322, 8 U. S. C. §1101(a)(43). In this case, we must decide if a state crime counts as an aggravated felony when it corresponds to a specified federal offense in all ways but one—namely, the state crime lacks the interstate commerce element used in the federal statute to establish legislative jurisdiction (i.e., Congress’s power to enact the law). We hold that the absence of such a jurisdictional element is immaterial: A state crime of that kind is an aggravated felony. I The INA makes any alien convicted of an “aggravated felony” after entering the United States deportable. See §1227(a)(2)(A)(iii). Such an alien is also ineligible for several forms of discretionary relief, including cancellation of removal—an order allowing a deportable alien to remain in the country. See §1229b(a)(3). And because of his felony, the alien faces expedited removal proceedings. See §1228(a)(3)(A). The Act defines the term “aggravated felony” by way of a long list of offenses, now codified at §1101(a)(43). In all, that provision’s 21 subparagraphs enumerate some 80 different crimes. In more than half of those subparagraphs, Congress specified the crimes by citing particular federal statutes. According to that common formulation, an offense is an aggravated felony if it is “described in,” say, 18 U. S. C. §2251 (relating to child pornography), §922(g) (relating to unlawful gun possession), or, of particular relevance here, §844(i) (relating to arson and explosives). 8 U. S. C. §§1101(a)(43)(E), (I). Most of the remaining subparagraphs refer to crimes by their generic labels, stating that an offense is an aggravated felony if, for example, it is “murder, rape, or sexual abuse of a minor.” §1101(a)(43)(A). Following the entire list of crimes, §1101(a)(43)’s penultimate sentence reads: “The term [aggravated felony] applies to an offense described in this paragraph whether in violation of Federal or State law and applies to such an offense in violation of the law of a foreign country for which the term of imprisonment was completed within the previous 15 years.” So, putting aside the 15-year curlicue, the penultimate sentence provides that an offense listed in §1101(a)(43) is an aggravated felony whether in violation of federal, state, or foreign law. Petitioner Jorge Luna Torres, who goes by the name George Luna, immigrated to the United States as a child and has lived here ever since as a lawful permanent resident. In 1999, he pleaded guilty to attempted arson in the third degree, in violation of New York law; he was sentenced to one day in prison and five years of probation. Seven years later, immigration officials discovered his conviction and initiated proceedings to remove him from the country. During those proceedings, Luna applied for cancellation of removal. But the Immigration Judge found him ineligible for that discretionary relief because his arson conviction qualified as an aggravated felony. See App. to Pet. for Cert. 21a–22a. The Board of Immigration Appeals (Board) affirmed, based on a comparison of the federal and New York arson statutes. See id., at 15a–17a. The INA, as just noted, provides that “an offense described in” 18 U. S. C. §844(i), the federal arson and explosives statute, is an aggravated felony. Section 844(i), in turn, makes it a crime to “maliciously damage[ ] or destroy[ ], or attempt[ ] to damage or destroy, by means of fire or an explosive, any building [or] vehicle . . . used in interstate or foreign commerce or in any activity affecting interstate or foreign commerce.” For its part, the New York law that Luna was convicted under prohibits “intentionally damag[ing],” or attempting to damage, “a building or motor vehicle by starting a fire or causing an explosion.” N. Y. Penal Law Ann. §§110, 150.10 (West 2010). The state law, the Board explained, thus matches the federal statute element-for-element with one exception: The New York law does not require a connection to interstate commerce. According to the Board, that single difference did not matter because the federal statute’s commerce element is “jurisdictional”—that is, its function is to establish Congress’s power to legislate. See App. to Pet for Cert. 16a–17a. Given that the two laws’ substantive (i.e., non-jurisdictional) elements map onto each other, the Board held, the New York arson offense is “described in” 18 U. S. C. §844(i). The Court of Appeals for the Second Circuit denied Luna’s petition for review of the Board’s ruling. See 764 F. 3d 152 (2014). The court’s decision added to a Circuit split over whether a state offense is an aggravated felony when it has all the elements of a listed federal crime except one requiring a connection to interstate commerce.[1] We granted certiorari. 576 U. S. ___ (2015). II The issue in this case arises because of the distinctive role interstate commerce elements play in federal criminal law. In our federal system, “Congress cannot punish felonies generally,” Cohens v. Virginia, 6 Wheat. 264, 428 (1821); it may enact only those criminal laws that are connected to one of its constitutionally enumerated powers, such as the authority to regulate interstate commerce. As a result, most federal offenses include, in addition to substantive elements, a jurisdictional one, like the interstate commerce requirement of §844(i). The substantive elements “primarily define[ ] the behavior that the statute calls a ‘violation’ of federal law,” Scheidler v. National Organization for Women, Inc., 547 U. S. 9, 18 (2006) —or, as the Model Penal Code puts the point, they relate to “the harm or evil” the law seeks to prevent, §1.13(10). The jurisdictional element, by contrast, ties the substantive offense (here, arson) to one of Congress’s constitutional powers (here, its authority over interstate commerce), thus spelling out the warrant for Congress to legislate. See id., at 17–18 (explaining that Congress intends “such statu-tory terms as ‘affect commerce’ or ‘in commerce’ . . . as terms of art connecting the congressional exercise of legislative authority with the constitutional provision (here, the Commerce Clause) that grants Congress that authority”). For obvious reasons, state criminal laws do not include the jurisdictional elements common in federal statutes.[2] State legislatures, exercising their plenary police powers, are not limited to Congress’s enumerated powers; and so States have no reason to tie their substantive offenses to those grants of authority. See, e.g., United States v. Lopez, 514 U. S. 549, 567 (1995) . In particular, state crimes do not contain interstate commerce elements because a State does not need such a jurisdictional hook. Accordingly, even state offenses whose substantive elements match up exactly with a federal law’s will part ways with respect to interstate commerce. That slight discrepancy creates the issue here: If a state offense lacks an interstate commerce element but otherwise mirrors one of the federal statutes listed in §1101(a)(43), does the state crime count as an aggravated felony? Or, alternatively, does the jurisdictional difference reflected in the state and federal laws preclude that result, no matter the laws’ substantive correspondence? Both parties begin with the statutory text most directly at issue, disputing when a state offense (here, arson) is “described in” an enumerated federal statute (here, 18 U. S. C. §844(i)). Luna, armed principally with Black’s Law Dictionary, argues that “described in” means “expressed” or “set forth” in—which, he says, requires the state offense to include each one of the federal law’s elements. Brief for Petitioner 15–16.[3] The Government, brandishing dictionaries of its own, contends that the statutory phrase has a looser meaning—that “describing entails . . . not precise replication,” but “convey[ance of ] an idea or impression” or of a thing’s “central features.” Brief for Respondent 17.[4] On that view, “described in,” as opposed to the more precise “defined in” sometimes found in statutes, denotes that the state offense need only incorporate the federal law’s core, substantive elements. But neither of those claims about the bare term “described in” can resolve this case. Like many words, “describe” takes on different meanings in different contexts. Consider two ways in which this Court has used the word. In one case, “describe” conveyed exactness: A contractual provision, we wrote, “describes the subject [matter] with great particularity[,] . . . giv[ing] the precise number of pounds [of tobacco], the tax for which each pound was liable, and the aggregate of the tax.” Ryan v. United States, 19 Wall. 514, 517 (1874). In another case, not: “The disclosure provision is meant,” we stated, “to describe the law to consumers in a manner that is concise and comprehensible to the layman—which necessarily means that it will be imprecise.” CompuCredit Corp. v. Greenwood, 565 U. S. 95, 102 (2012) . So staring at, or even looking up, the words “described in” cannot answer whether a state offense must replicate every last element of a listed federal statute, including its jurisdictional one, to qualify as an aggravated felony. In considering that issue, we must, as usual, “interpret the relevant words not in a vacuum, but with reference to the statutory context.” Abramski v. United States, 573 U. S. ___, ___ (2014) (slip op., at 9).[5] Here, two contextual considerations decide the matter. The first is §1101(a)(43)’s penultimate sentence, which shows that Congress meant the term “aggravated felony” to capture serious crimes regardless of whether they are prohibited by federal, state, or foreign law. The second is a well-established background principle distinguishing between substantive and jurisdictional elements in federal criminal statutes. We address each factor in turn. A Section 1101(a)(43)’s penultimate sentence, as noted above, provides: “The term [aggravated felony] applies to an offense described in this paragraph whether in violation of Federal or State law and applies to such an offense in violation of the law of a foreign country for which the term of imprisonment was completed within the previous 15 years.” See supra, at 2. That sentence (except for the time limit on foreign convictions) declares the source of criminal law irrelevant: The listed offenses count as aggravated felonies regardless of whether they are made illegal by the Federal Government, a State, or a foreign country. That is true of the crimes identified by reference to federal statutes (as here, an offense described in 18 U. S. C. §844(i)), as well as those employing generic labels (for example, murder). As even Luna recognizes, state and foreign analogues of the enumerated federal crimes qual-ify as aggravated felonies. See Brief for Petitioner 21 (contesting only what properly counts as such an analogue). The whole point of §1101(a)(43)’s penultimate sentence is to make clear that a listed offense should lead to swift removal, no matter whether it violates federal, state, or foreign law. Luna’s jot-for-jot view of “described in” would substantially undercut that function by excluding from the Act’s coverage all state and foreign versions of any enumerated federal offense that (like §844(i)) contains an interstate commerce element. Such an element appears in about half of §1101(a)(43)’s listed statutes—defining, altogether, 27 serious crimes.[6] Yet under Luna’s reading, only those federal crimes, and not their state and foreign counterparts, would provide a basis for an alien’s removal—because, as explained earlier, only Congress must ever show a link to interstate commerce. See supra, at 4–5. No state or foreign legislature needs to incorporate a commerce element to establish its jurisdiction, and so none ever does. Accordingly, state and foreign crimes will never precisely replicate a federal statute containing a commerce element. And that means, contrary to §1101(a)(43)’s penultimate sentence, that the term “aggravated felony” would not apply to many of the Act’s listed offenses irrespective of whether they are “in violation of Federal[,] State[, or foreign] law”; instead, that term would apply exclusively to the federal variants.[7] Indeed, Luna’s view would limit the penultimate sentence’s effect in a peculiarly perverse fashion—excluding state and foreign convictions for many of the gravest crimes listed in §1101(a)(43), while reaching those convictions for less harmful offenses. Consider some of the state and foreign crimes that would not count as aggravated felonies on Luna’s reading because the corresponding federal law has a commerce element: most child pornography offenses, including selling a child for the purpose of manufacturing such material, see §1101(a)(43)(I); demanding or receiving a ransom for kidnapping, see §1101(a)(43)(H); and possessing a firearm after a felony conviction, see §1101(a)(43)(E)(ii). Conversely, the term “aggravated felony” in Luna’s world would include state and foreign convictions for such comparatively minor offenses as operating an unlawful gambling business, see §1101(a)(43)(J), and possessing a firearm not identified by a serial number, see §1101(a)(43)(E)(iii), because Congress chose, for whatever reason, not to use a commerce element when barring that conduct. And similarly, the term would cover any state or foreign conviction for such nonviolent activity as receiving stolen property, see §1101(a)(43)(G), or forging documents, see §1101(a)(43)(R), because the INA happens to use generic labels to describe those crimes. This Court has previously refused to construe §1101(a)(43) so as to produce such “haphazard”—indeed, upside-down—coverage. Nijhawan v. Holder, 557 U. S. 29, 40 (2009) . We see no reason to follow a different path here: Congress would not have placed an alien convicted by a State of running an illegal casino at greater risk of removal than one found guilty under the same State’s law of selling a child.[8] In an attempt to make some sense of his reading, Luna posits that Congress might have believed that crimes having an interstate connection are generally more serious than those lacking one—for example, that interstate child pornography is “worse” than the intrastate variety. Brief for Petitioner 35. But to begin with, that theory cannot explain the set of crazy-quilt results just described: Not even Luna maintains that Congress thought local acts of selling a child, receiving explosives, or demanding a ransom are categorically less serious than, say, operating an unlawful casino or receiving stolen property (whether or not in interstate commerce). And it is scarcely more plausible to view an interstate commerce element in any given offense as separating serious from non-serious conduct: Why, for example, would Congress see an alien who carried out a kidnapping for ransom wholly within a State as materially less dangerous than one who crossed state lines in committing that crime? The essential harm of the crime is the same irrespective of state borders. Luna’s argument thus misconceives the function of interstate commerce elements: Rather than distinguishing greater from lesser evils, they serve (as earlier explained) to connect a given substantive offense to one of Congress’s enumerated powers. See supra, at 4–5. And still more fundamentally, Luna’s account runs counter to the penultimate sentence’s central message: that the national, local, or foreign character of a crime has no bearing on whether it is grave enough to warrant an alien’s automatic removal.[9] Luna (and the dissent, see post, at 6) must therefore fall back on a different defense: that his approach would exclude from the universe of aggravated felonies fewer serious state and foreign offenses than one might think. To make that argument, Luna relies primarily on a part of the Act specifying that the term “aggravated felony” shall include “a crime of violence (as defined in [ 18 U. S. C. §16]) for which the term of imprisonment [is] at least one year.” §1101(a)(43)(F); see 18 U. S. C. §16 (defining “crime of violence” as involving the use of “physical force” against the person or property of another). According to Luna, many state and foreign offenses failing to match the Act’s listed federal statutes (for want of an interstate commerce element) would count as crimes of violence and, by that alternative route, trigger automatic removal. A different statutory phrase, or so Luna says, would thus plug the holes opened by his construction of the “described in” provisions. Luna’s argument does not reassure us. We agree that state counterparts of some enumerated federal offenses would qualify as aggravated felonies through the “crime of violence” provision. But not nearly all such offenses, and not even the worst ones. Consider again some of the listed offenses described earlier. See supra, at 10. The “crime of violence” provision would not pick up demanding a ransom for kidnapping. See 18 U. S. C. §875(a) (defining the crime without any reference to physical force). It would not cover most of the listed child pornography offenses, involving the distribution, receipt, and possession of such materials. It would not reach felon-in-possession laws and other firearms offenses. And indeed, it would not reach arson in the many States defining that crime to include the destruction of one’s own property. See Jordison v. Gonzales, 501 F. 3d 1134, 1135 (CA9 2007) (holding that a violation of California’s arson statute does not count as a crime of violence for that reason); Tr. of Oral Arg. 28–29 (Solicitor General agreeing with that interpretation).[10] So under Luna’s reading, state and foreign counterparts to a broad swath of listed statutes would remain outside §1101(a)(43)’s coverage merely because they lack an explicit interstate commerce connection. And for all the reasons discussed above, that result would significantly restrict the penultimate sentence’s force and effect, and in an utterly random manner.[11] B Just as important, a settled practice of distinguishing between substantive and jurisdictional elements of federal criminal laws supports reading §1101(a)(43) to include state analogues lacking an interstate commerce requirement. As already explained, the substantive elements of a federal statute describe the evil Congress seeks to prevent; the jurisdictional element connects the law to one of Congress’s enumerated powers, thus establishing legislative authority. See supra, at 4–5; ALI, Model Penal Code §1.13(10) (1962). Both kinds of elements must be proved to a jury beyond a reasonable doubt; and because that is so, both may play a real role in a criminal case. But still, they are not created equal for every purpose. To the contrary, courts have often recognized—including when comparing federal and state offenses—that Congress uses substantive and jurisdictional elements for different reasons and does not expect them to receive identical treatment. Consider the law respecting mens rea. In general, courts interpret criminal statutes to require that a defendant possess a mens rea, or guilty mind, as to every element of an offense. See Elonis v. United States, 575 U. S. ___, ___ (2015) (slip op., at 10). That is so even when the “statute by its terms does not contain” any demand of that kind. United States v. X-Citement Video, Inc., 513 U. S. 64, 70 (1994) . In such cases, courts read the statute against a “background rule” that the defendant must know each fact making his conduct illegal. Staples v. United States, 511 U. S. 600, 619 (1994) . Or otherwise said, they infer, absent an express indication to the contrary, that Congress intended such a mental-state requirement. Except when it comes to jurisdictional elements. There, this Court has stated, “the existence of the fact that confers federal jurisdiction need not be one in the mind of the actor at the time he perpetrates the act made criminal by the federal statute.” United States v. Feola, 420 U. S. 671, 677, n. 9 (1975) ; see United States v. Yermian, 468 U. S. 63, 68 (1984) (“Jurisdictional language need not contain the same culpability requirement as other elements of the offense”); Model Penal Code §2.02. So when Congress has said nothing about the mental state pertaining to a jurisdictional element, the default rule flips: Courts assume that Congress wanted such an element to stand outside the otherwise applicable mens rea requirement. In line with that practice, courts have routinely held that a criminal defendant need not know of a federal crime’s interstate commerce connection to be found guilty. See, e.g., United States v. Jinian, 725 F. 3d 954, 964–966 (CA9 2013); United States v. Lindemann, 85 F. 3d 1232, 1241 (CA7 1996); United States v. Blackmon, 839 F. 2d 900, 907 (CA2 1988). Those courts have recognized, as we do here, that Congress viewed the commerce element as distinct from, and subject to a different rule than, the elements describing the substantive offense. Still more strikingly, courts have distinguished between the two kinds of elements in contexts, similar to this one, in which the judicial task is to compare federal and state offenses. The Assimilative Crimes Act (ACA), 18 U. S. C. §13(a), subjects federal enclaves, like military bases, to state criminal laws except when they punish the same conduct as a federal statute. The ACA thus requires courts to decide when a federal and a state law are sufficiently alike that only the federal one will apply. And we have held that, in making that assessment, courts should ignore jurisdictional elements: When the “differences among elements” of the state and federal crimes “reflect jurisdictional, or other technical, considerations” alone, then the state law will have no effect in the area. Lewis v. United States, 523 U. S. 155, 165 (1998) ; see also id., at 182 (Kennedy, J., dissenting) (agreeing that courts should “look beyond . . . jurisdictional elements,” and focus only on substantive ones, in determining whether “the elements of the two crimes are the same”). In such a case, we reasoned—just as we do now—that Congress meant for the federal jurisdictional element to be set aside. And lower courts have uniformly adopted the same approach when comparing federal and state crimes in order to apply the federal three-strikes statute. That law imposes mandatory life imprisonment on a person convicted on three separate occasions of a “serious violent felony.” 18 U. S. C. §3559(c)(1). Sounding very much like the INA, the three-strikes statute defines such a felony to include “a Federal or State offense, by whatever designation and wherever committed, consisting of” specified crimes (e.g., murder, manslaughter, robbery) “as described in” listed federal criminal statutes. §3559(c)(2)(F). In deciding whether a state crime of conviction thus corresponds to an enumerated federal statute, every court to have faced the issue has ignored the statute’s jurisdictional element. See, e.g., United States v. Rosario-Delgado, 198 F. 3d 1354, 1357 (CA11 1999) (per curiam); United States v. Wicks, 132 F. 3d 383, 386–387 (CA7 1997). Judge Wood, writing for the Seventh Circuit, highlighted the phrase “a Federal or State offense, by whatever designation and wherever committed”—the three-strikes law’s version of §1101(a)(43)’s penultimate sentence. “It is hard to see why Congress would have used this language,” she reasoned, “if it had meant that every detail of the federal offense, including its jurisdictional element[ ], had to be replicated in the state offense.” Id., at 386–387. Just so, too, in the INA—whose “aggravated felony” provisions operate against, and rely on, an established legal backdrop distinguishing between jurisdictional and substantive elements.[12] Luna objects to drawing that line on the ground that it is too hard to tell the difference between the two. See Brief for Petitioner 26–28 (discussing, in particular, statutes criminalizing the destruction of federal property and sending threats via the Postal Service). But that contention collides with the judicial experience just described. Courts regularly separate substantive from jurisdictional elements in applying federal criminal statutes’ mens rea requirements; so too in implementing other laws that require a comparison of federal and state offenses. And from all we can see, courts perform that task with no real trouble: Luna has not pointed to any divisions between or within Circuits arising from the practice. We do not deny that some tough questions may lurk on the margins—where an element that makes evident Congress’s regulatory power also might play a role in defining the behavior Congress thought harmful. But a standard interstate commerce element, of the kind appearing in a great many federal laws, is almost always a simple jurisdictional hook—and courts may as easily acknowledge that fact in enforcing the INA as they have done in other contexts. C Luna makes a final argument opposing our reading of §1101(a)(43): If Congress had meant for “ordinary state-law” crimes like arson to count as aggravated felonies, it would have drafted the provision to make that self-evident. Brief for Petitioner 20. Congress, Luna submits, would have used the generic term for those crimes—e.g., “arson”—rather than demanding that the state law of conviction correspond to a listed federal statute. See id., at 20–23. Or else, Luna (and the dissent) suggests, see id., at 24; post, at 13, Congress would have expressly distinguished between substantive and jurisdictional elements, as it did in an unrelated law mandating the pretrial detention of any person convicted of a federal offense “described in [a certain federal statute], or of a State or local offense that would have been an offense described in [that statute] if a circumstance giving rise to Federal jurisdiction had existed,” 18 U. S. C. §3142(e)(2)(A). But as an initial matter, Congress may have had good reason to think that a statutory reference would capture more accurately than a generic label the range of state convictions warranting automatic deportation. The clause of §1101(a)(43) applying to Luna’s case well illustrates the point. By referring to 18 U. S. C. §844(i), that provision incorporates not only the garden-variety arson offenses that a generic “arson” label would cover, but various explosives offenses too. See Brief for Petitioner 23, n. 7 (conceding that had Congress used the term “arson,” it would have had to separately identify the explosives crimes encompassed in §844(i)). And the elements of generic arson are themselves so uncertain as to pose problems for a court having to decide whether they are present in a given state law. See Poulos, The Metamorphosis of the Law of Arson, 51 Mo. L. Rev. 295, 364, 387–435 (1986) (describing multiple conflicts over what conduct the term “arson” includes). Nor is the clause at issue here unusual in those respects: Section 1101(a)(43) includes many other statutory references that do not convert easily to generic labels. See, e.g., §1101(a)(43)(E)(ii) (listing federal statutes defining various firearms offenses). To be sure, Congress used such labels to describe some crimes qualifying as aggravated felonies—for example, “murder, rape, or sexual abuse of a minor.” §1101(a)(43)(A). But what is good for some crimes is not for others. The use of a federal statutory reference shows only that Congress thought it the best way to identify certain substantive crimes—not that Congress wanted (in conflict with the penultimate sentence) to exclude state and foreign versions of those offenses for lack of a jurisdictional element. Still more, Congress’s omission of statutory language specifically directing courts to ignore those elements cannot tip the scales in Luna’s favor. We have little doubt that “Congress could have drafted [§1101(a)(43)] with more precision than it did.” Graham County Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U. S. 409, 422 (2005) . But the same could be said of many (even most) statutes; as to that feature, §1101(a)(43) can join a well-populated club. And we have long been mindful of that fact when interpreting laws. Rather than expecting (let alone demanding) perfection in drafting, we have routinely construed statutes to have a particular meaning even as we acknowledged that Congress could have expressed itself more clearly. See, e.g., ibid.; Florida Dept. of Revenue v. Piccadilly Cafeterias, Inc., 554 U. S. 33, 41 (2008) ; Scarborough v. United States, 431 U. S. 563 –571, 575 (1977). The question, then, is not: Could Congress have indicated (or even did Congress elsewhere indicate) in more crystalline fashion that comparisons of federal and state offenses should disregard elements that merely establish legislative jurisdiction? The question is instead, and more simply: Is that the right and fair reading of the statute before us? And the answer to that question, given the import of §1101(a)(43)’s penultimate sentence and the well-settled background rule distinguishing between jurisdictional and substantive elements, is yes. III That reading of §1101(a)(43) resolves this case. Luna has acknowledged that the New York arson law differs from the listed federal statute, 18 U. S. C. §844(i), in only one respect: It lacks an interstate commerce element. See Pet. for Cert. 3. And Luna nowhere contests that §844(i)’s commerce element—featuring the terms “in interstate or foreign commerce” and “affecting interstate or foreign commerce”—is of the standard, jurisdictional kind. See Tr. of Oral Arg. 12, 19; Scheidler, 547 U. S., at 17–18 (referring to the phrases “affect commerce” and “in commerce” as conventional “jurisdictional language”). For all the reasons we have given, such an element is properly ignored when determining if a state offense counts as an aggravated felony under §1101(a)(43). We accordingly affirm the judgment of the Second Circuit. It is so ordered.Notes 1 Compare Espinal-Andrades v. Holder, 777 F. 3d 163 (CA4 2015) (finding an aggravated felony in that circumstance); Spacek v. Holder, 688 F. 3d 536 (CA8 2012) (same); Nieto Hernandez v. Holder, 592 F. 3d 681 (CA5 2009) (same); Negrete-Rodriguez v. Mukasey, 518 F. 3d 497 (CA7 2008) (same); United States v. Castillo-Rivera, 244 F. 3d 1020 (CA9 2001) (same), with Bautista v. Attorney General, 744 F. 3d 54 (CA3 2014) (declining to find an aggravated felony). 2 That flat statement is infinitesimally shy of being wholly true. We have found a handful of state criminal laws with an interstate commerce element, out of the tens (or perhaps hundreds) of thousands of state crimes on the books. Mississippi, for example, lifted essentially verbatim the text of the federal money laundering statute when drafting its own, and thus wound up with such an element. See Miss. Code Ann. §97–23–101 (rev. 2014). But because the incidence of such laws is so vanishingly small, and the few that exist play no role in Luna’s arguments, we proceed without qualifying each statement of the kind above. 3 Black’s Law Dictionary 401 (5th ed. 1979) (defining “describe” as to “express, explain, set forth, relate, recount, narrate, depict, delineate, portray”). Luna also cites Webster’s New Collegiate Dictionary 307 (1976), which defines “describe” to mean “to represent or give an account of in words.” 4 See American Heritage Dictionary of the English Language 490 (5th ed. 2011) (defining “describe” as “[t]o convey an idea or impression of ”); Webster’s Third New International Dictionary 610 (1986) (defining “describe” as “to convey an image or notion of” or “trace or traverse the outline of ”). 5 The dissent disagrees, contending that the word “describe” decides this case in Luna’s favor because a “description cannot refer to features that the thing being described does not have.” Post, at 5 (opinion of Sotomayor, J.). Says the dissent: If a Craigslist ad “describes” an apartment as having an “in-unit laundry, a dishwasher, rooftop access, central A/C, and a walk-in closet,” it does not describe an apartment lacking rooftop access. Ibid. That is true enough, but irrelevant. The dissent is right that when someone describes an object by a list of specific characteristics, he means that the item has each of those attributes. But things are different when someone uses a more general descriptor—even when that descriptor (as here, a federal statute) itself has a determinate set of elements. It would be natural, for example, to say (in the exact syntax of §1101(a)(43)) that a person followed the itinerary for a journey through Brazil that is “described in” a Lonely Planet guide if he traveled every leg of the tour other than a brief “detour north to Petrópolis.” The Lonely Planet, On the Road: Destination Brazil, http://media.lonelyplanet.com/shop/pdfs/brazil-8-getting-started.pdf (all Internet materials as last visited May 16, 2016). And similarly, a person would say that she had followed the instructions for setting up an iPhone that are “described in” the user’s manual even if she in fact ignored the one (specifically highlighted there) telling her to begin by “read[ing] important safety information” to “avoid injury.” Apple, Set Up iPhone, http://help.apple.com/iphone/9/#iph3bf43d79. 6 See 8 U. S. C. §1101(a)(43)(D) (“an offense described in” 18 U. S. C. §1956, which criminalizes laundering of monetary instruments); ibid. (“an offense described in” 18 U. S. C. §1957, which criminalizes engaging in monetary transactions involving property derived from specified unlawful activities); §1101(a)(43)(E)(i) (three “offense[s] described in” 18 U. S. C. §§842(h)–(i), 844(d), which criminalize activities involving explosives); ibid. (“an offense described in” 18 U. S. C. §844(e), which criminalizes threatening to cause death, injury, or property damage using explosives); ibid. (“an offense described in” 18 U. S. C. §844(i), which criminalizes using fire or explosives to cause property damage); §1101(a)(43)(E)(ii) (six “offense[s] described in” 18 U. S. C. §§922(g)(1)–(5), ( j), which criminalize possessing a firearm in various circumstances); ibid. (two “offense[s] described in” 18 U. S. C. §§922(n), 924(b),which criminalize transporting or receiving a firearm under certain circumstances); §1101(a)(43)(E)(iii) (“an offense described in” 26 U. S. C. §5861( j), which criminalizes transporting an unregistered firearm); §1101(a)(43)(H) (“an offense described in” 18 U. S. C. §875, which criminalizes making a threat to kidnap or a ransom demand); ibid. (“an offense described in” 18 U. S. C. §1202(b), which criminalizes possessing, receiving, or transmitting proceeds of a kidnapping); §1101(a)(43)(I) (“an offense described in” 18 U. S. C. §2251, which criminalizes sexually exploiting a child); ibid. (“an offense described in” 18 U. S. C. §2251A, which criminalizes selling a child for purposes of child pornography); ibid. (“an offense described in 18 U. S. C. §2252, which criminalizes various activities relating to child pornography); §1101(a)(43)(J) (“an offense described in” 18 U. S. C. §1962, which criminalizes activities relating to racketeering); ibid. (“an offense described in” 18 U. S. C. §1084, which criminalizes transmitting information to facilitate gambling); §1101(a)(43)(K)(ii) (“an offense described in” 18 U. S. C. §2421, which criminalizes transporting a person for purposes of prostitution); ibid. (“an offense described in” 18 U. S. C. §2422, which criminalizes coercing or enticing a person to travel for purposes of prostitution); ibid. (“an offense described in” 18 U. S. C. §2423, which criminalizes transporting a child for purposes of prostitution); §1101(a)(43)(K)(iii) (“an offense described in” 18 U. S. C. §1591(a)(1), which criminalizes sex trafficking of children, or of adults by force, fraud, or coercion). 7 The dissent replies: What’s the big deal? See post, at 10. After all, it reasons, some listed federal statutes—specifically, those prohibiting treason, levying war against the United States, and disclosing national defense information—will lack state or foreign analogues even under our construction. See ibid. But Congress’s inclusion of a few federal offenses that, by their nature, have no state or foreign analogues hardly excuses expelling from the Act’s coverage the countless state and foreign versions of 27 other serious crimes. 8 Luna’s position, in addition to producing this bizarre patchwork of coverage, conflicts with our ordinary assumption that Congress, when drafting a statute, gives each provision independent meaning. See United States v. Butler, 297 U. S. 1, 65 (1936) (“These words cannot be meaningless, else they would not have been used”). Until its most recent amendment, §1101(a)(43)(J ) provided that the term “aggravated felony” included any “offense described in [ 18 U. S. C. §1962] (relating to racketeer influenced corrupt organizations) for which a sentence of 5 years’ imprisonment or more may be imposed.” 8 U. S. C. §1101(a)(43)(J ) (1994 ed., Supp. I). (That provision now incorporates two more federal crimes, and uses one year of prison as the threshold.) The federal racketeering statute cited has an interstate commerce element; analogous state and foreign laws (per usual) do not, and therefore would fall outside §1101(a)(43)(J ) on Luna’s reading. But if Congress had meant to so exclude those state and foreign counterparts, then §1101(a)(43)(J )’s final clause—“for which a sentence of 5 years’ imprisonment may be imposed”—would have been superfluous, because federal racketeering is always punishable by more than five years’ imprisonment, see 18 U. S. C. §1963(a). That language’s presence shows that Congress thought §1101(a)(43)(J ) would sweep in some state and foreign laws: The final clause served to filter out such statutes when—but only when—they applied to less serious conduct than the federal racketeering offense. 9 The dissent attempts a variant of Luna’s “not so serious” argument, but to no better effect. Claims the dissent: Even if Congress could not have viewed “interstate crimes [as] worse than wholly intrastate crimes,” it might have thought that, say, “arsons prosecuted as federal crimes are more uniformly serious than arsons prosecuted as state crimes.” Post, at 14 (emphasis added). But we see no call to suppose that Congress regarded state prosecutions as Grapefruit League versions of the Big Show. Cf. Mistretta v. United States, 488 U. S. 361, 427 (1989) (Scalia, J., dissenting). In our federal system, “States possess primary authority for defining and enforcing” criminal laws, including those prohibiting the gravest crimes. Brecht v. Abrahamson, 507 U. S. 619, 635 (1993) . For that reason, even when U. S. Attorneys have jurisdiction, they are generally to defer to, rather than supplant, state prosecutions of serious offenses. See U. S. Attorneys’ Manual: Principles of Federal Prosecution §9–27.240 (1997). And still more obviously, the dissent’s theory fails with respect to foreign convictions. That a foreign sovereign prosecutes a given crime reflects nothing about its gravity, but only about its location. 10 In all those States, arsons of every description (whether of one’s own or another’s property) would fall outside the “crime of violence” provision. See Tr. of Oral Arg. 29, 46 (Solicitor General noting that the categorical approach to comparing federal and state crimes produces that effect). And contrary to the dissent’s suggestion, post, at 6, n. 2, that would be true of the most dangerous arsons, as well as of less serious ones. The dissent similarly fails to take into account the categorical approach’s rigorous requirements when discussing a couple of the non-arson offenses discussed above. (Still others, the dissent wholly ignores.) It speculates that if the exact right state charge is filed, some of that conduct “may” qualify, through the crime-of-violence provision or some other route, as an aggravated felony. Ibid. “May” is very much the operative word there, because—depending on the elements of the state offense chosen—that conduct also “may not.” And the dissent never explains why Congress would have left the deportation of dangerous felons to such prosecutorial happenstance. 11 The dissent well-nigh embraces those consequences, arguing that a narrow reading of “aggravated felony” would make more convicted criminals removable under other statutory provisions, all of which allow for relief at the Attorney General’s discretion. See post, at 8, 15 (lamenting that aliens convicted of aggravated felonies may not “even appeal[ ] to the mercy of the Attorney General”). But Congress made a judgment that aliens convicted of certain serious offenses (irrespective of whether those convictions were based on federal, state, or foreign law) should be not only removable but also ineligible for discretionary relief. It is not our place to second-guess that decision. 12 The dissent declares our discussion of the three-strikes law, the Assimilative Crime Act (ACA), and mens rea “unhelpful” on the ground that all three contexts are somehow “differ[ent].” Post, at 10–13. But what makes them relevantly so the dissent fails to explain. First, the dissent errs in suggesting that the uniform judicial interpretation of the three-strikes law ignores only “place-based jurisdiction elements” (because, so says the dissent, of the phrase “wherever committed”). Post, at 13. As Judge Wood’s analysis indicates, that is a theory of the dissent’s own creation; the actual appellate decisions apply to all jurisdictional elements, not just territorial ones. Next, the dissent goes wrong in claiming that the ACA is not pertinent because this Court adopted a different method for matching substantive elements under that law than under the INA. See post, at 12. For even as the Court made that choice, it unanimously agreed that, however substantive elements should be compared, jurisdictional elements should be disregarded. See Lewis v. United States, 523 U. S. 155, 165 (1998) ; id., at 182 (Kennedy, J., dissenting). And finally, the dissent does nothing to undermine our point on mens rea by noting that Congress very occasionally dispenses with that requirement for substantive elements. See post, at 11. As just shown, the default rule respecting mental states flips as between jurisdictional and substantive elements, see supra, at 15–16—reflecting the view (also at play in the three-strikes and ACA contexts) that Congress generally means to treat the two differently. That leaves the dissent with nothing except its observation that when applying the beyond-a-reasonable-doubt and jury-trial requirements, the Court does not distinguish between jurisdictional and substantive elements. See post, at 10. But the dissent forgets that those commands are constitutional in nature; a principle of statutory interpretation distinguishing between the two kinds of elements, as best reflecting Congress’s intent, could not bear on those mandates.

579.US.2015_15-6092

The Armed Career Criminal Act (ACCA) imposes a 15-year mandatory minimum sentence on a defendant convicted of being a felon in possession of a firearm who also has three prior state or federal convictions “for a violent felony,” including “burglary, arson, or extortion.” 18 U. S. C. §§924(e)(1), (e)(2)(B)(ii). To determine whether a prior conviction is for one of those listed crimes, courts apply the “categorical approach”—they ask whether the elements of the offense forming the basis for the conviction sufficiently match the elements of the generic (or commonly understood) version of the enumerated crime. See Taylor v. United States, 495 U. S. 575 –601. “Elements” are the constituent parts of a crime’s legal definition, which must be proved beyond a reasonable doubt to sustain a conviction; they are distinct from “facts,” which are mere real-world things—extraneous to the crime’s legal requirements and thus ignored by the categorical approach. When a statute defines only a single crime with a single set of elements, application of the categorical approach is straightforward. But when a statute defines multiple crimes by listing multiple, alternative elements, the elements-matching required by the categorical approach is more difficult. To decide whether a conviction under such a statute is for a listed ACCA offense, a sentencing court must discern which of the alternative elements was integral to the defendant’s conviction. That determination is made possible by the “modified categorical approach,” which permits a court to look at a limited class of documents from the record of a prior conviction to determine what crime, with what elements, a defendant was convicted of before comparing that crime’s elements to those of the generic offense. See, e.g., Shepard v. United States, 544 U. S. 13 . This case involves a different type of alternatively worded statute—one that defines only one crime, with one set of elements, but which lists alternative factual means by which a defendant can satisfy those elements. Here, petitioner Richard Mathis pleaded guilty to being a felon in possession of a firearm. Because of his five prior Iowa burglary convictions, the Government requested an ACCA sentence enhancement. Under the generic offense, burglary requires unlawful entry into a “building or other structure.” Taylor, 495 U. S., at 598. The Iowa statute, however, reaches “any building, structure, [or] land, water, or air vehicle.” Iowa Code §702.12. Under Iowa law, that list of places does not set out alternative elements, but rather alternative means of fulfilling a single locational element. The District Court applied the modified categorical approach, found that Mathis had burgled structures, and imposed an enhanced sentence. The Eighth Circuit affirmed. Acknowledging that the Iowa statute swept more broadly than the generic statute, the court determined that, even if “structures” and “vehicles” were not separate elements but alternative means of fulfilling a single element, a sentencing court could still invoke the modified categorical approach. Because the record showed that Mathis had burgled structures, the court held, the District Court’s treatment of Mathis’s prior convictions as ACCA predicates was proper. Held: Because the elements of Iowa’s burglary law are broader than those of generic burglary, Mathis’s prior convictions cannot give rise to ACCA’s sentence enhancement. Pp. 7–19. (a) This case is resolved by this Court’s precedents, which have repeatedly held, and in no uncertain terms, that a state crime cannot qualify as an ACCA predicate if its elements are broader than those of a listed generic offense. See, e.g., Taylor, 495 U. S., at 602. The “underlying brute facts or means” by which the defendant commits his crime, Richardson v. United States, 526 U. S. 813 , make no difference; even if the defendant’s conduct, in fact, fits within the definition of the generic offense, the mismatch of elements saves him from an ACCA sentence. ACCA requires a sentencing judge to look only to “the elements of the [offense], not to the facts of [the] defendant’s conduct.” Taylor, 495 U. S., at 601. This Court’s cases establish three basic reasons for adhering to an elements-only inquiry. First, ACCA’s text, which asks only about a defendant’s “prior convictions,” indicates that Congress meant for the sentencing judge to ask only whether “the defendant had been convicted of crimes falling within certain categories,” id., at 600, not what he had done. Second, construing ACCA to allow a sentencing judge to go any further would raise serious Sixth Amendment concerns because only a jury, not a judge, may find facts that increase the maximum penalty. See Apprendi v. New Jersey, 530 U. S. 466 . And third, an elements-focus avoids unfairness to defendants, who otherwise might be sentenced based on statements of “non-elemental fact[s]” that are prone to error because their proof is unnecessary to a conviction. Descamps v. United States, 570 U. S. ___, ___. Those reasons remain as strong as ever when a statute, like Iowa’s burglary statute, lists alternative means of fulfilling one (or more) of a crime’s elements. ACCA’s term “convictions” still supports an elements-based inquiry. The Sixth Amendment problems associated with a court’s exploration of means rather than elements do not abate in the face of a statute like Iowa’s: Alternative factual scenarios remain just that, and thus off-limits to sentencing judges. Finally, a statute’s listing of disjunctive means does nothing to mitigate the possible unfairness of basing an increased penalty on something not legally necessary to a prior conviction. Accordingly, whether means are listed in a statute or not, ACCA does not care about them; rather, its focus, as always, remains on a crime’s elements. Pp. 7–16. (b) The first task for a court faced with an alternatively phrased statute is thus to determine whether the listed items are elements or means. That threshold inquiry is easy here, where a State Supreme Court ruling answers the question. A state statute on its face could also resolve the issue. And if state law fails to provide clear answers, the record of a prior conviction itself might prove useful to determining whether the listed items are elements of the offense. If such record materials do not speak plainly, a sentencing judge will be unable to satisfy “Taylor’s demand for certainty.” Shepard, 544 U. S., at 21. But between the record and state law, that kind of indeterminacy should prove more the exception than the rule. Pp. 16–18. 786 F. 3d 1068, reversed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, and Sotomayor, JJ., joined. Kennedy, J., and Thomas, J., filed concurring opinions. Breyer, J., filed a dissenting opinion, in which Ginsburg, J., joined. Alito, J., filed a dissenting opinion.

The Armed Career Criminal Act (ACCA or Act), 18 U. S. C. §924(e), imposes a 15-year mandatory minimum sentence on certain federal defendants who have three prior convictions for a “violent felony,” including “burglary, arson, or extortion.” To determine whether a past conviction is for one of those offenses, courts compare the elements of the crime of conviction with the elements of the “generic” version of the listed offense—i.e., the offense as commonly understood. For more than 25 years, our decisions have held that the prior crime qualifies as an ACCA predicate if, but only if, its elements are the same as, or narrower than, those of the generic offense. The question in this case is whether ACCA makes an exception to that rule when a defendant is convicted under a statute that lists multiple, alternative means of satisfying one (or more) of its elements. We decline to find such an exception. I A ACCA prescribes a 15-year mandatory minimum sentence if a defendant is convicted of being a felon in possession of a firearm following three prior convictions for a “violent felony.” §924(e)(1). (Absent that sentence enhancement, the felon-in-possession statute sets a 10-year maximum penalty. See §924(a)(2).) ACCA defines the term “violent felony” to include any felony, whether state or federal, that “is burglary, arson, or extortion.” §924(e)(2)(B)(ii). In listing those crimes, we have held, Congress referred only to their usual or (in our terminol-ogy) generic versions—not to all variants of the offenses. See Taylor v. United States, 495 U. S. 575, 598 (1990) . That means as to burglary—the offense relevant in this case—that Congress meant a crime “contain[ing] the following elements: an unlawful or unprivileged entry into . . . a building or other structure, with intent to commit a crime.” Ibid. To determine whether a prior conviction is for generic burglary (or other listed crime) courts apply what is known as the categorical approach: They focus solely on whether the elements of the crime of conviction sufficiently match the elements of generic burglary, while ignoring the particular facts of the case. See id., at 600–601. Distinguishing between elements and facts is therefore central to ACCA’s operation. “Elements” are the “constituent parts” of a crime’s legal definition—the things the “prosecution must prove to sustain a conviction.” Black’s Law Dictionary 634 (10th ed. 2014). At a trial, they are what the jury must find beyond a reasonable doubt to convict the defendant, see Richardson v. United States, 526 U. S. 813, 817 (1999) ; and at a plea hearing, they are what the defendant necessarily admits when he pleads guilty, see McCarthy v. United States, 394 U. S. 459, 466 (1969) . Facts, by contrast, are mere real-world things—extraneous to the crime’s legal requirements. (We have sometimes called them “brute facts” when distinguishing them from elements. Richardson, 526 U. S., at 817.) They are “circumstance[s]” or “event[s]” having no “legal effect [or] consequence”: In particular, they need neither be found by a jury nor admitted by a defendant. Black’s Law Dictionary 709. And ACCA, as we have always understood it, cares not a whit about them. See, e.g., Taylor, 495 U. S., at 599–602. A crime counts as “burglary” under the Act if its elements are the same as, or narrower than, those of the generic offense. But if the crime of conviction covers any more conduct than the generic offense, then it is not an ACCA “burglary”—even if the defendant’s actual conduct (i.e., the facts of the crime) fits within the generic offense’s boundaries. The comparison of elements that the categorical approach requires is straightforward when a statute sets out a single (or “indivisible”) set of elements to define a single crime. The court then lines up that crime’s elements alongside those of the generic offense and sees if they match. So, for example, this Court found that a California statute swept more broadly than generic burglary because it criminalized entering a location (even if lawfully) with the intent to steal, and thus encompassed mere shoplifting. See id., at 591; Descamps v. United States, 570 U. S. ___, ___–___ (2013) (slip op., at 5–6). Accordingly, no conviction under that law could count as an ACCA predicate, even if the defendant in fact made an illegal entry and so committed burglary in its generic form. See id., at ___–___ (slip op., at 22–23). Some statutes, however, have a more complicated (sometimes called “divisible”) structure, making the comparison of elements harder. Id., at ___ (slip op., at 5). A single statute may list elements in the alternative, and thereby define multiple crimes. Suppose, for example, that the California law noted above had prohibited “the lawful entry or the unlawful entry” of a premises with intent to steal, so as to create two different offenses, one more serious than the other. If the defendant were convicted of the offense with unlawful entry as an element, then his crime of conviction would match generic burglary and count as an ACCA predicate; but, conversely, the conviction would not qualify if it were for the offense with lawful entry as an element. A sentencing court thus requires a way of figuring out which of the alternative elements listed—lawful entry or unlawful entry—was integral to the defendant’s conviction (that is, which was necessarily found or admitted). See id., at ___ (slip op., at 6). To address that need, this Court approved the “modified categorical approach” for use with statutes having multiple alternative elements. See, e.g., Shepard v. United States, 544 U. S. 13, 26 (2005) . Under that approach, a sentencing court looks to a limited class of documents (for example, the indictment, jury instructions, or plea agreement and colloquy) to determine what crime, with what elements, a defendant was convicted of. See ibid.; Taylor, 495 U. S., at 602. The court can then compare that crime, as the categorical approach commands, with the relevant generic offense. This case concerns a different kind of alternatively phrased law: not one that lists multiple elements disjunctively, but instead one that enumerates various factual means of committing a single element. See generally Schad v. Arizona, 501 U. S. 624, 636 (1991) (plurality opinion) (“[L]egislatures frequently enumerate alternative means of committing a crime without intending to define separate elements or separate crimes”). To use a hypothetical adapted from two of our prior decisions, suppose a statute requires use of a “deadly weapon” as an element of a crime and further provides that the use of a “knife, gun, bat, or similar weapon” would all qualify. See Descamps, 570 U. S., at ___ (slip op., at 16); Richardson, 526 U. S., at 817. Because that kind of list merely specifies diverse means of satisfying a single element of a single crime—or otherwise said, spells out various factual ways of committing some component of the offense—a jury need not find (or a defendant admit) any particular item: A jury could convict even if some jurors “conclude[d] that the defendant used a knife” while others “conclude[d] he used a gun,” so long as all agreed that the defendant used a “deadly weapon.” Ibid.; see Descamps, 570 U. S., at ___ (slip op., at 14) (describing means, for this reason, as “legally extraneous circumstances”). And similarly, to bring the discussion back to burglary, a statute might—indeed, as soon discussed, Iowa’s burglary law does—itemize the various places that crime could occur as disjunctive factual scenarios rather than separate elements, so that a jury need not make any specific findings (or a defendant admissions) on that score. The issue before us is whether ACCA treats this kind of statute as it does all others, imposing a sentence enhancement only if the state crime’s elements correspond to those of a generic offense—or instead whether the Act makes an exception for such a law, so that a sentence can be enhanced when one of the statute’s specified means creates a match with the generic offense, even though the broader element would not. B Petitioner Richard Mathis pleaded guilty to being a felon in possession of a firearm. See §922(g). At sentencing, the Government asked the District Court to impose ACCA’s 15-year minimum penalty based on Mathis’s five prior convictions for burglary under Iowa law. Iowa’s burglary statute, all parties agree, covers more conduct than generic burglary does. See Brief for Petitioner 36; Brief for United States 44. The generic offense requires unlawful entry into a “building or other structure.” Taylor, 495 U. S., at 598; supra, at 2. Iowa’s statute, by contrast, reaches a broader range of places: “any building, structure, [or] land, water, or air vehicle.” Iowa Code §702.12 (2013) (emphasis added). And those listed locations are not alternative elements, going toward the creation of separate crimes. To the contrary, they lay out alternative ways of satisfying a single locational element, as the Iowa Supreme Court has held: Each of the terms serves as an “alternative method of committing [the] single crime” of burglary, so that a jury need not agree on which of the locations was actually involved. State v. Duncan, 312 N. W. 2d 519, 523 (Iowa 1981); see State v. Rooney, 862 N. W. 2d 367, 376 (Iowa 2015) (discussing the single “broadly phrased . . . element of place” in Iowa’s burglary law). In short, the statute defines one crime, with one set of elements, broader than generic burglary—while specifying multiple means of fulfilling its locational element, some but not all of which (i.e., buildings and other structures, but not vehicles) satisfy the generic definition. The District Court imposed an ACCA enhancement on Mathis after inspecting the records of his prior convictions and determining that he had burgled structures, rather than vehicles. See App. 34–35. The Court of Appeals for the Eighth Circuit affirmed. 786 F. 3d 1068 (2015). It acknowledged that Iowa’s burglary statute, by covering vehicles in addition to structures, swept more broadly than generic burglary. See id., at 1074. But it noted that if structures and vehicles were separate elements, each part of a different crime, then a sentencing court could invoke the modified categorical approach and look to old record materials to see which of those crimes the defendant had been convicted of. See id., at 1072–1074. And the Court of Appeals thought nothing changed if structures and vehicles were not distinct elements but only alternative means: “Whether [such locations] amount to alternative elements or merely alternative means to fulfilling an element,” the Eighth Circuit held, a sentencing court “must apply the modified categorical approach” and inspect the records of prior cases. Id., at 1075. If the court found from those materials that the defendant had in fact committed the offense in a way that satisfied the definition of generic burglary—here, by burgling a structure rather than a vehicle—then the court should treat the conviction as an ACCA predicate. And that was so, the Court of Appeals stated, even though the elements of the crime of conviction, in encompassing both types of locations, were broader than those of the relevant generic offense. See id., at 1074–1075. In this circumstance, the court thus found, ACCA’s usual elements-based inquiry would yield to a facts-based one. That decision added to a Circuit split over whether ACCA’s general rule—that a defendant’s crime of conviction can count as a predicate only if its elements match those of a generic offense—gives way when a statute happens to list various means by which a defendant can satisfy an element.[1] We granted certiorari to resolve that division, 577 U. S. ___ (2016), and now reverse. II A As just noted, the elements of Mathis’s crime of conviction (Iowa burglary) cover a greater swath of conduct than the elements of the relevant ACCA offense (generic burglary). See supra, at 5–6. Under our precedents, that undisputed disparity resolves this case. We have often held, and in no uncertain terms, that a state crime cannot qualify as an ACCA predicate if its elements are broader than those of a listed generic offense. See, e.g., Taylor, 495 U. S., at 602. How a given defendant actually perpetrated the crime—what we have referred to as the “underlying brute facts or means” of commission, Richardson, 526 U. S., at 817—makes no difference; even if his conduct fits within the generic offense, the mismatch of elements saves the defendant from an ACCA sentence. Those longstanding principles, and the reasoning that underlies them, apply regardless of whether a statute omits or instead specifies alternative possible means of commission. The itemized construction gives a sentencing court no special warrant to explore the facts of an offense, rather than to determine the crime’s elements and compare them with the generic definition. Taylor set out the essential rule governing ACCA cases more than a quarter century ago. All that counts under the Act, we held then, are “the elements of the statute of conviction.” 495 U. S., at 601. So, for example, the label a State assigns to a crime—whether “burglary,” “breaking and entering,” or something else entirely—has no relevance to whether that offense is an ACCA predicate. See id., at 590–592. And more to the point here: The same is true of “the particular facts underlying [the prior] convictions”—the means by which the defendant, in real life, committed his crimes. Id., at 600. That rule can seem counterintuitive: In some cases, a sentencing judge knows (or can easily discover) that the defendant carried out a “real” burglary, even though the crime of conviction also extends to other conduct. No matter. Under ACCA, Taylor stated, it is impermissible for “a particular crime [to] sometimes count towards enhancement and sometimes not, depending on the facts of the case.” Id., at 601. Accordingly, a sentencing judge may look only to “the elements of the [offense], not to the facts of [the] defendant’s conduct.” Ibid. That simple point became a mantra in our subsequent ACCA decisions.[2] At the risk of repetition (perhaps downright tedium), here are some examples. In Shepard: ACCA “refers to predicate offenses in terms not of prior conduct but of prior ‘convictions’ and the ‘element[s]’ of crimes.” 544 U. S., at 19 (alteration in original). In James v. United States: “[W]e have avoided any inquiry into the underlying facts of [the defendant’s] particular offense, and have looked solely to the elements of [burglary] as defined by [state] law.” 550 U. S. 192, 214 (2007) . In Sykes v. United States: “[W]e consider [only] the elements of the offense[,] without inquiring into the specific conduct of this particular offender.” 564 U. S. 1, 7 (2011) (quoting James, 550 U. S., at 202; emphasis in original). And most recently (and tersely) in Descamps: “The key [under ACCA] is elements, not facts.” 570 U. S., at ___ (slip op., at 5). Our decisions have given three basic reasons for adhering to an elements-only inquiry. First, ACCA’s text favors that approach. By enhancing the sentence of a defendant who has three “previous convictions” for generic burglary, §924(e)(1)—rather than one who has thrice committed that crime—Congress indicated that the sentencer should ask only about whether “the defendant had been convicted of crimes falling within certain categories,” and not about what the defendant had actually done. Taylor, 495 U. S., at 600. Congress well knows how to instruct sentencing judges to look into the facts of prior crimes: In other statutes, using different language, it has done just that. See United States v. Hayes, 555 U. S. 415, 421 (2009) (concluding that the phrase “an offense . . . committed” charged sentencers with considering non-elemental facts); Nijhawan v. Holder, 557 U. S. 29, 36 (2009) (construing an immigration statute to “call[ ] for a ‘circumstance-specific,’ not a ‘categorical’ interpretation”). But Congress chose another course in ACCA, focusing on only “the elements of the statute of conviction.” Taylor, 495 U. S., at 601. Second, a construction of ACCA allowing a sentencing judge to go any further would raise serious Sixth Amendment concerns. This Court has held that only a jury, and not a judge, may find facts that increase a maximum penalty, except for the simple fact of a prior conviction. See Apprendi v. New Jersey, 530 U. S. 466, 490 (2000) . That means a judge cannot go beyond identifying the crime of conviction to explore the manner in which the defendant committed that offense. See Shepard, 544 U. S., at 25 (plurality opinion); id., at 28 (Thomas, J., concurring in part and concurring in judgment) (stating that such an approach would amount to “constitutional error”). He is prohibited from conducting such an inquiry himself; and so too he is barred from making a disputed determination about “what the defendant and state judge must have understood as the factual basis of the prior plea” or “what the jury in a prior trial must have accepted as the theory of the crime.” See id., at 25 (plurality opinion); Descamps, 570 U. S., at ___ (slip op., at 14). He can do no more, consistent with the Sixth Amendment, than determine what crime, with what elements, the defendant was convicted of. And third, an elements-focus avoids unfairness to defendants. Statements of “non-elemental fact” in the records of prior convictions are prone to error precisely because their proof is unnecessary. Id., at ___ (slip op., at 15). At trial, and still more at plea hearings, a defendant may have no incentive to contest what does not matter under the law; to the contrary, he “may have good reason not to”—or even be precluded from doing so by the court. Ibid. When that is true, a prosecutor’s or judge’s mistake as to means, reflected in the record, is likely to go uncorrected. See ibid.[3] Such inaccuracies should not come back to haunt the defendant many years down the road by triggering a lengthy mandatory sentence. Those three reasons stay as strong as ever when a statute, instead of merely laying out a crime’s elements, lists alternative means of fulfilling one (or more) of them. ACCA’s use of the term “convictions” still supports an elements-based inquiry; indeed, that language directly refutes an approach that would treat as consequential a statute’s reference to factual circumstances not essential to any conviction. Similarly, the Sixth Amendment problems associated with a court’s exploration of means rather than elements do not abate in the face of a statute like Iowa’s: Whether or not mentioned in a statute’s text, alternative factual scenarios remain just that—and so remain off-limits to judges imposing ACCA enhancements. And finally, a statute’s listing of disjunctive means does nothing to mitigate the possible unfairness of basing an increased penalty on something not legally necessary to a prior conviction. Whatever the statute says, or leaves out, about diverse ways of committing a crime makes no difference to the defendant’s incentives (or lack thereof ) to contest such matters. For these reasons, the court below erred in applying the modified categorical approach to determine the means by which Mathis committed his prior crimes. 786 F. 3d, at 1075. ACCA, as just explained, treats such facts as irrelevant: Find them or not, by examining the record or anything else, a court still may not use them to enhance a sentence. And indeed, our cases involving the modified categorical approach have already made exactly that point. “[T]he only [use of that approach] we have ever allowed,” we stated a few Terms ago, is to determine “which element[s] played a part in the defendant’s conviction.” Descamps, 570 U. S., at ___, ___ (slip op., at 5, 8) (emphasis added); see Taylor, 495 U. S., at 602 (noting that the modified approach may be employed only to determine whether “a jury necessarily had to find” each element of generic burglary). In other words, the modified approach serves—and serves solely—as a tool to identify the elements of the crime of conviction when a statute’s disjunctive phrasing renders one (or more) of them opaque. See Descamps, 570 U. S., at ___ (slip op., at 8).[4] It is not to be repurposed as a technique for discovering whether a defendant’s prior conviction, even though for a too-broad crime, rested on facts (or otherwise said, involved means) that also could have satisfied the elements of a generic offense. B The Government and Justice Breyer claim that our longtime and exclusive focus on elements does not resolve this case because (so they say) when we talked about “elements,” we did not really mean it. “[T]he Court used ‘elements,’ ” the Government informs us, “not to distinguish between ‘means’ and ‘elements,’ ” but instead to refer to whatever the statute lists—whether means or elements. Brief for United States 8; see id., at 19. In a similar vein, Justice Breyer posits that every time we said the word “element,” we “used the word generally, simply to refer to the matter at issue,” without “intend[ing] to set forth a generally applicable rule.” Post, at 11–12 (dissenting opinion). But a good rule of thumb for reading our decisions is that what they say and what they mean are one and the same; and indeed, we have previously insisted on that point with reference to ACCA’s elements-only approach. In Descamps, the sole dissenting Justice made an argument identical to the one now advanced by the Government and Justice Breyer: that our prior caselaw had not intended to distinguish between statutes listing alternative elements and those setting out “merely alternative means” of commission. 570 U. S., at ___ (slip op., at 7) (opinion of Alito, J.).[5] The Court rejected that contention, stating that “[a]ll those decisions rested on the explicit premise that the laws contain[ed] statutory phrases that cover several different crimes, not several different methods of committing one offense”—in other words, that they listed alternative elements, not alternative means. Id., at ___, n. 2 (slip op., at 9, n. 2) (ellipsis and internal quotation marks omitted); see, e.g., Johnson v. United States, 559 U. S. 133, 144 (2010) ; Nijhawan, 557 U. S., at 35. That premise was important, we explained, because an ACCA penalty may be based only on what a jury “necessarily found” to convict a defendant (or what he necessar-ily admitted). Descamps, 570 U. S., at ___, ___ (slip op., at 11, 17). And elements alone fit that bill; a means, or (as we have called it) “non-elemental fact,” is “by definition[ ] not necessary to support a conviction.” Id., at ___, n. 3, __ (slip op., at 11, n. 3, 15); see supra, at 2.[6] Accordingly, Descamps made clear that when the Court had earlier said (and said and said) “elements,” it meant just that and nothing else. For that reason, this Court (including Justice Breyer) recently made clear that a court may not look behind the elements of a generally drafted statute to identify the means by which a defendant committed a crime. See Descamps, 570 U. S., at ___ (slip op., at 2). Consider if Iowa defined burglary as involving merely an unlawful entry into a “premises”—without any further elaboration of the types of premises that exist in the world (e.g., a house, a building, a car, a boat). Then, all agree, ACCA’s elements-focus would apply. No matter that the record of a prior conviction clearly indicated that the defendant burgled a house at 122 Maple Road—and that the jury found as much; because Iowa’s (hypothetical) law included an element broader than that of the generic offense, the defendant could not receive an ACCA sentence. Were that not so, this Court stated, “the categorical approach [would be] at an end”; the court would merely be asking “whether a particular set of facts leading to a conviction conforms to a generic ACCA offense.” Id., at ___ (slip op., at 19). That conclusion is common ground, and must serve as the baseline for anything Justice Breyer (or the Government) here argues. And contrary to his view, that baseline not only begins but also ends the analysis, because nothing material changes if Iowa’s law further notes (much as it does) that a “premises” may include “a house, a building, a car, or a boat.” That fortuity of legislative drafting affects neither the oddities of applying the categorical approach nor the reasons for doing so. On the one hand, a categorical inquiry can produce the same counter-intuitive conse-quences however a state law is written. Whether or not the statute lists various means of satisfying the “premises” element, the record of a prior conviction is just as likely to make plain that the defendant burgled that house on Maple Road and the jury knew it. On the other hand (and as already shown), the grounds—constitutional, statutory, and equitable—that we have offered for nonetheless using the categorical approach lose none of their force in the switch from a generally phrased statute (leaving means implicit) to a more particular one (expressly enumerating them). See supra, at 11. In every relevant sense, both functional and legal, the two statutes—one saying just “premises,” the other listing structures and vehicles—are the same. And so the same rule must apply: ACCA disregards the means by which the defendant committed his crime, and looks only to that offense’s elements. C The first task for a sentencing court faced with an alternatively phrased statute is thus to determine whether its listed items are elements or means. If they are elements, the court should do what we have previously approved: review the record materials to discover which of the enumerated alternatives played a part in the defendant’s prior conviction, and then compare that element (along with all others) to those of the generic crime. See ibid. But if instead they are means, the court has no call to decide which of the statutory alternatives was at issue in the earlier prosecution. Given ACCA’s indifference to how a defendant actually committed a prior offense, the court may ask only whether the elements of the state crime and generic offense make the requisite match. This threshold inquiry—elements or means?—is easy in this case, as it will be in many others. Here, a state court decision definitively answers the question: The listed premises in Iowa’s burglary law, the State Supreme Court held, are “alternative method[s]” of committing one offense, so that a jury need not agree whether the burgled location was a building, other structure, or vehicle. See Duncan, 312 N. W. 2d, at 523; supra, at 6. When a ruling of that kind exists, a sentencing judge need only follow what it says. See Schad, 501 U. S., at 636 (plurality opinion). Likewise, the statute on its face may resolve the issue. If statutory alternatives carry different punishments, then under Apprendi they must be elements. See, e.g., Colo. Rev. Stat. §18–4–203 (2015); Vt. Stat. Ann., Tit. 13, §1201 (Cum. Supp. 2015); see also 530 U. S., at 490 (requiring a jury to agree on any circumstance increasing a statutory penalty); supra, at 10. Conversely, if a statutory list is drafted to offer “illustrative examples,” then it includes only a crime’s means of commission. United States v. Howard, 742 F. 3d 1334, 1348 (CA11 2014); see United States v. Cabrera-Umanzor, 728 F. 3d 347, 353 (CA4 2013). And a statute may itself identify which things must be charged (and so are elements) and which need not be (and so are means). See, e.g., Cal. Penal Code Ann. §952 (West 2008). Armed with such authoritative sources of state law, federal sentencing courts can readily determine the nature of an alternatively phrased list. And if state law fails to provide clear answers, federal judges have another place to look: the record of a prior conviction itself. As Judge Kozinski has explained, such a “peek at the [record] documents” is for “the sole and limited purpose of determining whether [the listed items are] element[s] of the offense.” Rendon v. Holder, 782 F. 3d 466, 473–474 (CA9 2015) (opinion dissenting from denial of reh’g en banc).[7] (Only if the answer is yes can the court make further use of the materials, as previously described, see supra, at 12–13.) Suppose, for example, that one count of an indictment and correlative jury instructions charge a defendant with burgling a “building, structure, or vehicle”—thus reiterating all the terms of Iowa’s law. That is as clear an indication as any that each alternative is only a possible means of commission, not an element that the prosecutor must prove to a jury beyond a reasonable doubt. So too if those documents use a single umbrella term like “premises”: Once again, the record would then reveal what the prosecutor has to (and does not have to) demonstrate to prevail. See Descamps, 570 U. S., at ___ (slip op., at 17). Conversely, an indictment and jury instructions could indicate, by referencing one alternative term to the exclusion of all others, that the statute contains a list of elements, each one of which goes toward a separate crime. Of course, such record materials will not in every case speak plainly, and if they do not, a sentencing judge will not be able to satisfy “Taylor’s demand for certainty” when determining whether a defendant was convicted of a generic offense. Shepard, 544 U. S., at 21. But between those documents and state law, that kind of indeterminacy should prove more the exception than the rule. III Our precedents make this a straightforward case. For more than 25 years, we have repeatedly made clear that application of ACCA involves, and involves only, comparing elements. Courts must ask whether the crime of conviction is the same as, or narrower than, the relevant generic offense. They may not ask whether the defendant’s conduct—his particular means of committing the crime—falls within the generic definition. And that rule does not change when a statute happens to list possible alternative means of commission: Whether or not made explicit, they remain what they ever were—just the facts, which ACCA (so we have held, over and over) does not care about. Some have raised concerns about this line of decisions, and suggested to Congress that it reconsider how ACCA is written. See, e.g., Chambers v. United States, 555 U. S. 122, 133 (2009) (Alito, J., concurring in judgment); Descamps, 570 U. S., at ___ (slip op., at 2) (Kennedy, J., concurring). But whether for good or for ill, the elements-based approach remains the law. And we will not introduce inconsistency and arbitrariness into our ACCA decisions by here declining to follow its requirements. Everything this Court has ever said about ACCA runs counter to the Government’s position. That alone is sufficient reason to reject it: Coherence has a claim on the law. Because the elements of Iowa’s burglary law are broader than those of generic burglary, Mathis’s convictions under that law cannot give rise to an ACCA sentence. We accordingly reverse the judgment of the Court of Appeals. It is so ordered.Notes 1 Compare 786 F. 3d 1068 (CA8 2015) (case below) (recognizing such an exception); United States v. Ozier, 796 F. 3d 597 (CA6 2015) (same); United States v. Trent, 767 F. 3d 1046 (CA10 2014) (same), with Rendon v. Holder, 764 F. 3d 1077 (CA9 2014) (rejecting that exception); Omargharib v. Holder, 775 F. 3d 192 (CA4 2014) (same). 2 So too in our decisions applying the categorical approach outside the ACCA context—most prominently, in immigration cases. See, e.g., Kawashima v. Holder, 565 U. S. 478 –483 (2012) (stating that a judge must look to the “formal element[s] of a conviction[,] rather than to the specific facts underlying the crime,” in deciding whether to deport an alien for committing an “aggravated felony”). 3 To see the point most clearly, consider an example arising in the immigration context: A defendant charged under a statute that criminalizes “intentionally, knowingly, or recklessly” assaulting another—as exists in many States, see, e.g., Tex. Penal Code Ann. §22.01(a)(1) (West Cum. Supp. 2015)—has no apparent reason to dispute a prosecutor’s statement that he committed the crime intentionally (as opposed to recklessly) if those mental states are interchangeable means of satisfying a single mens rea element. But such a statement, if treated as reliable, could make a huge difference in a deportation proceeding years in the future, because an intentional assault (unlike a reckless one) qualifies as a “crime involving moral turpitude,” and so requires removal from the country. See In re Gomez-Perez, No. A200–958–511, p. 2 (BIA 2014). 4 Descamps made the point at some length, adding that the modified categorical approach “retains the categorical approach’s central feature: a focus on the elements, rather than the facts, of a crime. And it preserves the categorical approach’s basic method: comparing those elements with the generic offense’s. All the modified approach adds is a mechanism for making that comparison when a statute lists multiple, alternative elements, and so effectively creates ‘several different . . . crimes.’ If at least one, but not all of those crimes matches the generic version, a court needs a way to find out which the defendant was convicted of. That is the job, as we have always understood it, of the modified approach: to identify, from among several alternatives, the crime of conviction so that the court can compare it to the generic offense.” 570 U. S., at ___ (slip op., at 8) (citation omitted). 5 In another solo dissent, Justice Alito today switches gears, arguing not that our precedent is consistent with his means-based view, but instead that all of our ACCA decisions are misguided because all follow from an initial wrong turn in Taylor v. United States, 495 U. S. 575 (1990) . See post, at 2–3. To borrow the driving metaphor of his own dissent, Justice Alito thus locates himself entirely off the map of our caselaw. But that is not surprising; he has harshly criticized the categorical approach (and Apprendi too) for many years. See, e.g., Johnson v. United States, 576 U. S. ___, ___–___ (2015) (Alito, J., dissenting) (slip op., at 8–13); Descamps, 570 U. S., at ___–___ (Alito, J., dissenting) (slip op., at 4–5); Moncrieffe v. Holder, 569 U. S. ___, ___–___ (2013) (Alito, J., dissenting) (slip op., at 10–11); Chambers v. United States, 555 U. S. 122 –134 (2009) (Alito, J., concurring in judgment); see also Hurst v. Florida, 577 U. S. ___, ___ (2016) (Alito, J., dissenting) (slip op., at 2); Alleyne v. United States, 570 U. S. ___, ___–___ (2013) (Alito, J., dissenting) (slip op., at 1–2). 6 Justice Breyer’s dissent rests on the idea that, contrary to that long-accepted definition, a jury sometimes does “necessarily ha[ve] to find” a means of commission, see post, at 6 (quoting Taylor, 495 U. S., at 602)—but Descamps specifically refuted that argument too. In that case, Justice Alito made the selfsame claim: A jury, he averred, should be treated as having “necessarily found” any fact, even though non-elemental, that a later sentencing court can “infer[ ]” that the jury agreed on “as a practical matter.” 570 U. S., at ___ (Alito, J., dissenting) (slip op., at 15). The Court rejected that view, explaining that its ACCA decisions had always demanded that a jury necessarily agree as a legal matter—which meant on elements and not on means. See id., at ___, n. 3 (slip op., at 10, n. 3). The requirement, from the Court’s earliest decisions, was that a judge could impose a 15-year sentence based only on a legal “certainty,” not on his inference (however reasonable in a given case) about what a prior factfinder had thought. Shepard, 544 U. S., at 23; see Taylor, 495 U. S., at 602; supra, at 10. Or otherwise said, the relevant question was whether a defendant was legally convicted of a certain offense (with a certain set of elements), not whether a sentencing judge believes that the factfinder would have convicted him of that offense had it been on the books. See Carachuri-Rosendo v. Holder, 560 U. S. 563, 576 (2010) (rejecting such a “hypothetical” approach given a similar statute’s directive to “look to the conviction itself”). 7 Descamps previously recognized just this way of discerning whether a statutory list contains means or elements. See 570 U. S., at ___, n. 2 (slip op., at 8–9, n. 2). The Court there noted that indictments, jury instructions, plea colloquies and plea agreements will often “reflect the crime’s elements” and so can reveal—in some cases better than state law itself—whether a statutory list is of elements or means. Ibid. Accordingly, when state law does not resolve the means-or-elements question, courts should “resort[ ] to the [record] documents” for help in making that determination. Ibid.

579.US.2015_15-474

Petitioner, former Virginia Governor Robert McDonnell, and his wife, Maureen McDonnell, were indicted by the Federal Government on honest services fraud and Hobbs Act extortion charges related to their acceptance of $175,000 in loans, gifts, and other benefits from Virginia businessman Jonnie Williams, while Governor McDonnell was in office. Williams was the chief executive officer of Star Scientific, a Virginia-based company that had developed Anatabloc, a nutritional supplement made from anatabine, a compound found in tobacco. Star Scientific hoped that Virginia’s public universities would perform research studies on anatabine, and Williams wanted Governor McDonnell’s assistance in obtaining those studies. To convict the McDonnells, the Government was required to show that Governor McDonnell committed (or agreed to commit) an “official act” in exchange for the loans and gifts. An “official act” is defined as “any decision or action on any question, matter, cause, suit, proceeding or controversy, which may at any time be pending, or which may by law be brought before any public official, in such official’s official capacity, or in such official’s place of trust or profit.” 18 U. S. C. §201(a)(3). According to the Government, Governor McDonnell committed at least five “official acts,” including “arranging meetings” for Williams with other Virginia officials to discuss Star Scientific’s product, “hosting” events for Star Scientific at the Governor’s Mansion, and “contacting other government officials” concerning the research studies. The case was tried before a jury. The District Court instructed the jury that “official act” encompasses “acts that a public official customarily performs,” including acts “in furtherance of longer-term goals” or “in a series of steps to exercise influence or achieve an end.” Supp. App. 69–70. Governor McDonnell requested that the court further instruct the jury that “merely arranging a meeting, attending an event, hosting a reception, or making a speech are not, standing alone, ‘official acts,’ ” but the District Court declined to give that instruction. 792 F. 3d 478, 513 (internal quotation marks omitted). The jury convicted Governor McDonnell. Governor McDonnell moved to vacate his convictions on the ground that the definition of “official act” in the jury instructions was erroneous. He also moved for acquittal, arguing that there was insufficient evidence to convict him, and that the Hobbs Act and honest services statute were unconstitutionally vague. The District Court denied the motions, and the Fourth Circuit affirmed. Held: 1. An “official act” is a decision or action on a “question, matter, cause, suit, proceeding or controversy.” That question or matter must involve a formal exercise of governmental power, and must also be something specific and focused that is “pending” or “may by law be brought” before a public official. To qualify as an “official act,” the public official must make a decision or take an action on that question or matter, or agree to do so. Setting up a meeting, talking to another official, or organizing an event—without more—does not fit that definition of “official act.” Pp. 13–24. (a) The Government argues that the term “official act” encompasses nearly any activity by a public official concerning any subject, including a broad policy issue such as Virginia economic development. Governor McDonnell, in contrast, contends that statutory context compels a more circumscribed reading. Taking into account text, precedent, and constitutional concerns, the Court rejects the Government’s reading and adopts a more bounded interpretation of “official act.” Pp. 13–14. (b) Section 201(a)(3) sets forth two requirements for an “official act.” First, the Government must identify a “question, matter, cause, suit, proceeding or controversy” that “may at any time be pending” or “may by law be brought” before a public official. Second, the Government must prove that the public official made a decision or took an action “on” that “question, matter, cause, suit, proceeding or controversy,” or agreed to do so. Pp. 14–22. (1) The first inquiry is whether a typical meeting, call, or event is itself a “question, matter, cause, suit, proceeding or controversy.” The terms “cause,” “suit,” “proceeding,” and “controversy” connote a formal exercise of governmental power, such as a lawsuit, hearing, or administrative determination. Although it may be difficult to define the precise reach of those terms, a typical meeting, call, or event does not qualify. “Question” and “matter” could be defined more broadly, but under the familiar interpretive canon noscitur a sociis, a “word is known by the company it keeps.” Jarecki v. G. D. Searle & Co., 367 U. S. 303 . Because a typical meeting, call, or event is not of the same stripe as a lawsuit before a court, a determination before an agency, or a hearing before a committee, it does not count as a “question” or “matter” under §201(a)(3). That more limited reading also comports with the presumption “that statutory language is not superfluous.” Arlington Central School Dist. Bd. of Ed. v. Murphy, 548 U. S. 291 , n. 1. Pp. 14–16. (2) Because a typical meeting, call, or event is not itself a question or matter, the next step is to determine whether arranging a meeting, contacting another official, or hosting an event may qualify as a “decision or action” on a different question or matter. That first requires the Court to establish what counts as a question or matter in this case. Section 201(a)(3) states that the question or matter must be “pending” or “may by law be brought” before “any public official.” “Pending” and “may by law be brought” suggest something that is relatively circumscribed—the kind of thing that can be put on an agenda, tracked for progress, and then checked off as complete. “May by law be brought” conveys something within the specific duties of an official’s position. Although the District Court determined that the relevant matter in this case could be considered at a much higher level of generality as “Virginia business and economic development,” Supp. App. 88, the pertinent matter must instead be more focused and concrete. The Fourth Circuit identified at least three such questions or matters: (1) whether researchers at Virginia’s state universities would initiate a study of Anatabloc; (2) whether Virginia’s Tobacco Commission would allocate grant money for studying anatabine; and (3) whether Virginia’s health plan for state employees would cover Anatabloc. The Court agrees that those qualify as questions or matters under §201(a)(3). Pp. 16–18. (3) The question remains whether merely setting up a meeting, hosting an event, or calling another official qualifies as a decision or action on any of those three questions or matters. It is apparent from United States v. Sun-Diamond Growers of Cal., 526 U. S. 398 , that the answer is no. Something more is required: §201(a)(3) specifies that the public official must make a decision or take an action on the question or matter, or agree to do so. For example, a decision or action to initiate a research study would qualify as an “official act.” A public official may also make a decision or take an action by using his official position to exert pressure on another official to perform an “official act,” or by using his official position to provide advice to another official, knowing or intending that such advice will form the basis for an “official act” by another official. A public official is not required to actually make a decision or take an action on a “question, matter, cause, suit, proceeding or controversy”; it is enough that he agree to do so. Setting up a meeting, hosting an event, or calling an official (or agreeing to do so) merely to talk about a research study or to gather additional information, however, does not qualify as a decision or action on the pending question of whether to initiate the study. Pp. 18–22. (c) The Government’s expansive interpretation of “official act” would raise significant constitutional concerns. Conscientious public officials arrange meetings for constituents, contact other officials on their behalf, and include them in events all the time. Representative government assumes that public officials will hear from their constituents and act appropriately on their concerns. The Government’s position could cast a pall of potential prosecution over these relationships. This concern is substantial, as recognized by White House counsel from every administration from that of President Reagan to President Obama, as well as two bipartisan groups of former state attorneys general. The Government’s interpretation also raises due process and federalism concerns. Pp. 22–24. 2. Given the Court’s interpretation of “official act,” the District Court’s jury instructions were erroneous, and the jury may have convicted Governor McDonnell for conduct that is not unlawful. Because the errors in the jury instructions are not harmless beyond a reasonable doubt, the Court vacates Governor McDonnell’s convictions. Pp. 24–28. (a) The jury instructions lacked important qualifications, rendering them significantly overinclusive. First, they did not adequately explain to the jury how to identify the pertinent “question, matter, cause, suit, proceeding or controversy.” It is possible the jury thought that a typical meeting, call, or event was itself a “question, matter, cause, suit, proceeding or controversy.” If so, the jury could have convicted Governor McDonnell without finding that he committed or agreed to commit an “official act,” as properly defined. Second, the instructions did not inform the jury that the “question, matter, cause, suit, proceeding or controversy” must be more specific and focused than a broad policy objective. As a result, the jury could have thought that the relevant “question, matter, cause, suit, proceeding or controversy” was something as nebulous as Virginia economic development, and convicted Governor McDonnell on that basis. Third, the District Court did not instruct the jury that to convict Governor McDonnell, it had to find that he made a decision or took an action—or agreed to do so—on the identified “question, matter, cause, suit, proceeding or controversy,” as properly defined. At trial, several of Governor McDonnell’s subordinates testified that he asked them to attend a meeting, not that he expected them to do anything other than that. If that testimony reflects what Governor McDonnell agreed to do at the time he accepted the loans and gifts from Williams, then he did not agree to make a decision or take an action on any of the three questions or matters described by the Fourth Circuit. Pp. 24–27. (b) Governor McDonnell raises two additional claims. First, he argues that the honest services statute and the Hobbs Act are unconstitutionally vague. The Court rejects that claim. For purposes of this case, the parties defined those statutes with reference to §201 of the federal bribery statute. Because the Court interprets the term “official act” in §201(a)(3) in a way that avoids the vagueness concerns raised by Governor McDonnell, it declines to invalidate those statutes under the facts here. Second, Governor McDonnell argues that there is insufficient evidence that he committed an “official act,” or agreed to do so. Because the parties have not had an opportunity to address that question in light of the Court’s interpretation of “official act,” the Court leaves it for the Court of Appeals to resolve in the first instance. Pp. 27–28. 792 F. 3d 478, vacated and remanded. Roberts, C. J., delivered the opinion for a unanimous Court.

In 2014, the Federal Government indicted former Virginia Governor Robert McDonnell and his wife, Maureen McDonnell, on bribery charges. The charges related to the acceptance by the McDonnells of $175,000 in loans, gifts, and other benefits from Virginia businessman Jonnie Williams, while Governor McDonnell was in office. Williams was the chief executive officer of Star Scientific, a Virginia-based company that had developed a nutritional supplement made from anatabine, a compound found in tobacco. Star Scientific hoped that Virginia’s public universities would perform research studies on anatabine, and Williams wanted Governor McDonnell’s assistance in obtaining those studies. To convict the McDonnells of bribery, the Government was required to show that Governor McDonnell committed (or agreed to commit) an “official act” in exchange for the loans and gifts. The parties did not agree, however, on what counts as an “official act.” The Government alleged in the indictment, and maintains on appeal, that Governor McDonnell committed at least five “official acts.” Those acts included “arranging meetings” for Williams with other Virginia officials to discuss Star Scientific’s product, “hosting” events for Star Scientific at the Governor’s Mansion, and “contacting other government officials” concerning studies of anatabine. Supp. App. 47–48. The Government also argued more broadly that these activities constituted “official action” because they related to Vir-ginia business development, a priority of Governor Mc-Donnell’s administration. Governor McDonnell contends that merely setting up a meeting, hosting an event, or contacting an official—without more—does not count as an “official act.” At trial, the District Court instructed the jury according to the Government’s broad understanding of what constitutes an “official act,” and the jury convicted both Governor and Mrs. McDonnell on the bribery charges. The Fourth Circuit affirmed Governor McDonnell’s conviction, and we granted review to clarify the meaning of “official act.” I A On November 3, 2009, petitioner Robert McDonnell was elected the 71st Governor of Virginia. His campaign slogan was “Bob’s for Jobs,” and his focus in office was on promoting business in Virginia. As Governor, McDonnell spoke about economic development in Virginia “on a daily basis” and attended numerous “events, ribbon cuttings,” and “plant facility openings.” App. 4093, 5241. He also referred thousands of constituents to meetings with members of his staff and other government officials. According to longtime staffers, Governor McDonnell likely had more events at the Virginia Governor’s Mansion to promote Virginia business than had occurred in “any other administration.” Id., at 4093. This case concerns Governor McDonnell’s interactions with one of his constituents, Virginia businessman Jonnie Williams. Williams was the CEO of Star Scientific, a Virginia-based company that developed and marketed Anatabloc, a nutritional supplement made from anatabine, a compound found in tobacco. Star Scientific hoped to obtain Food and Drug Administration approval of Anatabloc as an anti-inflammatory drug. An important step in securing that approval was initiating independent research studies on the health benefits of anatabine. Star Scientific hoped Virginia’s public universities would undertake such studies, pursuant to a grant from Virginia’s Tobacco Commission. Governor McDonnell first met Williams in 2009, when Williams offered McDonnell transportation on his private airplane to assist with McDonnell’s election campaign. Shortly after the election, Williams had dinner with Governor and Mrs. McDonnell at a restaurant in New York. The conversation turned to Mrs. McDonnell’s search for a dress for the inauguration, which led Williams to offer to purchase a gown for her. Governor McDonnell’s counsel later instructed Williams not to buy the dress, and Mrs. McDonnell told Williams that she would take a rain check. Id., at 2203–2209. In October 2010, Governor McDonnell and Williams met again on Williams’s plane. During the flight, Williams told Governor McDonnell that he “needed his help” moving forward on the research studies at Virginia’s public universities, and he asked to be introduced to the person that he “needed to talk to.” Id., at 2210–2211. Governor McDonnell agreed to introduce Williams to Dr. William Hazel, Virginia’s Secretary of Health and Human Resources. Williams met with Dr. Hazel the following month, but the meeting was unfruitful; Dr. Hazel was skeptical of the science behind Anatabloc and did not assist Williams in obtaining the studies. Id., at 2211–2217, 3738–3749. Six months later, Governor McDonnell’s wife, Maureen McDonnell, offered to seat Williams next to the Governor at a political rally. Shortly before the event, Williams took Mrs. McDonnell on a shopping trip and bought her $20,000 worth of designer clothing. The McDonnells later had Williams over for dinner at the Governor’s Mansion, where they discussed research studies on Anatabloc. Id., at 6560. Two days after that dinner, Williams had an article about Star Scientific’s research e-mailed to Mrs. McDonnell, which she forwarded to her husband. Less than an hour later, Governor McDonnell texted his sister to discuss the financial situation of certain rental properties they owned in Virginia Beach. Governor McDonnell also e-mailed his daughter to ask about expenses for her upcoming wedding. The next day, Williams returned to the Governor’s Mansion for a meeting with Mrs. McDonnell. At the meeting, Mrs. McDonnell described the family’s financial problems, including their struggling rental properties in Virginia Beach and their daughter’s wedding expenses. Mrs. McDonnell, who had experience selling nutritional supplements, told Williams that she had a background in the area and could help him with Anatabloc. According to Williams, she explained that the “Governor says it’s okay for me to help you and—but I need you to help me. I need you to help me with this financial situation.” Id., at 2231. Mrs. McDonnell then asked Williams for a $50,000 loan, in addition to a $15,000 gift to help pay for her daughter’s wedding, and Williams agreed. Williams testified that he called Governor McDonnell after the meeting and said, “I understand the financial problems and I’m willing to help. I just wanted to make sure that you knew about this.” Id., at 2233. According to Williams, Governor McDonnell thanked him for his help. Ibid. Governor McDonnell testified, in contrast, that he did not know about the loan at the time, and that when he learned of it he was upset that Mrs. McDonnell had requested the loan from Williams. Id., at 6095–6096. Three days after the meeting between Williams and Mrs. McDonnell, Governor McDonnell directed his assistant to forward the article on Star Scientific to Dr. Hazel. In June 2011, Williams sent Mrs. McDonnell’s chief of staff a letter containing a proposed research protocol for the Anatabloc studies. The letter was addressed to Governor McDonnell, and it suggested that the Governor “use the attached protocol to initiate the ‘Virginia Study’ of Anatabloc at the Medical College of Virginia and the University of Virginia School of Medicine.” Id., at 2254. Governor McDonnell gave the letter to Dr. Hazel. Id., at 6121–6122. Williams testified at trial that he did not “recall any response” to the letter. Id., at 2256. In July 2011, the McDonnell family visited Williams’s vacation home for the weekend, and Governor McDonnell borrowed Williams’s Ferrari while there. Shortly thereafter, Governor McDonnell asked Dr. Hazel to send an aide to a meeting with Williams and Mrs. McDonnell to discuss research studies on Anatabloc. The aide later testified that she did not feel pressured by Governor or Mrs. McDonnell to do “anything other than have the meeting,” and that Williams did not ask anything of her at the meeting. Id., at 3075. After the meeting, the aide sent Williams a “polite blow-off” e-mail. Id., at 3081. At a subsequent meeting at the Governor’s Mansion, Mrs. McDonnell admired Williams’s Rolex and mentioned that she wanted to get one for Governor McDonnell. Williams asked if Mrs. McDonnell wanted him to purchase a Rolex for the Governor, and Mrs. McDonnell responded, “Yes, that would be nice.” Id., at 2274. Williams did so, and Mrs. McDonnell later gave the Rolex to Governor McDonnell as a Christmas present. In August 2011, the McDonnells hosted a lunch event for Star Scientific at the Governor’s Mansion. According to Williams, the purpose of the event was to launch Anatabloc. See id., at 2278. According to Governor McDonnell’s gubernatorial counsel, however, it was just lunch. See id., at 3229–3231. The guest list for the event included researchers at the University of Virginia and Virginia Commonwealth University. During the event, Star Scientific distributed free samples of Anatabloc, in addition to eight $25,000 checks that researchers could use in preparing grant proposals for studying Anatabloc. Governor McDonnell asked researchers at the event whether they thought “there was some scientific validity” to Anatabloc and “whether or not there was any reason to explore this further.” Id., at 3344. He also asked whether this could “be something good for the Commonwealth, particularly as it relates to economy or job creation.” Ibid. When Williams asked Governor McDonnell whether he would support funding for the research studies, Governor McDonnell “very politely” replied, “I have limited decision-making power in this area.” Id., at 3927. In January 2012, Mrs. McDonnell asked Williams for an additional loan for the Virginia Beach rental properties, and Williams agreed. On February 3, Governor McDonnell followed up on that conversation by calling Williams to discuss a $50,000 loan. Several days later, Williams complained to Mrs. McDonnell that the Virginia universities were not returning Star Scientific’s calls. She passed Williams’s complaint on to the Governor. While Mrs. McDonnell was driving with Governor McDonnell, she also e-mailed Governor McDonnell’s counsel, stating that the Governor “wants to know why nothing has developed” with the research studies after Williams had provided the eight $25,000 checks for preparing grant proposals, and that the Governor “wants to get this going” at the universities. Id., at 3214, 4931. According to Governor McDonnell, how-ever, Mrs. McDonnell acted without his knowledge or per-mission, and he never made the statements she attributed to him. Id., at 6306–6308. On February 16, Governor McDonnell e-mailed Williams to check on the status of documents related to the $50,000 loan. A few minutes later, Governor McDonnell e-mailed his counsel stating, “Please see me about Anatabloc issues at VCU and UVA. Thanks.” Id., at 3217. Governor McDonnell’s counsel replied, “Will do. We need to be careful with this issue.” Ibid. The next day, Governor McDonnell’s counsel called Star Scientific’s lobbyist in order to “change the expectations” of Star Scientific regarding the involvement of the Governor’s Office in the studies. Id., at 3219. At the end of February, Governor McDonnell hosted a healthcare industry reception at the Governor’s Mansion, which Williams attended. Mrs. McDonnell also invited a number of guests recommended by Williams, including researchers at the Virginia universities. Governor McDonnell was present, but did not mention Star Scientific, Williams, or Anatabloc during the event. Id., at 3671–3672. That same day, Governor McDonnell and Williams spoke about the $50,000 loan, and Williams loaned the money to the McDonnells shortly thereafter. Id., at 2306, 2353. In March 2012, Governor McDonnell met with Lisa Hicks-Thomas, the Virginia Secretary of Administration, and Sara Wilson, the Director of the Virginia Department of Human Resource Management. The purpose of the meeting was to discuss Virginia’s health plan for state employees. At that time, Governor McDonnell was taking Anatabloc several times a day. He took a pill during the meeting, and told Hicks-Thomas and Wilson that the pills “were working well for him” and “would be good for” state employees. Id., at 4227. Hicks-Thomas recalled Governor McDonnell asking them to meet with a representative from Star Scientific; Wilson had no such recollection. Id., at 4219, 4227. After the discussion with Governor McDonnell, Hicks-Thomas and Wilson looked up Anatabloc on the Internet, but they did not set up a meeting with Star Scientific or conduct any other follow-up. Id., at 4220, 4230. It is undisputed that Virginia’s health plan for state employees does not cover nutritional supplements such as Anatabloc. In May 2012, Governor McDonnell requested an additional $20,000 loan, which Williams provided. Throughout this period, Williams also paid for several rounds of golf for Governor McDonnell and his children, took the McDonnells on a weekend trip, and gave $10,000 as a wedding gift to one of the McDonnells’ daughters. In total, Williams gave the McDonnells over $175,000 in gifts and loans. B In January 2014, Governor McDonnell was indicted for accepting payments, loans, gifts, and other things of value from Williams and Star Scientific in exchange for “performing official actions on an as-needed basis, as opportunities arose, to legitimize, promote, and obtain research studies for Star Scientific’s products.” Supp. App. 46. The charges against him comprised one count of conspiracy to commit honest services fraud, three counts of honest services fraud, one count of conspiracy to commit Hobbs Act extortion, six counts of Hobbs Act extortion, and two counts of making a false statement. See 18 U. S. C. §§1343, 1349 (honest services fraud); §1951(a) (Hobbs Act extortion); §1014 (false statement). Mrs. McDonnell was indicted on similar charges, plus obstructing official proceedings, based on her alleged involvement in the scheme. See §1512(c)(2) (obstruction). The theory underlying both the honest services fraud and Hobbs Act extortion charges was that Governor McDonnell had accepted bribes from Williams. See Skilling v. United States, 561 U. S. 358, 404 (2010) (construing honest services fraud to forbid “fraudulent schemes to deprive another of honest services through bribes or kickbacks”); Evans v. United States, 504 U. S. 255, 260, 269 (1992) (construing Hobbs Act extortion to include “ ‘taking a bribe’ ”). The parties agreed that they would define honest services fraud with reference to the federal bribery statute, 18 U. S. C. §201. That statute makes it a crime for “a public official or person selected to be a public official, directly or indirectly, corruptly” to demand, seek, receive, accept, or agree “to receive or accept anything of value” in return for being “influenced in the performance of any official act.” §201(b)(2). An “official act” is defined as “any decision or action on any question, matter, cause, suit, proceeding or controversy, which may at any time be pending, or which may by law be brought before any public official, in such official’s official capacity, or in such official’s place of trust or profit.” §201(a)(3). The parties also agreed that obtaining a “thing of value . . . knowing that the thing of value was given in return for official action” was an element of Hobbs Act extortion, and that they would use the definition of “official act” found in the federal bribery statute to define “official action” under the Hobbs Act. 792 F. 3d 478, 505 (CA4 2015) (internal quotation marks omitted). As a result of all this, the Government was required to prove that Governor McDonnell committed or agreed to commit an “official act” in exchange for the loans and gifts from Williams. See Evans, 504 U. S., at 268 (“the offense is completed at the time when the public official receives a payment in return for his agreement to perform specific official acts; fulfillment of the quid pro quo is not an element of the offense”). The Government alleged that Governor McDonnell had committed at least five “official acts”: (1) “arranging meetings for [Williams] with Virginia government officials, who were subordinates of the Governor, to discuss and promote Anatabloc”; (2) “hosting, and . . . attending, events at the Governor’s Mansion designed to encourage Virginia university researchers to initiate studies of anatabine and to promote Star Scientific’s products to doctors for referral to their patients”; (3) “contacting other government officials in the [Governor’s Office] as part of an effort to encourage Vir-ginia state research universities to initiate studies of anatabine”; (4) “promoting Star Scientific’s products and facilitating its relationships with Virginia government officials by allowing [Williams] to invite individuals important to Star Scientific’s business to exclusive events at the Governor’s Mansion”; and (5) “recommending that senior government officials in the [Governor’s Office] meet with Star Scientific executives to discuss ways that the company’s products could lower healthcare costs.” Supp. App. 47–48(indictment). The case proceeded to a jury trial, which lasted five weeks. Pursuant to an immunity agreement, Williams testified that he had given the gifts and loans to the McDonnells to obtain the Governor’s “help with the testing” of Anatabloc at Virginia’s medical schools. App. 2234. Governor McDonnell acknowledged that he had requested loans and accepted gifts from Williams. He testified, however, that setting up meetings with government officials was something he did “literally thousands of times” as Governor, and that he did not expect his staff “to do anything other than to meet” with Williams. Id., at 6042. Several state officials testified that they had discussed Anatabloc with Williams or Governor McDonnell, but had not taken any action to further the research studies. Id., at 3739–3750 (Dr. Hazel), 3075–3077 (aide to Dr. Hazel), 4218–4220 (Sara Wilson), 4230–4231 (Lisa Hicks-Thomas). A UVA employee in the university research office, who had never spoken with the Governor about Anatabloc, testified that she wrote a pro/con list concerning research studies on Anatabloc. The first “pro” was the “[p]erception to Governor that UVA would like to work with local companies,” and the first “con” was the “[p]olitical pressure from Governor and impact on future UVA requests from the Governor.” Id., at 4321, 4323 (Sharon Krueger). Following closing arguments, the District Court instructed the jury that to convict Governor McDonnell it must find that he agreed “to accept a thing of value in exchange for official action.” Supp. App. 68. The court described the five alleged “official acts” set forth in the indictment, which involved arranging meetings, hosting events, and contacting other government officials. The court then quoted the statutory definition of “official act,” and—as the Government had requested—advised the jury that the term encompassed “acts that a public official customarily performs,” including acts “in furtherance of longer-term goals” or “in a series of steps to exercise influence or achieve an end.” Id., at 69–70. Governor McDonnell had requested the court to further instruct the jury that the “fact that an activity is a routine activity, or a ‘settled practice,’ of an office-holder does not alone make it an ‘official act,’ ” and that “merely arranging a meeting, attending an event, hosting a reception, or making a speech are not, standing alone, ‘official acts,’ even if they are settled practices of the official,” because they “are not decisions on matters pending before the government.” 792 F. 3d, at 513 (internal quotation marks omitted). He also asked the court to explain to the jury that an “official act” must intend to or “in fact influence a specific official decision the government actually makes—such as awarding a contract, hiring a government em-ployee, issuing a license, passing a law, or implementing a regulation.” App. to Pet. for Cert. 147a. The District Court declined to give Governor McDonnell’s proposed instruction to the jury. The jury convicted Governor McDonnell on the honest services fraud and Hobbs Act extortion charges, but acquitted him on the false statement charges. Mrs. McDonnell was also convicted on most of the charges against her. Although the Government requested a sentence of at least ten years for Governor McDonnell, the District Court sentenced him to two years in prison. Mrs. McDonnell received a one-year sentence. Following the verdict, Governor McDonnell moved to vacate his convictions on the ground that the jury instructions “were legally erroneous because they (i) allowed the jury to convict [him] on an erroneous understanding of ‘official act,’ and (ii) allowed a conviction on the theory that [he] accepted things of value that were given for future unspecified action.” 64 F. Supp. 3d 783, 787 (ED Va. 2014). The District Court denied the motion. Id., at 802. In addition, Governor McDonnell moved for acquittal on the basis that there was insufficient evidence to convict him, and that the Hobbs Act and honest services statute were unconstitutionally vague. Crim. No. 3:14–CR–12 (ED Va., Dec. 1, 2014), Supp. App. 80, 82–92. That motion was also denied. See id., at 92–94. (He also raised other challenges to his convictions, which are not at issue here.) Governor McDonnell appealed his convictions to the Fourth Circuit, challenging the definition of “official action” in the jury instructions on the ground that it deemed “virtually all of a public servant’s activities ‘official,’ no matter how minor or innocuous.” 792 F. 3d, at 506. He also reiterated his challenges to the sufficiency of the evidence and the constitutionality of the statutes under which he was convicted. Id., at 509, n. 19, 515. The Fourth Circuit affirmed, and we granted certiorari. 577 U. S. ___ (2016). Mrs. McDonnell’s separate appeal remains pending before the Court of Appeals. II The issue in this case is the proper interpretation of the term “official act.” Section 201(a)(3) defines an “official act” as “any decision or action on any question, matter, cause, suit, proceeding or controversy, which may at any time be pending, or which may by law be brought before any public official, in such official’s official capacity, or in such official’s place of trust or profit.” According to the Government, “Congress used intentionally broad language” in §201(a)(3) to embrace “any decision or action, on any question or matter, that may at any time be pending, or which may by law be brought before any public official, in such official’s official capac-ity.” Brief for United States 20–21 (Government’s emphasis; alteration and internal quotation marks omitted). The Government concludes that the term “official act” therefore encompasses nearly any activity by a public official. In the Government’s view, “official act” specifically includes arranging a meeting, contacting another public official, or hosting an event—without more—concerning any subject, including a broad policy issue such as Vir-ginia economic development. Id., at 47–49; Tr. of Oral Arg. 28–30. Governor McDonnell, in contrast, contends that statu-tory context compels a more circumscribed reading, limiting “official acts” to those acts that “direct[ ] a particular resolution of a specific governmental decision,” or that pressure another official to do so. Brief for Petitioner 44, 51. He also claims that “vague corruption laws” such as §201 implicate serious constitutional concerns, militating “in favor of a narrow, cautious reading of these criminal statutes.” Id., at 21. Taking into account the text of the statute, the precedent of this Court, and the constitutional concerns raised by Governor McDonnell, we reject the Government’s reading of §201(a)(3) and adopt a more bounded interpretation of “official act.” Under that interpretation, setting up a meeting, calling another public official, or hosting an event does not, standing alone, qualify as an “official act.” A The text of §201(a)(3) sets forth two requirements for an “official act”: First, the Government must identify a “question, matter, cause, suit, proceeding or controversy” that “may at any time be pending” or “may by law be brought” before a public official. Second, the Government must prove that the public official made a decision or took an action “on” that question, matter, cause, suit, proceeding, or controversy, or agreed to do so. The issue here is whether arranging a meeting, contacting another official, or hosting an event—without more—can be a “question, matter, cause, suit, proceeding or controversy,” and if not, whether it can be a decision or action on a “question, matter, cause, suit, proceeding or controversy.” The first inquiry is whether a typical meeting, call, or event is itself a “question, matter, cause, suit, proceeding or controversy.” The Government argues that nearly any activity by a public official qualifies as a question or matter—from workaday functions, such as the typical call, meeting, or event, to the broadest issues the government confronts, such as fostering economic development. We conclude, however, that the terms “question, matter, cause, suit, proceeding or controversy” do not sweep so broadly. The last four words in that list—“cause,” “suit,” “proceeding,” and “controversy”—connote a formal exercise of governmental power, such as a lawsuit, hearing, or administrative determination. See, e.g., Crimes Act of 1790, §21, 1Stat. 117 (using “cause,” “suit,” and “controversy” in a related statutory context to refer to judicial proceedings); Black’s Law Dictionary 278–279, 400, 1602–1603 (4th ed. 1951) (defining “cause,” “suit,” and “controversy” as judicial proceedings); 18 U. S. C. §201(b)(3) (using “proceeding” to refer to trials, hearings, or the like “before any court, any committee of either House or both Houses of Congress, or any agency, commission, or officer”). Al-though it may be difficult to define the precise reach of those terms, it seems clear that a typical meeting, telephone call, or event arranged by a public official does not qualify as a “cause, suit, proceeding or controversy.” But what about a “question” or “matter”? A “question” could mean any “subject or aspect that is in dispute, open for discussion, or to be inquired into,” and a “matter” any “subject” of “interest or relevance.” Webster’s Third New International Dictionary 1394, 1863 (1961). If those meanings were adopted, a typical meeting, call, or event would qualify as a “question” or “matter.” A “question” may also be interpreted more narrowly, however, as “a subject or point of debate or a proposition being or to be voted on in a meeting,” such as a question “before the senate.” Id., at 1863. Similarly, a “matter” may be limited to “a topic under active and usually serious or practical consideration,” such as a matter that “will come before the committee.” Id., at 1394. To choose between those competing definitions, we look to the context in which the words appear. Under the familiar interpretive canon noscitur a sociis, “a word is known by the company it keeps.” Jarecki v. G. D. Searle & Co., 367 U. S. 303, 307 (1961) . While “not an inescapable rule,” this canon “is often wisely applied where a word is capable of many meanings in order to avoid the giving of unintended breadth to the Acts of Congress.” Ibid. For example, in Gustafson v. Alloyd Co., 513 U. S. 561 (1995) , a statute defined the word “prospectus” as a “prospectus, notice, circular, advertisement, letter, or communication.” Id., at 573–574 (internal quotation marks omitted). We held that although the word “communication” could in the abstract mean any type of communication, “it is apparent that the list refers to documents of wide dissemination,” and that inclusion “of the term ‘communication’ in that list suggests that it too refers to a public communication.” Id., at 575. Applying that same approach here, we conclude that a “question” or “matter” must be similar in nature to a “cause, suit, proceeding or controversy.” Because a typical meeting, call, or event arranged by a public official is not of the same stripe as a lawsuit before a court, a determination before an agency, or a hearing before a committee, it does not qualify as a “question” or “matter” under §201(a)(3). That more limited reading also comports with the presumption “that statutory language is not superfluous.” Arlington Central School Dist. Bd. of Ed. v. Murphy, 548 U. S. 291, 299, n. 1 (2006) . If “question” and “matter” were as unlimited in scope as the Government argues, the terms “cause, suit, proceeding or controversy” would serve no role in the statute—every “cause, suit, proceeding or controversy” would also be a “question” or “matter.” Under a more confined interpretation, however, “question” and “matter” may be understood to refer to a formal exercise of governmental power that is similar in nature to a “cause, suit, proceeding or controversy,” but that does not necessarily fall into one of those prescribed categories. Because a typical meeting, call, or event is not itself a question or matter, the next step is to determine whether arranging a meeting, contacting another official, or hosting an event may qualify as a “decision or action” on a different question or matter. That requires us to first establish what counts as a question or matter in this case. In addition to the requirements we have described, §201(a)(3) states that the question or matter must be “pending” or “may by law be brought” before “any public official.” “Pending” and “may by law be brought” suggest something that is relatively circumscribed—the kind of thing that can be put on an agenda, tracked for progress, and then checked off as complete. In particular, “may by law be brought” conveys something within the specific duties of an official’s position—the function conferred by the authority of his office. The word “any” conveys that the matter may be pending either before the public official who is performing the official act, or before another public official. The District Court, however, determined that the relevant matter in this case could be considered at a much higher level of generality as “Virginia business and economic development,” or—as it was often put to the jury—“Bob’s for Jobs.” Supp. App. 88; see, e.g., App. 1775, 2858, 2912, 3733. Economic development is not naturally described as a matter “pending” before a public official—or something that may be brought “by law” before him—any more than “justice” is pending or may be brought by law before a judge, or “national security” is pending or may be brought by law before an officer of the Armed Forces. Under §201(a)(3), the pertinent “question, matter, cause, suit, proceeding or controversy” must be more focused and concrete. For its part, the Fourth Circuit found at least three questions or matters at issue in this case: (1) “whether researchers at any of Virginia’s state universities would initiate a study of Anatabloc”; (2) “whether the state-created Tobacco Indemnification and Community Revitalization Commission” would “allocate grant money for the study of anatabine”; and (3) “whether the health insurance plan for state employees in Virginia would include Anatabloc as a covered drug.” 792 F. 3d, at 515–516. We agree that those qualify as questions or matters under §201(a)(3). Each is focused and concrete, and each involves a formal exercise of governmental power that is similar in nature to a lawsuit, administrative determination, or hearing. The question remains whether—as the Government argues—merely setting up a meeting, hosting an event, or calling another official qualifies as a decision or action on any of those three questions or matters. Although the word “decision,” and especially the word “action,” could be read expansively to support the Government’s view, our opinion in United States v. Sun-Diamond Growers of Cal., 526 U. S. 398 (1999) , rejects that interpretation. In Sun-Diamond, the Court stated that it was not an “official act” under §201 for the President to host a championship sports team at the White House, the Secretary of Education to visit a high school, or the Secretary of Agriculture to deliver a speech to “farmers concerning various matters of USDA policy.” Id., at 407. We recognized that “the Secretary of Agriculture always has before him or in prospect matters that affect farmers, just as the President always has before him or in prospect matters that affect college and professional sports, and the Secretary of Education matters that affect high schools.” Ibid. But we concluded that the existence of such pending matters was not enough to find that any action related to them constituted an “official act.” Ibid. It was possible to avoid the “absurdities” of convicting individuals on corruption charges for engaging in such conduct, we explained, “through the definition of that term,” i.e., by adopting a more limited definition of “official acts.” Id., at 408. It is apparent from Sun-Diamond that hosting an event, meeting with other officials, or speaking with interested parties is not, standing alone, a “decision or action” within the meaning of §201(a)(3), even if the event, meeting, or speech is related to a pending question or matter. Instead, something more is required: §201(a)(3) specifies that the public official must make a decision or take an action on that question or matter, or agree to do so. For example, a decision or action to initiate a research study—or a decision or action on a qualifying step, such as narrowing down the list of potential research topics—would qualify as an “official act.” A public official may also make a decision or take an action on a “question, matter, cause, suit, proceeding or controversy” by using his official position to exert pressure on another official to perform an “official act.” In addition, if a public official uses his official position to provide advice to another official, knowing or intending that such advice will form the basis for an “official act” by another official, that too can qualify as a decision or action for purposes of §201(a)(3). See United States v. Birdsall, 233 U. S. 223, 234 (1914) (finding “official action” on the part of subordinates where their superiors “would necessarily rely largely upon the reports and advice of subordinates . . . who were more directly acquainted with” the “facts and circumstances of particular cases”). Under this Court’s precedents, a public official is not required to actually make a decision or take an action on a “question, matter, cause, suit, proceeding or controversy”; it is enough that the official agree to do so. See Evans, 504 U. S., at 268. The agreement need not be explicit, and the public official need not specify the means that he will use to perform his end of the bargain. Nor must the public official in fact intend to perform the “official act,” so long as he agrees to do so. A jury could, for example, conclude that an agreement was reached if the evidence shows that the public official received a thing of value knowing that it was given with the expectation that the official would perform an “official act” in return. See ibid. It is up to the jury, under the facts of the case, to determine whether the public official agreed to perform an “official act” at the time of the alleged quid pro quo. The jury may consider a broad range of pertinent evidence, including the nature of the transaction, to answer that question. Setting up a meeting, hosting an event, or calling an official (or agreeing to do so) merely to talk about a research study or to gather additional information, however, does not qualify as a decision or action on the pending question of whether to initiate the study. Simply expressing support for the research study at a meeting, event, or call—or sending a subordinate to such a meeting, event, or call—similarly does not qualify as a decision or action on the study, as long as the public official does not intend to exert pressure on another official or provide advice, knowing or intending such advice to form the basis for an “official act.” Otherwise, if every action somehow related to the research study were an “official act,” the requirement that the public official make a decision or take an action on that study, or agree to do so, would be meaningless. Of course, this is not to say that setting up a meeting, hosting an event, or making a phone call is always an innocent act, or is irrelevant, in cases like this one. If an official sets up a meeting, hosts an event, or makes a phone call on a question or matter that is or could be pending before another official, that could serve as evidence of an agreement to take an official act. A jury could conclude, for example, that the official was attempting to pressure or advise another official on a pending matter. And if the official agreed to exert that pressure or give that advice in exchange for a thing of value, that would be illegal. The Government relies on this Court’s decision in Birdsall to support a more expansive interpretation of “official act,” but Birdsall is fully consistent with our reading of §201(a)(3). We held in Birdsall that “official action” could be established by custom rather than “by statute” or “a written rule or regulation,” and need not be a formal part of an official’s decisionmaking process. 233 U. S., at 230–231. That does not mean, however, that every decision or action customarily performed by a public official—such as the myriad decisions to refer a constituent to another official—counts as an “official act.” The “official action” at issue in Birdsall was “advis[ing] the Commissioner of Indian Affairs, contrary to the truth,” that the facts of the case warranted granting leniency to certain defendants convicted of “unlawfully selling liquor to Indians.” Id., at 227–230. That “decision or action” fits neatly within our understanding of §201(a)(3): It reflected a decision or action to advise another official on the pending question whether to grant leniency. In sum, an “official act” is a decision or action on a “question, matter, cause, suit, proceeding or controversy.” The “question, matter, cause, suit, proceeding or controversy” must involve a formal exercise of governmental power that is similar in nature to a lawsuit before a court, a determination before an agency, or a hearing before a committee. It must also be something specific and focused that is “pending” or “may by law be brought” before a public official. To qualify as an “official act,” the public official must make a decision or take an action on that “question, matter, cause, suit, proceeding or controversy,” or agree to do so. That decision or action may include using his official position to exert pressure on another official to perform an “official act,” or to advise another official, knowing or intending that such advice will form the basis for an “official act” by another official. Setting up a meeting, talking to another official, or organizing an event (or agreeing to do so)—without more—does not fit that definition of “official act.” B In addition to being inconsistent with both text and precedent, the Government’s expansive interpretation of “official act” would raise significant constitutional concerns. Section 201 prohibits quid pro quo corruption—the exchange of a thing of value for an “official act.” In the Government’s view, nearly anything a public official accepts—from a campaign contribution to lunch—counts as a quid; and nearly anything a public official does—from arranging a meeting to inviting a guest to an event—counts as a quo. See Brief for United States 14, 27; Tr. of Oral Arg. 34–35, 44–46. But conscientious public officials arrange meetings for constituents, contact other officials on their behalf, and include them in events all the time. The basic compact underlying representative government assumes that public officials will hear from their constituents and act appropriately on their concerns—whether it is the union official worried about a plant closing or the homeowners who wonder why it took five days to restore power to their neighborhood after a storm. The Government’s position could cast a pall of potential prosecution over these relationships if the union had given a campaign contribution in the past or the homeowners invited the official to join them on their annual outing to the ballgame. Officials might wonder whether they could respond to even the most commonplace requests for assistance, and citizens with legitimate concerns might shrink from participating in democratic discourse. This concern is substantial. White House counsel who worked in every administration from that of President Reagan to President Obama warn that the Government’s “breathtaking expansion of public-corruption law would likely chill federal officials’ interactions with the people they serve and thus damage their ability effectively to perform their duties.” Brief for Former Federal Officials as Amici Curiae 6. Six former Virginia attorneys general—four Democrats and two Republicans—also filed an amicus brief in this Court echoing those concerns, as did 77 former state attorneys general from States other than Virginia—41 Democrats, 35 Republicans, and 1 independent. Brief for Former Virginia Attorneys General as Amici Curiae 1–2, 16; Brief for 77 Former State Attorneys General (Non-Virginia) as Amici Curiae 1–2. None of this, of course, is to suggest that the facts of this case typify normal political interaction between public officials and their constituents. Far from it. But the Government’s legal interpretation is not confined to cases involving extravagant gifts or large sums of money, and we cannotconstrue a criminal statute on the assumption that the Government will “use it responsibly.” United States v. Stevens, 559 U. S. 460, 480 (2010) . The Court in Sun-Diamond declined to rely on “the Government’s discretion” to protect against overzealous prosecutions under §201, concluding instead that “a statute in this field that can linguistically be interpreted to be either a meat axe or a scalpel should reasonably be taken to be the latter.” 526 U. S., at 408, 412. A related concern is that, under the Government’s interpretation, the term “official act” is not defined “with sufficient definiteness that ordinary people can understand what conduct is prohibited,” or “in a manner that does not encourage arbitrary and discriminatory enforcement.” Skilling, 561 U. S., at 402–403 (internal quotation marks omitted). Under the “ ‘standardless sweep’ ” of the Government’s reading, Kolender v. Lawson, 461 U. S. 352, 358 (1983) , public officials could be subject to prosecution, without fair notice, for the most prosaic interactions. “Invoking so shapeless a provision to condemn someone to prison” for up to 15 years raises the serious concern that the provision “does not comport with the Constitution’s guarantee of due process.” Johnson v. United States, 576 U. S. ___, ___ (2015) (slip op., at 10). Our more constrained interpretation of §201(a)(3) avoids this “vagueness shoal.” Skilling, 561 U. S., at 368. The Government’s position also raises significant federalism concerns. A State defines itself as a sovereign through “the structure of its government, and the character of those who exercise government authority.” Gregory v. Ashcroft, 501 U. S. 452, 460 (1991) . That includes the prerogative to regulate the permissible scope of interactions between state officials and their constituents. Here, where a more limited interpretation of “official act” is supported by both text and precedent, we decline to “construe the statute in a manner that leaves its outer boundaries ambiguous and involves the Federal Government in setting standards” of “good government for local and state officials.” McNally v. United States, 483 U. S. 350, 360 (1987) ; see also United States v. Enmons, 410 U. S. 396 –411 (1973) (rejecting a “broad concept of extortion” that would lead to “an unprecedented incursion into the criminal jurisdiction of the States”). III A Governor McDonnell argues that his convictions must be vacated because the jury was improperly instructed on the meaning of “official act” under §201(a)(3) of the federal bribery statute. According to Governor McDonnell, the District Court “refused to convey any meaningful limits on ‘official act,’ giving an instruction that allowed the jury to convict [him] for lawful conduct.” Brief for Petitioner 51. We agree. The jury instructions included the statutory definition of “official action,” and further defined the term to include “actions that have been clearly established by settled practice as part of a public official’s position, even if the action was not taken pursuant to responsibilities explicitly assigned by law.” Supp. App. 69–70. The instructions also stated that “official actions may include acts that a public official customarily performs,” including acts “in furtherance of longer-term goals” or “in a series of steps to exercise influence or achieve an end.” Id., at 70. In light of our interpretation of the term “official acts,” those instructions lacked important qualifications, rendering them significantly overinclusive. First, the instructions did not adequately explain to the jury how to identify the “question, matter, cause, suit, proceeding or controversy.” As noted, the Fourth Circuit held that “the Government presented evidence of three questions or matters”: (1) “whether researchers at any of Virginia’s state universities would initiate a study of Anatabloc”; (2) “whether the state-created Tobacco Indemnification and Community Revitalization Commission” would “allocate grant money for the study of anatabine”; and (3) “whether the health insurance plan for state employees in Virginia would include Anatabloc as a covered drug.” 792 F. 3d, at 515–516. The problem with the District Court’s instructions is that they provided no assurance that the jury reached its verdict after finding those questions or matters. The testimony at trial described how Governor McDonnell set up meetings, contacted other officials, and hosted events. It is possible the jury thought that a typical meeting, call, or event was itself a “question, matter, cause, suit, proceeding or controversy.” If so, the jury could have convicted Governor McDonnell without finding that he committedor agreed to commit an “official act,” as properly defined. To prevent this problem, the District Court should have instructed the jury that it must identify a “question, matter, cause, suit, proceeding or controversy” involving the formal exercise of governmental power. Second, the instructions did not inform the jury that the “question, matter, cause, suit, proceeding or controversy” must be more specific and focused than a broad policy objective. The Government told the jury in its closing argument that “[w]hatever it was” Governor McDonnell had done, “it’s all official action.” App. to Pet. for Cert. 263a–264a. Based on that remark, and the repeated references to “Bob’s for Jobs” at trial, the jury could have thought that the relevant “question, matter, cause, suit, proceeding or controversy” was something as nebulous as “Virginia business and economic development,” as the District Court itself concluded. Supp. App. 87–88 (“The alleged official actions in this case were within the range of actions on questions, matters, or causes pending before McDonnell as Governor as multiple witnesses testified that Virginia business and economic development was a top priority in McDonnell’s administration”). To avoid that misconception, the District Court should have instructed the jury that the pertinent “question, matter, cause, suit, proceeding or controversy” must be something specific and focused that is “pending” or “may by law be brought before any public official,” such as the question whether to initiate the research studies. Third, the District Court did not instruct the jury that to convict Governor McDonnell, it had to find that he made a decision or took an action—or agreed to do so—on the identified “question, matter, cause, suit, proceeding or controversy,” as we have construed that requirement. At trial, several of Governor McDonnell’s subordinates testified that he asked them to attend a meeting, not that he expected them to do anything other than that. See, e.g., App. 3075, 3739–3740, 4220. If that testimony reflects what Governor McDonnell agreed to do at the time he accepted the loans and gifts from Williams, then he did not agree to make a decision or take an action on any of the three questions or matters described by the Fourth Circuit. The jury may have disbelieved that testimony or found other evidence that Governor McDonnell agreed to exert pressure on those officials to initiate the research studies or add Anatabloc to the state health plan, but it is also possible that the jury convicted Governor McDonnell without finding that he agreed to make a decision or take an action on a properly defined “question, matter, cause, suit, proceeding or controversy.” To forestall that possibility, the District Court should have instructed the jury that merely arranging a meeting or hosting an event to discuss a matter does not count as a decision or action on that matter. Because the jury was not correctly instructed on the meaning of “official act,” it may have convicted Governor McDonnell for conduct that is not unlawful. For that reason, we cannot conclude that the errors in the jury instructions were “harmless beyond a reasonable doubt.” Neder v. United States, 527 U. S. 1, 16 (1999) (internal quotation marks omitted). We accordingly vacate Governor McDonnell’s convictions. B Governor McDonnell raises two additional claims. First, he argues that the charges against him must be dismissed because the honest services statute and the Hobbs Act are unconstitutionally vague. See Brief for Petitioner 58–61. We reject that claim. For purposes of this case, the parties defined honest services fraud and Hobbs Act extortion with reference to §201 of the federal bribery statute. Because we have interpreted the term “official act” in §201(a)(3) in a way that avoids the vagueness concerns raised by Governor McDonnell, we decline to invalidate those statutes under the facts here. See Skilling, 561 U. S., at 403 (seeking “to construe, not condemn, Congress’ enactments”). Second, Governor McDonnell argues that the charges must be dismissed because there is insufficient evidence that he committed an “official act,” or that he agreed to do so. Brief for Petitioner 44–45. Because the parties have not had an opportunity to address that question in light of the interpretation of §201(a)(3) adopted by this Court, we leave it for the Court of Appeals to resolve in the first instance. If the court below determines that there is sufficient evidence for a jury to convict Governor McDonnell of committing or agreeing to commit an “official act,” his case may be set for a new trial. If the court instead determines that the evidence is insufficient, the charges against him must be dismissed. We express no view on that question. * * * There is no doubt that this case is distasteful; it may be worse than that. But our concern is not with tawdry tales of Ferraris, Rolexes, and ball gowns. It is instead with the broader legal implications of the Government’s boundless interpretation of the federal bribery statute. A more limited interpretation of the term “official act” leaves ample room for prosecuting corruption, while comporting with the text of the statute and the precedent of this Court. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

577.US.2015_14-510

Pursuant to the Indian Self-Determination and Education Assistance Act (ISDA), petitioner Menominee Indian Tribe of Wisconsin contracted with the Indian Health Service (IHS) to operate what would otherwise have been a federal program and to receive an amount of money equal to what the Government would have spent on operating the program itself, including reimbursement for reasonable contract support costs. 25 U. S. C. §§450f, 450j–1(a). After other tribal entities successfully litigated complaints against the Federal Government for failing to honor its obligation to pay contract support costs, the Menominee Tribe presented its own contract support claims to the IHS in accordance with the Contract Disputes Act of 1978 (CDA), which requires contractors to present each claim to a contracting officer for decision, 41 U. S. C. §7103(a)(1). The contracting officer denied some of the Tribe’s claims because they were not presented within the CDA’s 6-year limitations period. See §7103(a)(4)(A). The Tribe challenged the denials in Federal District Court, arguing that the limitations period should be tolled for the nearly two years in which a putative class action, brought by tribes with parallel complaints, was pending. As relevant here, the District Court eventually denied the Tribe’s equitable-tolling claim, and the Court of Appeals affirmed, holding that no extraordinary circumstances beyond the Tribe’s control caused the delay. Held: Equitable tolling does not apply to the presentment of petitioner’s claims. Pp. 5–9. (a) To be entitled to equitable tolling of a statute of limitations, a litigant must establish “(1) that he has been pursuing his rights diligently, and (2) that some extraordinary circumstance stood in his way and prevented timely filing.” Holland v. Florida, 560 U. S. 631 . The Tribe argues that diligence and extraordinary circumstances should be considered together as factors in a unitary test, and it faults the Court of Appeals for declining to consider the Tribe’s diligence in connection with its finding that no extraordinary circumstances existed. But this Court has expressly characterized these two components as “elements,” not merely factors of indeterminate or commensurable weight, Pace v. DiGuglielmo, 544 U. S. 408 , and has treated them as such in practice, see Lawrence v. Florida, 549 U. S. 327 –337. The Tribe also objects to the Court of Appeals’ interpretation of the “extraordinary circumstances” prong as requiring the showing of an “external obstacle” to timely filing. This Court reaffirms that this prong is met only where the circumstances that caused a litigant’s delay are both extraordinary and beyond its control. Pp. 5–7. (b) None of the Tribe’s excuses satisfy the “extraordinary circumstances” prong of the test. The Tribe had unilateral authority to present its claims in a timely manner. Its claimed obstacles, namely, a mistaken reliance on a putative class action and a belief that presentment was futile, were not outside the Tribe’s control. And the significant risk and expense associated with presenting and litigating its claims are far from extraordinary. Finally, the special relationship between the United States and Indian tribes, as articulated in the ISDA, does not override clear statutory language. Pp. 7–8. 764 F. 3d 51, affirmed. Alito, J., delivered the opinion for a unanimous Court.

Petitioner Menominee Indian Tribe of Wisconsin (Tribe) seeks equitable tolling to preserve contract claims not timely presented to a federal contracting officer. Because the Tribe cannot establish extraordinary circumstances that stood in the way of timely filing, we hold that equit-able tolling does not apply. I Congress enacted the Indian Self-Determination and Education Assistance Act (ISDA), Pub. L. 93–638, 88Stat. 2203, 25 U. S. C. §450 et seq., in 1975 to help Indian tribes assume responsibility for aid programs that benefit their members. Under the ISDA, tribes may enter into “self-determination contracts” with federal agencies to take control of a variety of federally funded programs. §450f. A contracting tribe is eligible to receive the amount of money that the Government would have otherwise spent on the program, see §450j–1(a)(1), as well as reimbursement for reasonable “contract support costs,” which include administrative and overhead costs associated with carrying out the contracted programs, §§450j–1(a)(2), (3), (5). In 1988, Congress amended the ISDA to apply the Contract Disputes Act of 1978 (CDA), 41 U. S. C. §7101 et seq., to disputes arising under the ISDA. See 25 U. S. C. §450m–1(d); Indian Self-Determination and Education Assistance Act Amendments of 1988, §206(2), 102Stat. 2295. As part of its mandatory administrative process for resolving contract disputes, the CDA requires contractors to present “[e]ach claim” they may have to a contracting officer for decision. 41 U. S. C. §7103(a)(1). Congress later amended the CDA to include a 6-year statute of limitations for presentment of each claim. Federal Acquisition Streamlining Act of 1994, 41 U. S. C. §7103(a)(4)(A). Under the CDA, the contracting officer’s decision is generally final, unless challenged through one of the statutorily authorized routes. §7103(g). A contractor dissatisfied with the officer’s decision may either take an administrative appeal to a board of contract appeals or file an action for breach of contract in the United States Court of Federal Claims. §§7104(a), (b)(1), 7105(b). Both routes then lead to the United States Court of Appeals for the Federal Circuit for any further review. 28 U. S. C. §1295(a)(3); 41 U. S. C. §7107(a)(1); see 25 U. S. C. §450m–1(d). Under the ISDA, tribal contractors have a third option. They may file a claim for money damages in federal district court, §§450m–1(a), (d), and if they lose, they may pursue an appeal in one of the regional courts of appeals, 28 U. S. C. §1291. Tribal contractors have repeatedly complained that the Federal Government has not fully honored its obligations to pay contract support costs. Three lawsuits making such claims are relevant here. The first was a class action filed by the Ramah Navajo Chapter alleging that the Bureau of Indian Affairs (BIA) systematically underpaid certain contract support costs. Ramah Navajo Chapter v. Lujan, No. 1:90–cv–0957 (D NM) (filed Oct. 4, 1990). In 1993, Ramah successfully moved for certification of a nationwide class of all tribes that had contracted with the BIA under the ISDA. See Order and Memorandum Opinion in Ramah Navajo Chapter v. Lujan, No. 1:90–cv–0957 (D NM, Oct. 1, 1993), App. 35–40. The Government argued that each tribe needed to present its claims to a contracting officer before it could participate in the class. Id., at 37–38. But the trial court held that tribal contractors could participate in the class without presentment, because the suit alleged systemwide flaws in the BIA’s contracting scheme, not merely breaches of individual contracts. Id., at 39. The Government did not appeal the certification order, and the Ramah class action proceeded to further litigation and settlement. The second relevant ISDA suit raised similar claims about contract support costs but arose from contracts with the Indian Health Service (IHS). Cherokee Nation of Okla. v. United States, No. 6:99–cv–0092 (ED Okla.) (filed Mar. 5, 1999). In Cherokee Nation, two tribes filed a putative class action against IHS. On February 9, 2001, the District Court denied class certification without addressing whether tribes would need to present claims to join the class. Cherokee Nation of Okla. v. United States, 199 F. R. D. 357, 363–366 (ED Okla.). The two plaintiff tribes did not appeal the denial of class certification but proceeded to the merits on their own, eventually prevailing before this Court in a parallel suit. See Cherokee Nation of Okla. v. Leavitt, 543 U. S. 631 (2005) . The third relevant case is the one now before us. In this case, the Tribe presented its contract support claims (for contract years 1995 through 2004) to IHS on September 7, 2005, shortly after our Cherokee Nation ruling. As relevant here, the contracting officer denied the Tribe’s claims based on its 1996, 1997, and 1998 contracts because, inter alia, those claims were barred by the CDA’s 6-year statute of limitations.[1] The Tribe challenged the denials in the United States District Court for the District of Columbia, arguing, based on theories of class-action and equitable tolling, that the limitations period should be tolled for the 707 days that the putative Cherokee Nation class had been pending. See American Pipe & Constr. Co. v. Utah, 414 U. S. 538 (1974) (class-action tolling); Holland v. Florida, 560 U. S. 631 (2010) (equitable tolling). Initially, the District Court held that the limitations period was jurisdictional and thus forbade tolling of any sort. 539 F. Supp. 2d 152, 154, and n. 2 (DDC 2008). On appeal, the United States Court of Appeals for the District of Columbia Circuit concluded that the limitations period was not jurisdictional and thus did not necessarily bar tolling. 614 F. 3d 519, 526 (2010). But the court held that the Tribe was ineligible for class-action tolling during the pendency of the putative Cherokee Nation class, because the Tribe’s failure to present its claims to IHS made it “ineligible to participate in the class action at the time class certification [was] denied.” 614 F. 3d, at 527 (applying American Pipe). The court then remanded the case to the District Court to determine the Tribe’s eligibility for equitable tolling. On remand, the District Court concluded that the Tribe’s asserted reasons for failing to present its claims within the specified time “do not, individually or collec-tively, amount to an extraordinary circumstance” that could warrant equitable tolling. 841 F. Supp. 2d 99, 107 (DC 2012) (internal quotation marks omitted). This time, the Court of Appeals affirmed. 764 F. 3d 51 (CADC 2014). It explained that, “[t]o count as sufficiently ‘extraordinary’ to support equitable tolling, the circumstances that caused a litigant’s delay must have been beyond its control,” and “cannot be a product of that litigant’s own misunderstanding of the law or tactical mistakes in litigation.” Id., at 58. Because none of the Tribe’s proffered circumstances was beyond its control, the court held, there were no extraordinary circumstances that could merit equitable tolling. The Court of Appeals’ decision created a split with the Federal Circuit, which granted another tribal entity equitable tolling under similar circumstances. See Arctic Slope Native Assn., Ltd. v. Sebelius, 699 F. 3d 1289 (CA Fed. 2012). We granted certiorari to resolve the conflict. 576 U. S. ___ (2015). II The Court of Appeals denied the Tribe’s request for equitable tolling by applying the test that we articulated in Holland v. Florida, 560 U. S. 631 . Under Holland, a litigant is entitled to equitable tolling of a statute of limitations only if the litigant establishes two elements: “(1) that he has been pursuing his rights diligently, and (2) that some extraordinary circumstance stood in his way and prevented timely filing.” Id., at 649 (internal quotation marks omitted). The Tribe calls this formulation of the equitable tolling test overly rigid, given the doctrine’s equitable nature. First, it argues that diligence and extraordinary circumstances should be considered together as two factors in a unitary test, and it faults the Court of Appeals for declining to consider the Tribe’s diligence in connection with its finding that no extraordinary circumstances existed. But we have expressly characterized equitable tolling’s two components as “elements,” not merely factors of indeterminate or commensurable weight. Pace v. DiGuglielmo, 544 U. S. 408, 418 (2005) (“Generally, a litigant seeking equitable tolling bears the burden of establishing two elements”). And we have treated the two requirements as distinct elements in practice, too, rejecting requests for equitable tolling where a litigant failed to satisfy one without addressing whether he satisfied the other. See, e.g., Lawrence v. Florida, 549 U. S. 327 –337 (2007) (rejecting equitable tolling without addressing diligence because habeas petitioner fell “far short of showing ‘extraordinary circumstances’ ”); Pace, supra, at 418 (holding, without resolving litigant’s argument that he had “satisfied the extraordinary circumstance test,” that, “[e]ven if we were to accept [his argument], he would not be entitled to relief because he has not established the requisite diligence”). Second, the Tribe objects to the Court of Appeals’ interpretation of the “extraordinary circumstances” prong as requiring a litigant seeking tolling to show an “external obstacl[e]” to timely filing, i.e., that “the circumstances that caused a litigant’s delay must have been beyond its control.” 764 F. 3d, at 58–59. The Tribe complains that this “external obstacle” formulation amounts to the same kind of “ ‘overly rigid per se approach’ ” we rejected in Holland. Brief for Petitioner 32 (quoting 560 U. S., at 653). But in truth, the phrase “external obstacle” merely reflects our requirement that a litigant seeking tolling show “that some extraordinary circumstance stood in his way.” Id., at 649 (emphasis added; internal quotation marks omitted). This phrasing in Holland (and in Pace before that) would make little sense if equitable tolling were available when a litigant was responsible for its own delay. Indeed, the diligence prong already covers those affairs within the litigant’s control; the extraordinary-circumstances prong, by contrast, is meant to cover matters outside its control. We therefore reaffirm that the second prong of the equitable tolling test is met only where the circumstances that caused a litigant’s delay are both extraordinary and beyond its control.[2] III The Tribe offers no circumstances that meet this standard. Its mistaken reliance on the putative Cherokee Nation class action was not an obstacle beyond its control.[3] As the Tribe conceded below, see 614 F. 3d, at 526–527, it could not have been a member of the putative Cherokee Nation class because it did not present its claims to an IHS contracting officer before class certification was denied. Before then, the Tribe had unilateral authority to present its claims and to join the putative class. Presentment was blocked not by an obstacle outside its control, but by the Tribe’s mistaken belief that presentment was unneeded. The Tribe’s mistake, in essence, was its inference that the reasoning of the Ramah class certification decision (allowing tribes to participate—without presentment—in the class challenging underpayment of BIA contract support costs) applied to the putative Cherokee Nation class. This mistake was fundamentally no different from “a garden variety claim of excusable neglect,” Irwin v. Department of Veterans Affairs, 498 U. S. 89, 96 (1990) , “such as a simple ‘miscalculation’ that leads a lawyer to miss a filing deadline,” Holland, supra, at 651 (quoting Lawrence, supra, at 336). And it is quite different from relying on actually binding precedent that is subsequently reversed.[4] The Tribe’s other excuses are even less compelling. Its belief that presentment was futile was not an obstacle beyond its control but a species of the same mistake that kept it out of the putative Cherokee Nation class. And the fact that there may have been significant risk and expense associated with presenting and litigating its claims is far from extraordinary. As the District Court noted below, “it is common for a litigant to be confronted with significant costs to litigation, limited financial resources, an uncertain outcome based upon an uncertain legal landscape, and impending deadlines. These circumstances are not ‘extraordinary.’ ” 841 F. Supp. 2d, at 107. Finally, the Tribe also urges us to consider the special relationship between the United States and Indian tribes, as articulated in the ISDA. See 25 U. S. C. §450a(b) (“Congress declares its commitment to the maintenance of the Federal Government’s unique and continuing relationship with, and responsibility to, individual Indian tribes and to the Indian people as a whole”). We do not question the “general trust relationship between the United States and the Indian tribes,” but any specific obligations the Government may have under that relationship are “governed by statute rather than the common law.” United States v. Jicarilla Apache Nation, 564 U. S. 162, 165 (2011) . The ISDA and CDA establish a clear procedure for the resolution of disputes over ISDA contracts, with an unambiguous 6-year deadline for presentment of claims. The “general trust relationship” does not override the clear language of those statutes.[5] IV For these reasons, the judgment of the United States Court of Appeals for the District of Columbia Circuit is affirmed. It is so ordered.Notes 1 Because the contract claims accrued no later than the end of each calendar-year contract, the District Court determined, the statute of limitations for the 1996, 1997, and 1998 contracts had run by January 1st of the years 2003, 2004, and 2005, respectively. 539 F. Supp. 2d 152, 154, n. 1 (DC 2008). The Tribe does not dispute the timing of accrual before this Court. 2 Holland v. Florida, 560 U. S. 631 (2010) , is a habeas case, and we have never held that its equitable-tolling test necessarily applies outside the habeas context. Nevertheless, because we agree that the Tribe cannot meet Holland’s test, we have no occasion to decide whether an even stricter test might apply to a nonhabeas case. Nor does the Tribe argue that a more generous test than Holland’s should apply here. 3 Because we conclude that the Tribe’s mistake of law was not outside its control, we need not decide whether a mistake of law, however reasonable, could ever be extraordinary. 4 The Court of Appeals speculated, without deciding, that such a development might merit tolling, but like that court we have no occasion to decide the question. 5 Because we hold that there were no extraordinary circumstances, we need not decide whether the Tribe was diligently pursuing its rights. We also need not accept the Tribe’s invitation to assess prejudice to the Government, because the absence of prejudice to the opposing party “is not an independent basis for invoking the doctrine [of equitable tolling] and sanctioning deviations from established procedures.” Baldwin County Welcome Center v. Brown, 466 U. S. 147, 152 (1984) (per curiam). Rather, the absence of prejudice is “a factor to be considered in determining whether the doctrine of equitable tolling should apply once a factor that might justify such tolling is identified.” Ibid. (emphasis added).

578.US.2015_14-1132

Respondent Greg Manning held over two million shares of stock in Escala Group, Inc. He claims that he lost most of his investment when the share price plummeted after petitioners, Merrill Lynch and other financial institutions (collectively, Merrill Lynch), devalued Escala through “naked short sales” of its stock. Unlike a typical short sale, where a person borrows stock from a broker, sells it to a buyer on the open market, and later purchases the same number of shares to return to the broker, the seller in a “naked” short sale does not borrow the stock he puts on the market, and so never delivers the promised shares to the buyer. This practice, which can injure shareholders by driving down a stock’s price, is regulated by the Securities and Exchange Commission’s Regulation SHO, which prohibits short-sellers from intentionally failing to deliver securities, thereby curbing market manipulation. Manning and other former Escala shareholders (collectively, Manning) filed suit in New Jersey state court, alleging that Merrill Lynch’s actions violated New Jersey law. Though Manning chose not to bring any claims under federal securities laws or rules, his complaint referred explicitly to Regulation SHO, cataloguing past accusations against Merrill Lynch for flouting its requirements and suggesting that the transactions at issue had again violated the regulation. Merrill Lynch removed the case to Federal District Court, asserting federal jurisdiction on two grounds. First, it invoked the general federal question statute, 28 U. S. C. §1331, which grants district courts jurisdiction of “all civil actions arising under” federal law. It also invoked §27 the Securities Exchange Act of 1934 (Exchange Act), which grants federal district courts exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules or regulations thereunder.” 15 U. S. C. §78aa(a). Manning moved to remand the case to state court, arguing that neither statute gave the federal court authority to adjudicate his state-law claims. The District Court denied his motion, but the Third Circuit reversed. The court first decided that §1331 did not confer jurisdiction, because Manning’s claims all arose under state law and did not necessarily raise any federal issues. Nor was the District Court the appropriate forum under §27 of the Exchange Act, which, the court held, covers only those cases that would satisfy §1331’s “arising under” test for general federal jurisdiction. Held: The jurisdictional test established by §27 is the same as §1331’s test for deciding if a case “arises under” a federal law. Pp. 4–18. (a) Section 27’s text more readily supports this meaning than it does the parties’ two alternatives. Merrill Lynch argues that §27’s plain language requires an expansive rule: Any suit that either explicitly or implicitly asserts a breach of an Exchange Act duty is “brought to enforce” that duty even if the plaintiff seeks relief solely under state law. Under the natural reading of that text, however, §27 confers federal jurisdiction when an action is commenced in order to give effect to an Exchange Act requirement. The “brought to enforce” language thus stops short of embracing any complaint that happens to mention a duty established by the Exchange Act. Meanwhile, Manning’s far more restrictive interpretation—that a suit is “brought to enforce” only if it is brought directly under that statute—veers too far in the opposite direction. Instead, §27’s language is best read to capture both suits brought under the Exchange Act and the rare suit in which a state-law claim rises and falls on the plaintiff’s ability to prove the violation of a federal duty. An existing jurisdictional test well captures both of these classes of suits “brought to enforce” such a duty: 28 U. S. C. §1331’s provision of federal jurisdiction of all civil actions “arising under” federal law. Federal jurisdiction most often attaches when federal law creates the cause of action asserted, but it may also attach when the state-law claim “necessarily raise[s] a stated federal issue, actually disputed and substantial, which a federal forum may entertain without disturbing any congressionally approved balance” of federal and state power. Grable & Sons Metal Products, Inc. v. Darue Engineering & Mfg., 545 U. S. 308 . Pp. 5–10. (b) This Court’s precedents interpreting the term “brought to enforce” have likewise interpreted §27’s jurisdictional grant as coextensive with the Court’s construction of §1331’s “arising under” standard. See Pan American, 366 U. S. 656 ; Matsushita Elec. Industrial Co. v. Epstein, 516 U. S. 367 . Pp. 10–14. (c) Construing §27, consistent with both text and precedent, to cover suits that arise under the Exchange Act serves the goals the Court has consistently underscored in interpreting jurisdictional statutes. It gives due deference to the important role of state courts. And it promotes “administrative simplicity[, which] is a major virtue in a jurisdictional statute.” Hertz Corp. v. Friend, 559 U. S. 77 . Both judges and litigants are familiar with the “arising under” standard and how it works, and that test generally provides ready answers to jurisdictional questions. Pp. 14–18. 772 F. 3d 158, affirmed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Alito, JJ., joined. Thomas, J., filed an opinion concurring in the judgment, in which Sotomayor, J., joined.

Section 27 of the Securities Exchange Act of 1934 (Exchange Act), 48Stat. 992, as amended, 15 U. S. C. §78a, et seq., grants federal district courts exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules or regulations thereunder.” §78aa(a). We hold today that the jurisdictional test established by that provision is the same as the one used to decide if a case “arises under” a federal law. See 28 U. S. C. §1331. I Respondent Greg Manning held more than two million shares of stock in Escala Group, Inc., a company traded on the NASDAQ. Between 2006 and 2007, Escala’s share price plummeted and Manning lost most of his investment. Manning blames petitioners, Merrill Lynch and several other financial institutions (collectively, Merrill Lynch), for devaluing Escala during that period through “naked short sales” of its stock. A typical short sale of a security is one made by a borrower, rather than an owner, of stock. In such a transaction, a person borrows stock from a broker, sells it to a buyer on the open market, and later purchases the same number of shares to return to the broker. The short seller’s hope is that the stock price will decline between the time he sells the borrowed shares and the time he buys replacements to pay back his loan. If that happens, the seller gets to pocket the difference (minus associated transaction costs). In a “naked” short sale, by contrast, the seller has not borrowed (or otherwise obtained) the stock he puts on the market, and so never delivers the promised shares to the buyer. See “Naked” Short Selling Antifraud Rule, Securities Exchange Commission (SEC) Release No. 34–58774, 73 Fed. Reg. 61667 (2008). That practice (beyond its effect on individual purchasers) can serve “as a tool to drive down a company’s stock price”—which, of course, injures shareholders like Manning. Id., at 61670. The SEC regulates such short sales at the federal level: The Commission’s Regulation SHO, issued under the Exchange Act, prohibits short sellers from intentionally failing to deliver securities and thereby curbs market manipulation. See 17 CFR §§242.203–242.204 (2015). In this lawsuit, Manning (joined by six other former Escala shareholders) alleges that Merrill Lynch facilitated and engaged in naked short sales of Escala stock, in violation of New Jersey law. His complaint asserts that Merrill Lynch participated in “short sales at times when [it] neither possessed, nor had any intention of obtaining[,] sufficient stock” to deliver to buyers. App. to Pet. for Cert. 57a, Amended Complaint ¶39. That conduct, Manning charges, contravened provisions of the New Jersey RacketeerInfluenced and Corrupt Organizations Act (RICO), New Jersey Criminal Code, and New Jersey Uniform Securities Law; it also, he adds, ran afoul of the New Jersey common law of negligence, unjust enrichment, and interference with contractual relations. See id., at 82a–101a, ¶¶88–161. Manning chose not to bring any claims under federal securities laws or rules. His complaint, however, referred explicitly to Regulation SHO, both describing the purposes of that rule and cataloguing past accusations against Merrill Lynch for flouting its requirements. See id., at 51a–54a, ¶¶28–30; 75a–82a, ¶¶81–87. And the complaint couched its description of the short selling at issue here in terms suggesting that Merrill Lynch had again violated that regulation, in addition to infringing New Jersey law. See id., at 57a–59a, ¶¶39–43. Manning brought his complaint in New Jersey state court, but Merrill Lynch removed the case to Federal District Court. See 28 U. S. C. §1441 (allowing removal of any civil action of which federal district courts have original jurisdiction). Merrill Lynch asserted federal jurisdiction on two grounds. First, it invoked the general federal question statute, §1331, which grants district courts jurisdiction of “all civil actions arising under” federal law. Second, it maintained that the suit belonged in federal court by virtue of §27 of the Exchange Act. That provision, in relevant part, grants district courts exclusive jurisdiction of “all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a). Manning moved to remand the case to state court, arguing that neither statute gave the federal court authority to adjudicate his collection of state-law claims. The District Court denied his motion. See No. 12–4466 (D NJ, Mar. 18, 2013), App. to Pet. for Cert. 24a–38a. The Court of Appeals for the Third Circuit reversed, ordering a remand of the case to state court. See 772 F. 3d 158 (2014). The Third Circuit first decided that the fed-eral question statute, 28 U. S. C. §1331, did not confer juris-diction of the suit, because all Manning’s claims were “brought under state law” and none “necessarily raised” a federal issue. 772 F. 3d, at 161, 163. Nor, the court held, did §27 of the Exchange Act make the district court the appropriate forum. Relying on this Court’s construction of a nearly identical jurisdictional provision, the Court of Appeals found that §27 covers only those cases involving the Exchange Act that would satisfy the “arising under” test of the federal question statute. See id., at 166–167 (citing Pan American Petroleum Corp. v. Superior Court of Del. for New Castle Cty., 366 U. S. 656 (1961) ). Because the District Court lacked jurisdiction of Manning’s suit under §1331, so too it was not the exclusive forum under §27. Merrill Lynch sought this Court’s review solely as to whether §27 commits Manning’s case to federal court. See Pet. for Cert. i. Because of a Circuit split about that provision’s meaning,[1] we granted certiorari. 576 U. S. ___ (2015). We now affirm. II Like the Third Circuit, we read §27 as conferring exclusive federal jurisdiction of the same suits as “aris[e] under” the Exchange Act pursuant to the general federal question statute. See 28 U. S. C. §1331. The text of §27 more readily supports that meaning than it does either of the parties’ two alternatives. This Court’s precedents interpreting identical statutory language positively compel that conclusion. And the construction fits with our practice of reading jurisdictional laws, so long as consistent with their language, to respect the traditional role of state courts in our federal system and to establish clear and administrable rules. A Section 27, as noted earlier, provides federal district courts with exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a); see supra, at 3.[2] Much the same wording appears in nine other federal jurisdictional provisions—mostly enacted, like §27, as part of New Deal-era regulatory statutes.[3] Merrill Lynch argues that the “plain, unambiguous language” of §27 requires an expansive understanding of its scope. Brief for Petitioners 23. Whenever (says Merrill Lynch) a plaintiff’s complaint either explicitly or implic-itly “assert[s]” that “the defendant breached an Exchange Act duty,” then the suit is “brought to enforce” that duty and a federal court has exclusive jurisdiction. Id., at 22; Reply Brief 10–11; see Tr. of Oral Arg. 7–8 (confirming that such allegations need not be express). That is so, Merrill Lynch contends, even if the plaintiff, as in this case, brings only state-law claims in his complaint—that is, seeks relief solely under state law. See Reply Brief 3–6. And it is so, Merrill Lynch continues, even if the plaintiff can prevail on those claims without proving that the alleged breach of an Exchange Act duty—here, the violation of Regulation SHO—actually occurred. See id., at 7–13; Tr. of Oral Arg. 3 (“[T]he words ‘brought to enforce’ [donot focus] on what the court would necessarily have to decide”). But a natural reading of §27’s text does not extend so far. “Brought” in this context means “commenced,” Black’s Law Dictionary 254 (3d ed. 1933); “to” is a word “expressing purpose [or] consequence,” The Concise Oxford Dictionary 1288 (1931); and “enforce” means “give force [or] effect to,” 1 Webster’s New International Dictionary of the English Language 725 (1927). So §27 confers federal jurisdiction when an action is commenced in order to give effect to an Exchange Act requirement. That language, in emphasizing what the suit is designed to accomplish, stops short of embracing any complaint that happens to mention a duty established by the Exchange Act. Consider, for example, a simple state-law action for breach of contract, in which the plaintiff alleges, for atmospheric reasons, that the defendant’s conduct also violated the Exchange Act—or still less, that the defendant is a bad actor who infringed that statute on another occasion. On Merrill Lynch’s view, §27 would cover that suit; indeed, Merrill Lynch points to just such incidental assertions as the basis for federal jurisdiction here. See Brief for Petitioners 20–21; supra, at 3. But that hypothetical suit is “brought to enforce” state contract law, not the Exchange Act—because the plaintiff can get all the relief he seeks just by showing the breach of an agreement, without proving any violation of federal securities law. The suit, that is, can achieve all it is supposed to even if issues involving the Exchange Act never come up. Critiquing Merrill Lynch’s position on similar grounds, Manning proposes a far more restrictive interpretation of §27’s language—one going beyond what he needs to prevail. See Brief for Respondents 27–33. According to Manning, a suit is “brought to enforce” the Exchange Act’s duties or liabilities only if it is brought directly under that statute—that is, only if the claims it asserts (and not just the duties it means to vindicate) are created by the Exchange Act. On that view, everything depends (as Justice Holmes famously said in another jurisdictional context) on which law “creates the cause of action.” American Well Works Co. v. Layne & Bowler Co., 241 U. S. 257, 260 (1916) . If a complaint asserts a right of action deriving from the Exchange Act (or an associated regulation), the suit must proceed in federal court. But if, as here, the complaint brings only state-created claims, then the case belongs in a state forum. And that is so, Manning claims, even if—contrary to what the Third Circuit held below—the success of the state claim necessarily hinges on proving that the defendant breached an Exchange Act duty. See Brief for Respondents 31. Manning’s view of the text’s requirements, although better than Merrill Lynch’s, veers too far in the opposite direction. There is no doubt, as Manning says, that a suit asserting an Exchange Act cause of action fits within §27’s scope: Bringing such a suit is the prototypical way of enforcing an Exchange Act duty. But it is not the only way. On rare occasions, as just suggested, a suit raising a state-law claim rises or falls on the plaintiff’s ability to prove the violation of a federal duty. See, e.g., Grable & Sons Metal Products, Inc. v. Darue Engineering & Mfg., 545 U. S. 308 –315 (2005); Smith v. Kansas City Title & Trust Co., 255 U. S. 180, 201 (1921) . If in that manner, a state-law action necessarily depends on a showing that the defendant breached the Exchange Act, then that suit could also fall within §27’s compass. Suppose, for example, that a state statute simply makes illegal “any violation of the Exchange Act involving naked short selling.” A plaintiff seeking relief under that state law must undertake to prove, as the cornerstone of his suit, that the defendant infringed a requirement of the federal statute. (Indeed, in this hypothetical, that is the plaintiff’s only project.) Accordingly, his suit, even though asserting a state-created claim, is also “brought to enforce” a duty created by the Exchange Act. An existing jurisdictional test well captures both classes of suits “brought to enforce” such a duty. As noted earlier, 28 U. S. C. §1331 provides federal jurisdiction of all civil actions “arising under” federal law. See supra, at 3. This Court has found that statutory term satisfied in either of two circumstances. Most directly, and most often, federal jurisdiction attaches when federal law creates the cause of action asserted. That set of cases is what Manning highlights in offering his view of §27. But even when “a claim finds its origins” in state law, there is “a special and small category of cases in which arising under jurisdiction still lies.” Gunn v. Minton, 568 U. S. ___, ___ (2013) (slip op., at 6) (internal quotation marks omitted). As this Court has explained, a federal court has jurisdiction of a state-law claim if it “necessarily raise[s] a stated federal issue, actually disputed and substantial, which a federal forum may entertain without disturbing any congressionally approved balance” of federal and state power. Grable, 545 U. S., at 314; see Gunn, 568 U. S., at ___ (slip op., at 6) (framing the same standard as a four-part test). That description typically fits cases, like those described just above, in which a state-law cause of action is “brought to enforce” a duty created by the Exchange Act because the claim’s very success depends on giving effect to a federal requirement. Accordingly, we agree with the court below that §27’s jurisdictional test matches the one we have formulated for §1331, as applied to cases involving the Exchange Act. If (but only if) such a case meets the “arising under” standard, §27 commands that it go to federal court.[4] Merrill Lynch objects that our rule construes “completely different language”—i.e., the phrases “arising under”and “brought to enforce” in §1331 and §27, respectively—“to mean exactly the same thing.” Reply Brief 7. We cannot deny that point. But we think it far less odd than Merrill Lynch does. After all, the test for §1331 jurisdiction is not grounded in that provision’s particular phrasing. This Court has long read the words “arising under” in Article III to extend quite broadly, “to all cases in which a federal question is ‘an ingredient’ of the action.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804 (1986) (quoting Osborn v. Bank of United States, 9 Wheat. 738, 823 (1824)). In the statutory context, however, we opted to give those same words a narrower scope “in the light of [§1331’s] history[,] the demands of reason and coherence, and the dictates of sound judicial policy.” Romero v. International Terminal Operating Co., 358 U. S. 354, 379 (1959) . Because the resulting test does not turn on §1331’s text, there is nothing remarkable in its fitting as, or even more, neatly a differently worded statutory provision. Nor can Merrill Lynch claim that Congress’s use of the new “brought to enforce” language in §27 shows an intent to depart from a settled (even if linguistically ungrounded) test for statutory “arising under” jurisdiction. That is because no such well-defined test then existed. As we recently noted, our caselaw construing §1331 was for many decades—including when the Exchange Act passed—highly “unruly.” Gunn, 568 U. S., at __ (slip op., at 6) (referring to the “canvas” of our old opinions as “look[ing] like one that Jackson Pollock got to first”). Against that muddled backdrop, it is impossible to infer that Congress, in enacting §27, wished to depart from what we now understand as the “arising under” standard. B This Court has reached the same conclusion before. In two unrelated decisions, we addressed the “brought to enforce” language at issue here. See Pan American, 366 U. S. 656 ; Matsushita Elec. Industrial Co. v. Epstein, 516 U. S. 367 (1996) . Each time, we viewed that phrase as coextensive with our construction of “arising under.” Pan American involved §22 of the Natural Gas Act (NGA), 15 U. S. C. §717u—an exclusive jurisdiction provision containing language materially indistinguishable from §27’s.[5] The case began in state court when a natural gas purchaser sued a producer for breach of a contract setting sale prices. Prior to the alleged breach, the producer had filed those contractual rates with the Federal Power Commission, as the NGA required. Relying on that submission (which the complaint did not mention), the producer claimed that the buyer’s suit was “brought to enforce” a liability deriving from the NGA—i.e., a filed rate—and so must proceed in federal court. See 366 U. S., at 662. This Court rejected the argument. Our decision explained that §22’s use of the term “brought to enforce,” rather than “arising under,” made no difference to the jurisdictional analysis. The inquiry, we wrote, was “not affected by want” of the language contained in the federal question statute. Id., at 665, n. 2. The “limitation[s]” associated with “arising under” jurisdiction, we continued, were “clearly implied” in §22’s alternative phrasing. Ibid. In short, the linguistic distinction between the two jurisdictional provisions did not extend to their meaning. Pan American thus went on to analyze the jurisdictional issue in the manner set out in our “arising under” precedents. Federal question jurisdiction lies, the Court wrote, only if “it appears from the face of the complaint that determination of the suit depends upon a question of federal law.” Id., at 663. That inquiry focuses on “the particular claims a suitor makes” in his complaint—meaning, whether the plaintiff seeks relief under state or federal law. Id., at 662. In addition, the Court suggested, a federal court could adjudicate a suit stating only a state-law claim if it included as “an element, and an essential one,” the violation of a federal right. Id., at 663 (quoting Gully v. First Nat. Bank in Meridian, 299 U. S. 109, 112 (1936) ). With those principles of “arising under” jurisdiction laid out, the Court held that §22 did not enable a federal court to resolve the buyer’s case, because he could prevail merely by proving breach of the contract. See 366 U. S., at 663–665. Pan American establishes, then, that an action “brought to enforce” a duty or liability created by a federal statute is nothing more (and nothing less) than an action “arising under” that law. Merrill Lynch reads Pan American more narrowly, as holding only that §22 does not confer federal jurisdiction when a complaint (unlike Manning’s) fails to reference federal law at all. See Brief for Petitioners 32–33, 38. But that argument ignores Pan American’s express statement of equivalence between §27’s language and the federal question statute’s: “Brought to enforce” has the same “limitation[s]” (meaning, the same scope) as “arising under.” 366 U. S., at 665, n. 2. And just as important, Merrill Lynch disregards Pan American’s analytical structure: The decision proceeds by reviewing this Court’s “arising under” precedents, articulating the principles animating that caselaw, and then applying those tenets to the dispute at hand. Id., at 662–665. The Court thus showed (as well as told) that “brought to enforce” jurisdiction mirrors that of “arising under.” As a fallback, Merrill Lynch claims that Pan American is irrelevant here because it relied on legislative history distinct to the NGA in finding §22’s “brought to enforce” language coterminous with “arising under.” See Brief for Petitioners 38–39. The premise of that argument is true enough: In support of its holding, the Court quoted a Committee Report describing §22 as conferring federal jurisdiction “over cases arising under the act.” 366 U. S., at 665, n. 2. But we cannot accept the conclusion Merrill Lynch draws from that statement: that courts should give two identically worded statutory provisions, passed less than five years apart, markedly different meanings. Indeed, the result of Merrill Lynch’s approach is still odder, for what of the eight other jurisdictional provisions containing “brought to enforce” language? See n. 3, supra. Presumably, Merrill Lynch would have courts inspect each of their legislative histories to decide whether to read those statutes as reproducing the “arising under” standard, adopting Merrill Lynch’s alternative view, or demanding yet another jurisdictional test. We are hard pressed to imagine a less sensible way of construing the repeated iterations of the phrase “brought to enforce” in the jurisdictional provisions of the Federal Code. In any event, this Court in Matsushita addressed §27 itself, and once again equated the “brought to enforce” and “arising under” standards. That decision arose from a state-law action against corporate directors for breach of fiduciary duty. The issue was whether the state court handling the suit could approve a settlement releasing, in addition to the state claims actually brought, potential Exchange Act claims that §27 would have committed to federal court. In deciding that the state court could do so, we described §27—not once, not twice, but three times—as conferring exclusive jurisdiction of suits “arising under” the Exchange Act. See 516 U. S., at 380 (Section 27 “confers exclusive jurisdiction upon the federal courts for suits arising under the [Exchange] Act”); id., at 381 (Section 27 “prohibits state courts from adjudicating claims arising under the Exchange Act”); id., at 385 (Section 27 “prohibit[s] state courts from exercising jurisdiction over suits arising under the Exchange Act”) (emphases added). Over and over, then, the Court took as a given that §27’s jurisdictional test mimicked the one in the general federal question statute. And still more: The Matsushita Court thought clear that the suit as filed—which closely resembled Manning’s in its mix of state and federal law—fell outside §27’s grant of exclusive jurisdiction. As just noted, the claims brought in the Matsushita complaint sought relief for breach of a state-law duty. But in support of those claims, the plaintiffs charged, much as Manning did here, that the defendants’ conduct also violated federal securities laws. See 516 U. S., at 370; supra, at 2–3. We found the presence of that accusation insufficient to trigger §27. “[T]he cause pleaded,” we wrote, remained “a state common-law action,” 516 U. S., at 382, n. 7: Notwithstanding the potential federal issue, the suit “was not ‘brought to enforce’ any rights or obligations under the [Exchange] Act,” id., at 381. The Court thus rejected the very position Merrill Lynch takes here—i.e., that §27 precludes a state court from adjudicating any case, even if brought under state law, in which the plaintiff asserts an Exchange Act breach. C Construing §27, consistent with both text and precedent, to cover suits that arise under the Exchange Act serves the goals we have consistently underscored in interpreting jurisdictional statutes. Our reading, unlike Merrill Lynch’s, gives due deference to the important role of state courts in our federal system. And the standard we adopt is more straightforward and administrable than the alternative Merrill Lynch offers. Out of respect for state courts, this Court has time and again declined to construe federal jurisdictional statutes more expansively than their language, most fairly read, requires. We have reiterated the need to give “[d]ue regard [to] the rightful independence of state governments”—and more particularly, to the power of the States “to provide for the determination of controversies in their courts.” Romero, 358 U. S., at 380 (quoting Healy v. Ratta, 292 U. S. 263, 270 (1934) ; Shamrock Oil & Gas Corp. v. Sheets, 313 U. S. 100, 109 (1941) . Our decisions, as we once put the point, reflect a “deeply felt and traditional reluctance . . . to expand the jurisdiction of federal courts through a broad reading of jurisdictional statutes.” Romero, 358 U. S., at 379.[6] That interpretive stance serves, among other things, to keep state-law actions like Manning’s in state court, and thus to help maintainthe constitutional balance between state and federaljudiciaries. Nor does this Court’s concern for state court prerogatives disappear, as Merrill Lynch suggests it should, in the face of a statute granting exclusive federal jurisdiction. See Brief for Petitioners 23–27. To the contrary, when a statute mandates, rather than permits, federal jurisdiction—thus depriving state courts of all ability to adjudicate certain claims—our reluctance to endorse “broad reading[s],” Romero, 358 U. S., at 379, if anything, grows stronger. And that is especially so when, as here, the construction offered would place in federal court actions bringing only claims created by state law—even if those claims might raise federal issues. To be sure, a grant of exclusive federal jurisdiction, as Merrill Lynch reminds us, indicates that Congress wanted “greater uniformity of construction and more effective and expert application” of federal law than usual. Brief for Petitioners 24 (quoting Matsushita, 516 U. S., at 383). But “greater” and “more” do not mean “total,” and the critical question remains how far such a grant extends. In resolving that issue, we will not lightly read the statute to alter the usual constitu-tional balance, as it would by sending actions with all state-law claims to federal court just because a complaint references a federal duty. Our precedents construing other exclusive grants of federal jurisdiction illustrate those principles. In Pan American, for example, we denied that a state court’s resolution of state-law claims potentially implicating the NGA’s meaning would “jeopardize the uniform system of regulation” that the statute established. 366 U. S., at 665. We reasoned that this Court’s ability to review state court decisions of federal questions would sufficiently protect federal interests. And similarly, in Tafflin v. Levitt, 493 U. S. 455 –467 (1990), we permitted state courts to adjudicate civil RICO actions that might raise issues about the scope of federal crimes alleged as predicate acts, even though federal courts have exclusive jurisdiction “of all offenses against the laws of the United States,” 18 U. S. C. §3231. There, we expressed confidence that state courts would look to federal court interpretations of the relevant criminal statutes. Accordingly, we saw “no significant danger of inconsistent application of federal criminal law” and no “incompatibility with federal interests.” Tafflin, 493 U. S., at 464–465, 467 (internal quotation marks omitted). So too here, when state courts, in deciding state-law claims, address possible issues of the Exchange Act’s meaning. Not even Merrill Lynch thinks those decisions wholly avoidable: It admits that §27 does nothing to prevent state courts from resolving Exchange Act questions that result from defenses or counterclaims. See Brief for Petitioners 32–33; Pan American, 366 U. S., at 664–665. We see little difference, in terms of the uniformity-based policies Merrill Lynch invokes, if those issues instead appear in a complaint like Manning’s. And indeed, Congress likely contemplated that some complaints intermingling state and federal questions would be brought in state court: After all, Congress specifically affirmed the capacity of such courts to hear state-law securities actions, which predictably raise issues coinciding, overlapping, or intersecting with those under the Act itself. See 15 U. S. C. §78bb(a)(2); Matsushita, 516 U. S., at 383. So, for example, it is hardly surprising in a suit like this one, alleging short sales in violation of state securities law, that a plaintiff might say the defendant previously breached a federal prohibition of similar conduct. See supra, at 2–3 (describing Manning’s complaint). And it is less troubling for a state court to consider such an issue than to lose all ability to adjudicate a suit raising only state-law causes of action. Reading §27 in line with our §1331 caselaw also promotes “administrative simplicity[, which] is a major virtue in a jurisdictional statute.” Hertz Corp. v. Friend, 559 U. S. 77, 94 (2010) . Both judges and litigants are familiar with the “arising under” standard and how it works. For the most part, that test provides ready answers to jurisdictional questions. And an existing body of precedent gives guidance whenever borderline cases crop up. See supra, at 8–9. By contrast, no one has experience with Merrill Lynch’s alternative standard, which would spring out of nothing to govern suits involving not only the Exchange Act but up to nine other discrete spheres of federal law. See n. 3, supra (listing statutes with “brought to enforce” language); supra, at 12–13 (noting Merrill Lynch’s backup claim that legislative histories might compel different tests for different statutes). Adopting such an untested approach, and forcing courts to toggle back and forth between it and the “arising under” standard, would undermine consistency and predictability in litigation. That result disserves courts and parties alike. Making matters worse, Merrill Lynch’s rule is simple for plaintiffs to avoid—or else, excruciating for courts to police. Under that rule, a plaintiff electing to bring state-law claims in state court will purge his complaint of any references to federal securities law, so as to escape re-moval. Such omissions, after all, will do nothing to change the way the plaintiff can present his case at trial; they will merely make the complaint less informative. Recognizing the potential for that kind of avoidance, Merrill Lynch argues that a judge should go behind the face of a complaint to determine whether it is the product of “artful pleading.” See Tr. of Oral Arg. 7 (If the plaintiffs “had just literally whited out, deleted the references to Reg[ulation] SHO,” the court should still understand the complaint to allege a breach of that rule; “the fact [that the plaintiffs] didn’t cite it wouldn’t change the fact”). We have no idea how a court would make that judgment, and get cold comfort from Merrill Lynch’s assurance that the question would arise not in this case but in “the next third, fourth, fifth case down the road.” Id., at 8. Jurisdictional tests are built for more than a single dispute: That Merrill Lynch’s threatens to become either a useless drafting rule or a tortuous inquiry into artful pleading is one more good reason to reject it. III Section 27 provides exclusive federal jurisdiction of the same class of cases as “arise under” the Exchange Act for purposes of §1331. The text of §27, most naturally read, supports that rule. This Court has adopted the same view in two prior cases. And that reading of the statute promotes the twin goals, important in interpreting jurisdictional grants, of respecting state courts and providing administrable standards. Our holding requires remanding Manning’s suit to state court. The Third Circuit found that the District Court did not have jurisdiction of Manning’s suit under §1331 because all his claims sought relief under state law and none necessarily raised a federal issue. See supra, at 3. Merrill Lynch did not challenge that ruling, and we therefore take it as a given. And that means, under our decision today, that the District Court also lacked jurisdiction under §27. Accordingly, we affirm the judgment below. It is so ordered.Notes 1 Compare 772 F. 3d 158 (CA3 2014) (case below) with Barbara v. New York Stock Exchange, Inc., 99 F. 3d 49, 55 (CA2 1996) (construing §27 more narrowly), Sparta Surgical Corp. v. National Assn. of Securities Dealers, Inc., 159 F. 3d 1209, 1211–1212 (CA9 1998) (construing §27 more broadly), and Hawkins v. National Assn. of Securities Dealers, Inc., 149 F. 3d 330, 331–332 (CA5 1998) (per curiam) (same). 2 Section 27 also grants federal courts exclusive jurisdiction of “violations of [the Exchange Act] or the rules and regulations thereunder.” 15 U. S. C. §78aa(a). Manning argues that the “violations” language applies only to criminal proceedings and SEC enforcement actions. See Brief for Respondents 28. Merrill Lynch, although not conceding that much, believes the “violations” clause irrelevant here because, in private suits for damages, it goes no further than the “brought to enforce” language quoted in the text. See Reply Brief 1, n. 1. Given that both parties have thus taken the “violations” language off the table, we do not address its meaning. 3 See Securities Act of 1933, 15 U. S. C. §77v(a); Federal Power Act of 1935, 16 U. S. C. §825p; Connally Hot Oil Act of 1935, 15 U. S. C. §715i(c); Natural Gas Act of 1938, 15 U. S. C. §717u; Trust Indenture Act of 1939, 15 U. S. C. §77vvv(b); Investment Company Act of 1940, 15 U. S. C. §80a–43; Investment Advisers Act of 1940, 15 U. S. C. §80b–14(a); International Wheat Agreement Act of 1949, 7 U. S. C. §1642(e); Interstate Land Sales Full Disclosure Act of 1968, 15 U. S. C. §1719. 4 The concurrence adopts a slightly different approach, placing in federal court Exchange Act claims plus all state-law claims necessarily raising an Exchange Act issue. See post, at 2–3 (Thomas, J., concurring in judgment). In other words, the concurrence would not ask, as the “arising under” test does, whether the federal issue embedded in such a state-law claim is also substantial, actually disputed, and capable of resolution in federal court without disrupting the congressionally approved federal-state balance. See post, at 6–7; Grable & Sons Metal Products, Inc. v. Darue Engineering & Mfg., 545 U. S. 308, 314 (2005) . But this Court has not construed any jurisdictional statute, whether using the words “brought to enforce” or “arising under” (or for that matter, any other), to draw the concurrence’s line. For as long as we have contemplated exercising federal jurisdiction over state-law claims necessarily raising federal issues, we have inquired as well into whether those issues are “really and substantially” disputed. See, e.g., Hop-kins v. Walker, 244 U. S. 486, 489 (1917) ; Shulthis v. McDougal, 225 U. S. 561, 569 (1912) . And similarly, we have long emphasized the need in such circumstances to make “sensitive judgments about congressional intent, judicial power, and the federal system.” Merrell Dow Pharmaceuticals Inc. v. Thompson, 478 U. S. 804, 810 (1986) . At this late juncture, we see no virtue in trying to pull apart these interconnected strands of necessity and substantiality-plus. Indeed, doing so here—and thus creating a gap between our “brought to enforce” and “arising under” standards—would conflict with this Court’s precedent and undermine important goals of interpreting jurisdictional statutes. See infra, at 10–14 (discussing our prior decisions equating the two tests), 14–17 (highlighting the need to respect state courts and the benefits of using a single, time-tested standard). 5 Section 22 grants federal courts exclusive jurisdiction “of all suits in equity and actions at law brought to enforce any liability or duty created by . . . [the NGA] or any rule, regulation, or order thereunder.” 52Stat. 833. 6 The Romero Court continued: “A reluctance which must be even more forcefully felt when the expansion is proposed, for the first time, eighty-three years after the jurisdiction has been conferred.” 358 U. S., at 379. The Exchange Act was passed a mere 82 years ago, but we believe the point still stands.

578.US.2015_14-8913

The Federal Sentencing Guidelines first enter the sentencing process when the United States Probation Office prepares a presentence report containing, as relevant here, an advisory Guidelines range based on the seriousness of a defendant’s offense and the extent of his criminal history. A district court may depart from the Guidelines, but it “must consult [them] and take them into account when sentencing.” United States v. Booker, 543 U. S. 220 . Given the Guidelines’ complexity, a District Court’s use of an incorrect Guidelines range may go unnoticed. That error can be remedied on appeal pursuant to Federal Rule of Criminal Procedure 52(b), provided that (1) there is an error that was not intentionally relinquished or abandoned, United States v. Olano, 507 U. S. 725 –733; (2) the error is plain, i.e., clear or obvious, id., at 734; and (3) the error affected the defendant’s substantial rights, ibid., which in the ordinary case means he or she must “show a reasonable probability that, but for the error,” the outcome of the proceeding would have been different, United States v. Dominguez Benitez, 542 U. S. 74 . Once these three conditions have been met, the court of appeals should exercise its discretion to correct the forfeited error if the error “ ‘seriously affects the fairness, integrity or public reputation of judicial proceedings.’ ” Olano, 507 U. S., at 736 (brackets omitted). Petitioner Molina-Martinez pleaded guilty to being unlawfully present in the United States after having been deported following an aggravated felony conviction. The Guidelines range in his presentence report was 77 to 96 months. He requested, and the Probation Office recommended, a 77-month sentence, while the Government requested 96 months. The District Court, with little explanation, sentenced him to the lowest end of what it believed to be the applicable Guidelines range—77 months. On appeal, Molina-Martinez argued for the first time that the Probation Office and the District Court miscalculated his Guidelines range, which should have been 70 to 87 months, and noted that his 77-month sentence would have been in the middle of the correct range, not at the bottom. The Fifth Circuit agreed that the District Court used an incorrect Guidelines range but found that Molina-Martinez could not satisfy Rule 52(b)’s requirement that the error affect his substantial rights. It reasoned that a defendant whose sentence falls within what would have been the correct Guidelines range must, on appeal, identify “additional evidence” showing that use of the incorrect Guidelines range in fact affected his sentence. Held: Courts reviewing Guidelines errors cannot apply a categorical “additional evidence” rule in cases, like this one, where a district court applies an incorrect range but sentences the defendant within the correct range. Pp. 8–16. (a) The Guidelines establish the essential framework for sentencing proceedings. Sentencing courts “ ‘must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process.’ ” Peugh v. United States, 569 U. S. ___, ___. Sentencing Commission statistics confirm that the Guidelines inform and instruct the district court’s determination of an appropriate sentence. In the usual case, the systemic function of the selected Guidelines range will affect a defendant’s sentence. As a result, a defendant who shows that the district court mistakenly deemed applicable an incorrect, higher range will, in the ordinary case, have demonstrated a reasonable probability of a different outcome. That showing will suffice for relief if Rule 52(b)’s other requirements are met. Pp. 9–12. (b) The unworkable nature of the Fifth Circuit’s “additional evidence” rule is evident here, where the record shows that the District Court gave little explanation for the sentence it selected, rejected the Government’s request for a sentence at the top of the erroneous Guidelines range, and chose the sentence requested by the defendant and recommended by the Probation Office—a sentence at the bottom of the erroneous Guidelines range. This demonstrates that the Guidelines served as the starting point for the sentencing and were the focal point for the proceedings that followed. Given the sentence the District Court chose, and because the court said nothing to suggest that it would have imposed the same sentence regardless of the Guidelines range, there is at least a reasonable probability that the court would have imposed a different sentence had it known that 70 months was the lowest sentence the Commission deemed appropriate. Pp. 12–13. (c) Rejection of the Fifth Circuit’s rule means only that a defendant can rely on the application of an incorrect Guidelines range to show an effect on his substantial rights, not that the Government will have to prove that every Guidelines error was harmless. And the Government’s concern over the judicial resources needed for the resentencing proceedings that might result from today’s holding is unfounded because the holding is consistent with the approach taken by most Courts of Appeals and because remanding for resentencing is less costly than remanding for retrial. Pp. 13–15. 588 Fed. Appx. 333, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in part and concurring in the judgment, in which Thomas, J., joined.

This case involves the Federal Sentencing Guidelines. In sentencing petitioner, the District Court applied a Guidelines range higher than the applicable one. The error went unnoticed by the court and the parties, so no timely objection was entered. The error was first noted when, during briefing to the Court of Appeals for the Fifth Circuit, petitioner himself raised the mistake. The Court of Appeals refused to correct the error because, in its view, petitioner could not establish a reasonable probability that but for the error he would have received a different sentence. Under that court’s decisions, if a defendant’s ultimate sentence falls within what would have been the correct Guidelines range, the defendant, on appeal, must identify “additional evidence” to show that use of the incorrect Guidelines range did in fact affect his sentence. Absent that evidence, in the Court of Appeals’ view, a defendant who is sentenced under an incorrect range but whose sentence is also within what would have been the correct range cannot demonstrate he has been prejudiced by the error. Most Courts of Appeals have not adopted so rigid a standard. Instead, in recognition of the Guidelines’ central role in sentencing, other Courts of Appeals have concluded that a district court’s application of an incorrect Guidelines range can itself serve as evidence of an effect on substantial rights. See, e.g., United States v. Sabillon-Umana, 772 F. 3d 1328, 1333 (CA10 2014) (application of an erroneous Guidelines range “ ‘runs the risk of affecting the ultimate sentence regardless of whether the court ultimately imposes a sentence within or outside’ ” that range); United States v. Vargem, 747 F. 3d 724, 728–729 (CA9 2014); United States v. Story, 503 F. 3d 436, 440 (CA6 2007). These courts recognize that, in most cases, when a district court adopts an incorrect Guidelinesrange, there is a reasonable probability that the defendant’s sentence would be different absent the error. This Court granted certiorari to reconcile the difference in approaches. I A The Sentencing Guidelines provide the framework for the tens of thousands of federal sentencing proceedings that occur each year. Congress directed the United States Sentencing Commission (USSC or Commission) to establish the Guidelines. 28 U. S. C. §994(a)(1). The goal was to achieve “ ‘uniformity in sentencing . . . imposed by different federal courts for similar criminal conduct,’ as well as ‘proportionality in sentencing through a system that imposes appropriately different sentences for criminal conduct of different severity.’ ” Rita v. United States, 551 U. S. 338, 349 (2007) . To those ends, the Commission engaged in “a deliberative and dynamic process” to create Guidelines that account for a variety of offenses and circumstances. USSC, Guidelines Manual §2, ch. 1, pt. A, intro. comment., p. 14 (Nov. 2015) (USSG). As part of that process, the Commission considered the objectives of federal sentencing identified in the Sentencing Reform Act of 1984—the same objectives that federal judges must consider when sentencing defendants. Compare 28 U. S. C. §991(b) with 18 U. S. C. §3553(a). The result is a set of elaborate, detailed Guidelines that aim to embody federal sentencing objectives “both in principle and in practice.” Rita, supra, at 350. Uniformity and proportionality in sentencing are achieved, in part, by the Guidelines’ significant role in sentencing. See Peugh v. United States, 569 U. S. ___, ___ (2013) (slip op., at 10). The Guidelines enter the sentencing process long before the district court imposes the sentence. The United States Probation Office first prepares a presentence report which includes a calculation of the advisory Guidelines range it considers to be applicable. Fed. Rules Crim. Proc. 32(d)(1)(A)–(C); see generally 18 U. S. C. §3552(a). The applicable Guidelines range is based on the seriousness of a defendant’s offense (indicated by his “offense level”) and his criminal history (indi-cated by his “criminal history category”). Rules 32(d)(1)(B)–(C). The presentence report explains the basis for the Probation Office’s calculations and sets out the sentencing options under the applicable statutes and Guidelines. Rule 32(d)(1). It also contains detailed information about the defendant’s criminal history and personal characteristics, such as education and employment his-tory. Rule 32(d)(2). At the outset of the sentencing proceedings, the district court must determine the applicable Guidelines range. Peugh, supra, at ___ (slip op., at 10). To do so, the court considers the presentence report as well as any objections the parties might have. The court then entertains the parties’ arguments regarding an appropriate sentence, including whether the sentence should be within the Guidelines range or not. Although the district court has discretion to depart from the Guidelines, the court “must consult those Guidelines and take them into account when sentencing.” United States v. Booker, 543 U. S. 220, 264 (2005) . B The Guidelines are complex, and so there will be instances when a district court’s sentencing of a defendant within the framework of an incorrect Guidelines range goes unnoticed. In that circumstance, because the defendant failed to object to the miscalculation, appellate review of the error is governed by Federal Rule of Criminal Procedure 52(b). Rule 52, in both its parts, is brief. It states: “(a) Harmless Error. Any error, defect, irregularity, or variance that does not affect substantial rights must be disregarded. “(b) Plain Error. A plain error that affects substantial rights may be considered even though it was not brought to the court’s attention.” The starting point for interpreting and applying paragraph (b) of the Rule, upon which this case turns, is the Court’s decision in United States v. Olano, 507 U. S. 725 (1993) . Olano instructs that a court of appeals has discretion to remedy a forfeited error provided certain conditions are met. First, there must be an error that has not been intentionally relinquished or abandoned. Id., at 732–733. Second, the error must be plain—that is to say, clear or obvious. Id., at 734. Third, the error must have affected the defendant’s substantial rights, ibid., which in the ordinary case means he or she must “show a reasonable probability that, but for the error,” the outcome of the proceeding would have been different, United States v. Dominguez Benitez, 542 U. S. 74, 76, 82 (2004) . Once these three conditions have been met, the court of appeals should exercise its discretion to correct the forfeited error if the error “ ‘seriously affects the fairness, integrity or public reputation of judicial proceedings.’ ” Olano, supra, at 736 (brackets omitted). II The petitioner here, Saul Molina-Martinez, pleaded guilty to being unlawfully present in the United States after having been deported following an aggravated felony conviction, in violation of 8 U. S. C. §§1326(a) and (b). As required, the Probation Office prepared a presentence report that related Molina-Martinez’s offense of conviction, his criminal history, his personal characteristics, and the available sentencing options. The report also included the Probation Office’s calculation of what it believed to be Molina-Martinez’s Guidelines range. The Probation Office calculated Molina-Martinez’s total offense level as 21. It concluded that Molina-Martinez’s criminal history warranted 18 points, which included 11 points for five aggravated burglary convictions from 2011. Those 18 criminal history points resulted in a criminal history category of VI. That category, combined with an offense level of 21, resulted in a Guidelines range of 77 to 96 months. At the sentencing hearing Molina-Martinez’s counsel and the Government addressed the court. The Government acknowledged that the Probation Office had “recommended the low end on this case, 77 months.” App. 30. But, the prosecution told the court, it “disagree[d] with that recommendation,” and was “asking for a high end sentence of 96 months”—the top of the Guidelines range. Ibid. Like the Probation Office, counsel for Molina-Martinez urged the court to enter a sentence at the bottom of the Guidelines range. Counsel asserted that “77 months is a severe sentence” and that “after the 77 months, he’ll be deported with probably a special release term.” Id., at 32. A sentence of 77 months, counsel continued, “is more than adequate to ensure he doesn’t come back again.” Ibid. After hearing from the parties, the court stated it was adopting the presentence report’s factual findings and Guidelines calculations. It then ordered Molina-Martinez’s sentence: “It’s the judgment of the Court that the defendant, Saul Molina-Martinez, is hereby committed to the custody of the Bureau of Prisons to be imprisoned for a term of 77 months. Upon release from imprisonment, Defendant shall be placed on supervised release for a term of three years without supervision.” Id.,at 33. The court provided no further explanation for thesentence. On appeal, Molina-Martinez’s attorney submitted a brief pursuant to Anders v. California, 386 U. S. 738 (1967) . The attorney explained that, in his opinion, there were no nonfrivolous grounds for appeal. Molina-Martinez, however, submitted a pro se response to his attorney’s Anders brief. In it he identified for the first time what he believed to be an error in the calculation of his criminal history points under the Guidelines. The Court of Appeals concluded that Molina-Martinez’s argument did not appear frivolous. It directed his lawyer to file either a supplemental Anders brief or a brief on the merits of the Guidelines issue. Molina-Martinez, through his attorney, filed a merits brief arguing that the Probation Office and the District Court erred in calculating his criminal history points, resulting in the application of a higher Guidelines range. The error, Molina-Martinez explained, occurred because the Probation Office failed to apply §4A1.2(a)(2) of the Guidelines. See USSG §4A1.2(a)(2) (Nov. 2012). That provision addresses how multiple sentences imposed on the same day are to be counted for purposes of determining a defendant’s criminal history. It instructs that, when prior sentences were imposed on the same day, they should be counted as a single sentence unless the offenses “were separated by an intervening arrest (i.e., the defendant is arrested for the first offense prior to committing the second offense).” Ibid. Molina-Martinez’s presentence report included five aggravated burglary convictions for which he had been sentenced on the same day. The Probation Office counted each sentence separately, which resulted in the imposition of 11 criminal history points. Molina-Martinez contended this was error because none of the offenses were separated by an intervening arrest and because he had been sentenced for all five burglaries on the same day. Under a correct calculation, in his view, the burglaries should have resulted in 5 criminal history points instead of 11. That would have lowered his criminal history category from VI to V. The correct criminal history category, in turn, would have resulted in a Guidelines range of 70 to 87 months rather than 77 to 96 months. Had the correct range been used, Molina-Martinez’s 77-month sentence would have been in the middle of the range, not at the bottom. Molina-Martinez acknowledged that, because he did not object in the District Court, he was entitled to relief only if he could satisfy Rule 52(b)’s requirements. He nevertheless maintained relief was warranted because the error was plain, affected his substantial rights, and impugned the fairness, integrity, and public reputation of judicial proceedings. The Court of Appeals disagreed. It held that Molina-Martinez had not established that the District Court’s application of an incorrect Guidelines range affected his substantial rights. It reasoned that, when a correct sentencing range overlaps with an incorrect range, the reviewing court “ ‘do[es] not assume, in the absence of additional evidence, that the sentence [imposed] affects a defendant’s substantial rights.’ ” 588 Fed. Appx. 333, 335 (CA5 2014) (per curiam); see also United States v. Blocker, 612 F. 3d 413, 416 (CA5 2010). Molina-Martinez, the court ruled, had not put forth the additional evidence necessary to show that the error affected his substantial rights. “The mere fact that the court sentenced Molina-Martinez to a low-end sentence,” the Court of Appeals reasoned, “is insufficient on its own to show that Molina-Martinez would have received a similar low-end sentence had the district court used the correct Guidelines range.” 588 Fed. Appx., at 335. Instead, Molina-Martinez needed to identify “ ‘additional evidence’ ” in the record showing that the Guidelines had an effect on the District Court’s selection of his sentence. Ibid. The court noted that “the district court made no explicit statement suggesting that the Guidelines range was a primary factor in sentencing.” Ibid. And the court did not view as probative “the parties’ anchoring of their sentencing arguments in the Guidelines” or “the district court’s refusal to grant the government’s request for a high-end sentence of 96 months.” Ibid. This Court granted certiorari to resolve the disagreement among Courts of Appeals over how to determine whether the application of an incorrect Guidelines range at sentencing affected the defendant’s substantial rights. See 576 U. S. ___ (2015). III The Court of Appeals for the Fifth Circuit stands generally apart from other Courts of Appeals with respect to its consideration of unpreserved Guidelines errors. This Court now holds that its approach is incorrect. Nothing in the text of Rule 52(b), its rationale, or the Court’s precedents supports a requirement that a defendant seeking appellate review of an unpreserved Guidelines error make some further showing of prejudice beyond the fact that the erroneous, and higher, Guidelines range set the wrong framework for the sentencing proceedings. This is so even if the ultimate sentence falls within both the correct and incorrect range. When a defendant is sentenced under an incorrect Guidelines range—whether or not the defendant’s ultimate sentence falls within the correct range—the error itself can, and most often will, be sufficient to show a reasonable probability of a different outcome absent the error. A Today’s holding follows from the essential framework the Guidelines establish for sentencing proceedings. The Court has made clear that the Guidelines are to be the sentencing court’s “starting point and . . . initial benchmark.” Gall v. United States, 552 U. S. 38, 49 (2007) . Federal courts understand that they “ ‘must begin their analysis with the Guidelines and remain cognizant of them throughout the sentencing process.’ ” Peugh, 569 U. S., at ___ (slip op., at 10). The Guidelines are “the framework for sentencing” and “anchor . . . the district court’s discretion.” Id., at ___, ___ (slip op., at 11, 18) “Even if the sentencing judge sees a reason to vary from the Guidelines, ‘if the judge uses the sentencing range as the beginning point to explain the decision to deviate from it, then the Guidelines are in a real sense the basis for the sentence.’ ” Id., at ___ (slip op., at 11). The Guidelines’ central role in sentencing means that an error related to the Guidelines can be particularly serious. A district court that “improperly calculat[es]” a defendant’s Guidelines range, for example, has committed a “significant procedural error.” Gall, supra, at 51. That same principle explains the Court’s ruling that a “retrospective increase in the Guidelines range applicable to a defendant creates a sufficient risk of a higher sentence to constitute an ex post facto violation.” Peugh, 569 U. S., at ___ (slip op., at 13). The Commission’s statistics demonstrate the real and pervasive effect the Guidelines have on sentencing.In most cases district courts continue to impose “either within-Guidelines sentences or sentences that depart downward from the Guidelines on the Government’smotion.” Id., at ___ (slip op., at 12); see USSC, 2014Annual Report and 2014 Sourcebook of Federal Sen-tencing Statistics S–50 (19th ed.) (Table N) (2014 Sourcebook). In less than 20% of cases since 2007 have district courts “imposed above- or below-Guidelines sentences absent a Government motion.” Peugh, supra, at ___–___ (slip op., at 12–13); see also 2011 Annual Report and 2011 Sourcebook of Federal Sentencing Statistics 63 (16th ed.) (Figure G); 2015 Annual Report and 2015 Sourcebook of Federal Sentencing Statistics (20th ed.) (Figure G), on-line at http://www.ussc.gov/sites/default/files/pdf/research-and - publications / annual - reports - and - sourcebooks / 2015/FigureG.pdf (as last visited Apr. 15, 2016). As the Court has recognized, “when a Guidelines range moves up or down, offenders’ sentences [tend to] move with it.” Peugh, supra, at ___ (slip op., at 13); USSC, Final Quarterly Data Report, FY 2014, pp. 32–37 (Figures C to H). These realities have led the Court to observe that there is “consider-able empirical evidence indicating that the Sentencing Guidelines have the intended effect of influencing the sentences imposed by judges.” Peugh, supra, at ___ (slip op., at 12). These sources confirm that the Guidelines are not only the starting point for most federal sentencing proceedings but also the lodestar. The Guidelines inform and instruct the district court’s determination of an appropriate sentence. In the usual case, then, the systemic function of the selected Guidelines range will affect the sentence. This fact is essential to the application of Rule 52(b) to a Guidelines error. From the centrality of the Guidelines in the sentencing process it must follow that, when a defendant shows that the district court used an incorrect range, he should not be barred from relief on appeal simply because there is no other evidence that the sentencing outcome would have been different had the correct range been used. In most cases a defendant who has shown that the district court mistakenly deemed applicable an incorrect, higher Guidelines range has demonstrated a reasonable probability of a different outcome. And, again in most cases, that will suffice for relief if the other requirements of Rule 52(b) are met. There may be instances when, despite application of an erroneous Guidelines range, a reasonable probability of prejudice does not exist. The sentencing process is particular to each defendant, of course, and a reviewing court must consider the facts and circumstances of the case before it. See United States v. Davila, 569 U. S. ___, ___ (2013) (slip op., at 13) (“Our essential point is that particular facts and circumstances matter”). The record in a case may show, for example, that the district court thought the sentence it chose was appropriate irrespective of the Guidelines range. Judges may find that some cases merit a detailed explanation of the reasons the selected sentence is appropriate. And that explanation could make it clear that the judge based the sentence he or she selected on factors independent of the Guidelines. The Government remains free to “poin[t] to parts of the record”—including relevant statements by the judge—“to counter any ostensible showing of prejudice the defendant may make.” United States v. Vonn, 535 U. S. 55, 68 (2002) . Where, however, the record is silent as to what the district court might have done had it considered the correct Guidelines range, the court’s reliance on an incorrect range in most instances will suffice to show an effect on the defendant’s substantial rights. Indeed, in the ordinary case a defendant will satisfy his burden to show prejudice by pointing to the application of an incorrect, higher Guidelines range and the sentence he received thereunder. Absent unusual circumstances, he will not be required to show more. The Court of Appeals’ rule to the contrary fails to take account of the dynamics of federal sentencing. In a significant number of cases the sentenced defendant will lack the additional evidence the Court of Appeals’ rule would require, for sentencing judges often say little about the degree to which the Guidelines influenced their determination. District courts, as a matter of course, use the Guidelines range to instruct them regarding the appropriate balance of the relevant federal sentencing factors. This Court has told judges that they need not provide extensive explanations for within-Guidelines sentences because “[c]ircumstances may well make clear that the judge rests his decision upon the Commission’s own reasoning.” Rita, 551 U. S., at 356–357. In these situations, reviewing courts may presume that a sentence imposed within a properly calculated Guidelines range is reason-able. Id., at 341. As a result, the cases where the Guidelines are most likely to have influenced the district court’s sentencing decision—those where the court chose a sentence within what it believed to be the applicable Guidelines range—are also the cases least likely to provide the defendant with evidence of the Guidelines’ influence beyond the sentence itself. The defendants in these cases should not be prevented by a categorical rule from establishing on appeal that there is a reasonable probability the Guidelines range applied by the sentencing court had an effect on their within-Guidelines sentence. B This case illustrates the unworkable nature of the Court of Appeals’ additional evidence rule. Here the court held that Molina-Martinez could not establish an effect on his substantial rights. Yet the record points to a different conclusion. The District Court said nothing specific about why it chose the sentence it imposed. It merely “adopt[ed] the . . . guideline applications in the presentence investigation report,” App. 33, which set the range at 77 to 96 months; rejected the Government’s argument for a sentence at the top of the Guidelines range; and agreed with the defendant’s request for, and the Probation Office’s recommendation of, a sentence at the bottom of the range. As intended, the Guidelines served as the starting point for the sentencing and were the focal point for the proceedings that followed. The 77-month sentence the District Court selected is conspicuous for its position as the lowest sentence within what the District Court believed to be the applicable range. As Molina-Martinez explained to the Court of Appeals, the District Court’s selection of a sentence at the bottom of the range, despite the Government’s request for the maximum Guidelines sentence, “evinced an intention . . . to give the minimum recommended by the Guidelines.” Brief for Appellant in No. 13–40324 (CA5), p. 18. The District Court said nothing to suggest that it would have imposed a 77-month sentence regardless of the Guidelines range. Given these circumstances, there is at least a reasonable probability that the District Court would have imposed a different sentence had it known that 70 months was in fact the lowest sentence the Commission deemed appropriate. IV The Government contends that permitting a defendant to establish prejudice through the fact of a Guidelines error alone eliminates the main difference between Rules 52(a) and 52(b)—which party must prove whether the complained-of error had an effect. Brief for United States 21. As noted, Rule 52(a) states: “Any error, defect, irregularity, or variance that does not affect substantial rights must be disregarded.” When a defendant makes a timely objection, the Government can rely on Rule 52(a) to argue that the error does not warrant correction because it was harmless. Although Rules 52(a) and (b) both require an inquiry into whether the complained-of error was prejudicial, there is “ ‘one important difference’ ” between the subparts—under (b), but not (a), “ ‘[i]t is the defendant rather than the Government who bears the burden of persuasion with respect to prejudice.’ ” Brief for United States 18 (quoting Olano, 507 U. S., at 734). In the Government’s view, ruling for Molina-Martinez will require the Government to prove the harmlessness of every Guidelines error raised on appeal regardless of whether it was preserved. Brief for United States 27–28. The holding here does not lead to that result. The decision today simply states that courts reviewing sentencing errors cannot apply a categorical rule requiring additional evidence in cases, like this one, where the district court applied an incorrect range but nevertheless sentenced the defendant within the correct range. Rejection of that rule means only that a defendant can rely on the application of an incorrect Guidelines range to show an effect on his substantial rights. The Government expresses concern over the judicial resources needed for the resentencing proceedings that might result from the Court’s holding. It is doubtful today’s holding will result in much of an increased burden. As already noted, today’s holding is consistent with the approach taken by most Courts of Appeals. See, e.g., Sabillon-Umana, 772 F. 3d, at 1333 (collecting cases). Yet only a small fraction of cases are remanded for resentencing because of Guidelines related errors. See 2014 Sourcebook S–6, S–153 (Tables 2 and 62) (of the roughly 75,000 cases sentenced in 2014, only 620 resulted in a remand for resentencing because of a statutory or Guidelines related error). Under the Olano framework, appellate courts retain broad discretion in determining whether a remand for resentencing is necessary. Courts have, for example, developed mechanisms short of a full remand to determine whether a district court in fact would have imposed a different sentence absent the error. See, e.g., United States v. Currie, 739 F. 3d 960, 967 (CA7 2014) (ordering “limited remand so that the district judge [could] consider, and state on the record, whether she would have imposed the same sentence . . . knowing that [the defendant] was subject to a five-year rather than a ten-year statutory minimum term of imprisonment”). And even when a Court of Appeals does decide that resentencing is appropriate, “a remand for resentencing, while not costless, does not invoke the same difficulties as a remand for retrial does.” United States v. Wernick, 691 F. 3d 108, 117–118 (CA2 2012); see also Sabillon-Umana, supra, at 1334 (noting that the “cost of correction is . . . small” because “[a] remand for sentencing . . . doesn’t require that a defendant be released or retried”). The Government’s concern about additional, burdensome procedures appears unfounded, and, in any event, does not warrant reading into Rule 52(b) a requirement that does not exist. * * * In the ordinary case the Guidelines accomplish their purpose. They serve as the starting point for the district court’s decision and anchor the court’s discretion in selecting an appropriate sentence. It follows, then, that in most cases the Guidelines range will affect the sentence. When that is so, a defendant sentenced under an incorrect Guidelines range should be able to rely on that fact to show a reasonable probability that the district court would have imposed a different sentence under the correct range. That probability is all that is needed to establish an effect on substantial rights for purposes of obtaining relief under Rule 52(b). The contrary judgment of the Court of Appeals for the Fifth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

577.US.2015_14-723

Employee benefits plans regulated by the Employee Retirement Income Security Act of 1974 (ERISA or Act) often contain subrogation clauses requiring a plan participant to reimburse the plan for medical expenses if the participant later recovers money from a third party for his injuries. Here, petitioner Montanile was seriously injured by a drunk driver, and his ERISA plan paid more than $120,000 for his medical expenses. Montanile later sued the drunk driver, obtaining a $500,000 settlement. Pursuant to the plan’s subrogation clause, respondent plan administrator (the Board of Trustees of the National Elevator Industry Health Benefit Plan, or Board), sought reimbursement from the settlement. Montanile’s attorney refused that request and subsequently informed the Board that the fund would be transferred from a client trust account to Montanile unless the Board objected. The Board did not respond, and Montanile received the settlement. Six months later, the Board sued Montanile in Federal District Court under §502(a)(3) of ERISA, which authorizes plan fiduciaries to file suit “to obtain . . . appropriate equitable relief . . . to enforce . . . the terms of the plan.” 29 U. S. C. §1132(a)(3). The Board sought an equitable lien on any settlement funds or property in Montanile’s possession and an order enjoining Montanile from dissipating any such funds. Montanile argued that because he had already spent almost all of the settlement, no identifiable fund existed against which to enforce the lien. The District Court rejected Montanile’s argument, and the Eleventh Circuit affirmed, holding that even if Montanile had completely dissipated the fund, the plan was entitled to reimbursement from Montanile’s general assets. Held: When an ERISA-plan participant wholly dissipates a third-party settlement on nontraceable items, the plan fiduciary may not bring suit under §502(a)(3) to attach the participant’s separate assets. Pp. 5–15. (a) Plan fiduciaries are limited by §502(a)(3) to filing suits “to obtain . . . equitable relief.” Whether the relief requested “is legal or equitable depends on [1] the basis for [the plaintiff’s] claim and [2] the nature of the underlying remedies sought.” Sereboff v. Mid Atlantic Medical Services, Inc., 547 U. S. 356 . Pp. 5–9. (1) This Court’s precedents establish that the basis for the Board’s claim—the enforcement of a lien created by an agreement to convey a particular fund to another party—is equitable. See Sereboff, 547 U. S., at 363–364. The Court’s precedents also establish that the nature of the Board’s underlying remedy—enforcement of a lien against “specifically identifiable funds that were within [Montanile’s] possession and control,” id., at 362–363—would also have been equitable had the Board immediately sued to enforce the lien against the fund. But those propositions do not resolve the question here: whether a plan is still seeking an equitable remedy when the defendant has dissipated all of a separate settlement fund, and the plan then seeks to recover out of the defendant’s general assets. Pp. 5–7. (2) This Court holds today that a plan is not seeking equitable relief under those circumstances. In premerger equity courts, a plaintiff could ordinarily enforce an equitable lien, including, as here, an equitable lien by agreement, only against specifically identified funds that remained in the defendant’s possession or against traceable items that the defendant purchased with the funds. See 4 S. Symons, Pomeroy’s Equity Jurisprudence §1234, pp. 692–695. If a defendant dissipated the entire fund on nontraceable items, the lien was eliminated and the plaintiff could not attach the defendant’s general assets instead. See Restatement of Restitution, §215(1), p. 866. Pp. 8–9. (b) The Board’s arguments in favor of the enforcement of an equitable lien against Montanile’s general assets are unsuccessful. Sereboff does not contain an exception to the general asset-tracing requirement for equitable liens by agreement. See 547 U. S., at 365. Nor does historical equity practice support the enforcement of an equitable lien against general assets. And the Board’s claim that ERISA’s objectives are best served by allowing plans to enforce such liens is a “vague notio[n] of [the] statute’s ‘basic purpose’ . . . inadequate to overcome the words of its text regarding the specific issue under consideration.” Mertens v. Hewitt Associates, 508 U. S. 248 . Pp. 9–14. (c) The case is remanded for the District Court to determine, in the first instance, whether Montanile kept his settlement fund separate from his general assets and whether he dissipated the entire fund on nontraceable assets. P. 14. 593 Fed. Appx. 903, reversed and remanded. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Scalia, Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined, and in which Alito, J., joined except for Part III–C. Ginsburg, J., filed a dissenting opinion.

liability for plan-related misdeeds in reasonable proportion to respective actors’ power to control and prevent the misdeeds.” Mertens, supra, at 262. More than a decade has passed since we decided Great-West, and plans have developed safeguards against participants’ and beneficiaries’ efforts to evade reimbursement obligations. Plans that cover medical expenses know how much medical care that participants and beneficiaries require, and have the incentive to investigate and track expensive claims. Plan provisions—like the ones here—obligate participants and beneficiaries to notify the plan of legal process against third parties and to give the plan a right of subrogation. The Board protests that tracking and participating in legal proceedings is hard and costly, and that settlements are often shrouded in secrecy. The facts of this case undercut that argument. The Board had sufficient notice of Montanile’s settlement to have taken various steps to preserve those funds. Most notably, when negotiations broke down and Montanile’s lawyer expressed his intent to disburse the remaining settlement funds to Montanile unless the plan objected within 14 days, the Board could have—but did not—object. Moreover, the Board could have filed suit immediately, rather than waiting half a year. IV Because the lower courts erroneously held that the plan could recover out of Montanile’s general assets, they did not determine whether Montanile kept his settlement fund separate from his general assets or dissipated the entire fund on nontraceable assets. At oral argument, Montanile’s counsel acknowledged “a genuine issue of . . . material fact on how much dissipation there was” anda lack of record evidence as to whether Montanile mixed the settlement fund with his general assets. A remandis necessary so that the District Court can make that determination. * * * We reverse the judgment of the Eleventh Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 * Justice Alito joins this opinion, except for Part III–C. 2 In full, the provision states: “A civil action may be brought— . . . (3) by a participant, beneficiary, or fiduciary (A) to enjoin any act or practice which violates any provision of this subchapter or the terms of the plan, or (B) to obtain other appropriate equitable relief (i) to redress such violations or (ii) to enforce any provisions of this subchapter or the terms of the plan.” 29 U. S. C. §1132(a)(3). 3 Compare Thurber v. Aetna Life Ins. Co., 712 F. 3d 654 (CA2 2013), Funk v. CIGNA Group Ins., 648 F. 3d 182 (CA3 2011), Cusson v. Liberty Life Assurance Co. of Boston, 592 F. 3d 215 (CA1 2010), Longaberger Co. v. Kolt, 586 F. 3d 459 (CA6 2009), and Gutta v. Standard Select Trust Ins. Plans, 530 F. 3d 614 (CA7 2008), with Treasurer, Trustees of Drury Industries, Inc. Health Care Plan & Trust v. Goding, 692 F. 3d 888 (CA8 2012), and Bilyeu v. Morgan Stanley Long Term Disability Plan, 683 F. 3d 1083 (CA9 2012). 4 The Board also interprets CIGNA Corp. v. Amara, 563 U. S. 421 (2011) , as all but overruling Mertens v. Hewitt Associates, 508 U. S. 248 (1993) , and Great-West Life & Annuity Ins. Co. v. Knudson, 534 U. S. 204 (2002) , in favor of the Board’s broad interpretation of “equitable relief ” under §502(a)(3). But CIGNA reaffirmed that “traditionally speaking, relief that sought a lien or a constructive trust was legal relief, not equitable relief, unless the funds in question were ‘particular funds or property in the defendant’s possession.’ ” 563 U. S, at 439 (quoting Great-West, supra, at 213; emphasis deleted). In any event, the Court’s discussion of §502(a)(3) in CIGNA was not essential to resolving that case, and—as our later analysis in US Airways, Inc. v. McCutchen, 569 U. S. ___ (2013), reinforces—our interpretation of “equitable relief” in Mertens, Great-West, and Sereboff v. Mid Atlantic Medical Services, Inc., 547 U. S. 356 (2006) , remains unchanged. See McCutchen, supra, at ___–___ (slip op., at 5–6).

577.US.2015_14-280

Petitioner Montgomery was 17 years old in 1963, when he killed a deputy sheriff in Louisiana. The jury returned a verdict of “guilty without capital punishment,” which carried an automatic sentence of life without parole. Nearly 50 years after Montgomery was taken into custody, this Court decided that mandatory life without parole for juvenile homicide offenders violates the Eighth Amendment’s prohibition on “ ‘cruel and unusual punishments.’ ” Miller v. Alabama, 567 U. S. ___, ___. Montgomery sought state collateral relief, arguing that Miller rendered his mandatory life-without-parole sentence illegal. The trial court denied his motion, and his application for a supervisory writ was denied by the Louisiana Supreme Court, which had previously held that Miller does not have retroactive effect in cases on state collateral review. Held: 1. This Court has jurisdiction to decide whether the Louisiana Supreme Court correctly refused to give retroactive effect to Miller. Pp. 5–14. (a) Teague v. Lane, 489 U. S. 288 , a federal habeas case, set forth a framework for the retroactive application of a new constitutional rule to convictions that were final when the new rule was announced. While the Court held that new constitutional rules of criminal procedure are generally not retroactive, it recognized that courts must give retroactive effect to new watershed procedural rules and to substantive rules of constitutional law. Substantive constitutional rules include “rules forbidding criminal punishment of certain primary conduct” and “rules prohibiting a certain category of punishment for a class of defendants because of their status or offense,” Penry v. Lynaugh, 492 U. S. 302 . Court-appointed amicus contends that because Teague was an interpretation of the federal habeas statute, not a constitutional command, its retroactivity holding has no application in state collateral review proceedings. However, neither Teague nor Danforth v. Minnesota, 552 U. S. 264 —which concerned only Teague’s general retroactivity bar for new constitutional rules of criminal procedure—had occasion to address whether States are required as a constitutional matter to give retroactive effect to new substantive rules. Pp. 5–8. (b) When a new substantive rule of constitutional law controls the outcome of a case, the Constitution requires state collateral review courts to give retroactive effect to that rule. This conclusion is established by precedents addressing the nature of substantive rules, their differences from procedural rules, and their history of retroactive application. As Teague, supra, at 292, 312, and Penry, supra, at 330, indicate, substantive rules set forth categorical constitutional guarantees that place certain criminal laws and punishments altogether beyond the State’s power to impose. It follows that when a State enforces a proscription or penalty barred by the Constitution, the resulting conviction or sentence is, by definition, unlawful. In contrast, where procedural error has infected a trial, a conviction or sentence may still be accurate and the defendant’s continued confinement may still be lawful, see Schriro v. Summerlin, 542 U. S. 348 –353; for this reason, a trial conducted under a procedure found unconstitutional in a later case does not automatically invalidate a defendant’s conviction or sentence. The same possibility of a valid result does not exist where a substantive rule has eliminated a State’s power to proscribe the defendant’s conduct or impose a given punishment. See United States v. United States Coin & Currency, 401 U. S. 715 . By holding that new substantive rules are, indeed, retroactive, Teague continued a long tradition of recognizing that substantive rules must have retroactive effect regardless of when the defendant’s conviction became final; for a conviction under an unconstitutional law “is not merely erroneous, but is illegal and void, and cannot be a legal cause of imprisonment,” Ex parte Siebold, 100 U. S. 371 –377. The same logic governs a challenge to a punishment that the Constitution deprives States of authority to impose, Penry, supra, at 330. It follows that a court has no authority to leave in place a conviction or sentence that violates a substantive rule, regardless of whether the conviction or sentence became final before the rule was announced. This Court’s precedents may not directly control the question here, but they bear on the necessary analysis, for a State that may not constitutionally insist that a prisoner remain in jail on federal habeas review may not constitutionally insist on the same result in its own postconviction proceedings. Pp. 8–14. 2. Miller’s prohibition on mandatory life without parole for juvenile offenders announced a new substantive rule that, under the Constitution, is retroactive in cases on state collateral review. The “foundation stone” for Miller’s analysis was the line of precedent holding certain punishments disproportionate when applied to juveniles, 567 U. S., at ___, n. 4. Relying on Roper v. Simmons, 543 U. S. 551 , and Graham v. Florida, 560 U. S. 48 , Miller recognized that children differ from adults in their “diminished culpability and greater prospects for reform,” 567 U. S., at ___, and that these distinctions “diminish the penological justifications” for imposing life without parole on juvenile offenders, id., at ___. Because Miller determined that sentencing a child to life without parole is excessive for all but “ ‘the rare juvenile offender whose crime reflects irreparable corruption,’ ” id., at ___, it rendered life without parole an unconstitutional penalty for “a class of defendants because of their status”—i.e., juvenile offenders whose crimes reflect the transient immaturity of youth, Penry, 492 U. S., at 330. Miller therefore announced a substantive rule of constitutional law, which, like other substantive rules, is retroactive because it “ ‘necessarily carr[ies] a significant risk that a defendant’ ”—here, the vast majority of juvenile offenders—“ ‘faces a punishment that the law cannot impose upon him.’ ” Schriro, supra, at 352. A State may remedy a Miller violation by extending parole eligibility to juvenile offenders. This would neither impose an onerous burden on the States nor disturb the finality of state convictions. And it would afford someone like Montgomery, who submits that he has evolved from a troubled, misguided youth to a model member of the prison community, the opportunity to demonstrate the truth of Miller’s central intuition—that children who commit even heinous crimes are capable of change. Pp. 14–21. 2013–1163 (La. 6/20/14), 141 So. 3d 264, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Scalia, J., filed a dissenting opinion, in which Thomas and Alito, JJ., joined. Thomas, J., filed a dissenting opinion.

This is another case in a series of decisions involving the sentencing of offenders who were juveniles when their crimes were committed. In Miller v. Alabama, 567 U. S. ___ (2012), the Court held that a juvenile convicted of a homicide offense could not be sentenced to life in prison without parole absent consideration of the juvenile’s special circumstances in light of the principles and purposes of juvenile sentencing. In the wake of Miller, the question has arisen whether its holding is retroactive to juvenile offenders whose convictions and sentences were final when Miller was decided. Courts have reached different conclusions on this point. Compare, e.g., Martin v. Symmes, 782 F. 3d 939, 943 (CA8 2015); Johnson v. Ponton, 780 F. 3d 219, 224–226 (CA4 2015); Chambers v. State, 831 N. W. 2d 311, 331 (Minn. 2013); and State v. Tate, 2012–2763, p. 17 (La. 11/5/13), 130 So. 3d 829, 841, with Diatchenko v. District Attorney for Suffolk Dist., 466 Mass. 655, 661–667, 1 N. E. 3d 270, 278–282 (2013); Aiken v. Byars, 410 S. C. 534, 548, 765 S. E. 2d 572, 578 (2014); State v. Mares, 2014 WY 126, ¶¶47–63, 335 P. 3d 487, 504–508; and People v. Davis, 2014 IL 115595, ¶41, 6 N. E. 3d 709, 722. Certiorari was granted in this case to resolve the question. I Petitioner is Henry Montgomery. In 1963, Montgomery killed Charles Hurt, a deputy sheriff in East Baton Rouge, Louisiana. Montgomery was 17 years old at the time of the crime. He was convicted of murder and sentenced to death, but the Louisiana Supreme Court reversed his conviction after finding that public prejudice had pre-vented a fair trial. State v. Montgomery, 181 So. 2d 756, 762 (La. 1966). Montgomery was retried. The jury returned a verdict of “guilty without capital punishment.” State v. Montgomery, 242 So. 2d 818 (La. 1970). Under Louisiana law, this verdict required the trial court to impose a sentence of life without parole. The sentence was automatic upon the jury’s verdict, so Montgomery had no opportunity to present mitigation evidence to justify a less severe sentence. That evidence might have included Montgomery’s young age at the time of the crime; expert testimony regarding his limited capacity for foresight, self-discipline, and judgment; and his potential for rehabilitation. Montgomery, now 69 years old, has spent almost his entire life in prison. Almost 50 years after Montgomery was first taken into custody, this Court decided Miller v. Alabama, 567 U. S. ___. Miller held that mandatory life without parole for juvenile homicide offenders violates the Eighth Amendment’s prohibition on “ ‘cruel and unusual punishments.’ ” Id., at ___ (slip op., at 2). “By making youth (and all that accompanies it) irrelevant to imposition of that harshest prison sentence,” mandatory life without parole “poses too great a risk of disproportionate punishment.” Id., at ___ (slip op., at 17). Miller required that sentencing courts consider a child’s “diminished culpability and heightened capacity for change” before condemning him or her to die in prison. Ibid. Although Miller did not foreclose a sentencer’s ability to impose life without parole on a juvenile, the Court explained that a lifetime in prison is a disproportionate sentence for all but the rarest of children, those whose crimes reflect “ ‘irreparable corruption.’ ” Ibid. (quoting Roper v. Simmons, 543 U. S. 551, 573 (2005) ). After this Court issued its decision in Miller, Montgomery sought collateral review of his mandatory life-without-parole sentence. In Louisiana there are two principal mechanisms for collateral challenge to the lawfulness of imprisonment. Each begins with a filing in the trial court where the prisoner was convicted and sentenced. La. Code Crim. Proc. Ann., Arts. 882, 926 (West 2008). The first procedure permits a prisoner to file an application for postconviction relief on one or more of seven grounds set forth in the statute. Art. 930.3. The Louisiana Supreme Court has held that none of those grounds provides a basis for collateral review of sentencing errors. See State ex rel. Melinie v. State, 93–1380 (La. 1/12/96), 665 So. 2d 1172 (per curiam). Sentencing errors must instead be raised through Louisiana’s second collateral review procedure. This second mechanism allows a prisoner to bring a collateral attack on his or her sentence by filing a motion to correct an illegal sentence. See Art. 882. Montgomery invoked this procedure in the East Baton Rouge Parish District Court. The state statute provides that “[a]n illegal sentence may be corrected at any time by the court that imposed the sentence.” Ibid. An illegal sentence “is primarily restricted to those instances in which the term of the prisoner’s sentence is not authorized by the statute or statutes which govern the penalty” for the crime of conviction. State v. Mead, 2014–1051, p. 3 (La. App. 4 Cir. 4/22/15), 165 So. 3d 1044, 1047; see also State v. Alexander, 2014–0401 (La. 11/7/14), 152 So. 3d 137 (per curiam). In the ordinary course Louisiana courts will not consider a challenge to a disproportionate sentence on collateral review; rather, as a general matter, it appears that prisoners must raise Eighth Amendment sentencing chal-lenges on direct review. See State v. Gibbs, 620 So. 2d 296, 296–297 (La. App. 1993); Mead, 165 So. 3d, at 1047. Louisiana’s collateral review courts will, however, consider a motion to correct an illegal sentence based on a decision of this Court holding that the Eighth Amendment to the Federal Constitution prohibits a punishment for a type of crime or a class of offenders. When, for example, this Court held in Graham v. Florida, 560 U. S. 48 (2010) , that the Eighth Amendment bars life-without-parole sentences for juvenile nonhomicide offenders, Louisiana courts heard Graham claims brought by prisoners whose sentences had long been final. See, e.g., State v. Shaffer, 2011–1756, pp. 1–4 (La. 11/23/11), 77 So. 3d 939, 940–942 (per curiam) (considering motion to correct an illegal sentence on the ground that Graham rendered illegal a life-without-parole sentence for a juvenile nonhomicide offender). Montgomery’s motion argued that Miller rendered his mandatory life-without-parole sentence illegal. The trial court denied Montgomery’s motion on the ground that Miller is not retroactive on collateral review. Montgomery then filed an application for a supervisory writ. The Louisiana Supreme Court denied the application. 2013–1163 (6/20/14), 141 So. 3d 264. The court relied on its earlier decision in State v. Tate, 2012–2763, 130 So. 3d 829, which held that Miller does not have retroactive effect in cases on state collateral review. Chief Justice Johnson and Justice Hughes dissented in Tate, and Chief Justice Johnson again noted his dissent in Montgomery’s case. This Court granted Montgomery’s petition for certiorari. The petition presented the question “whether Miller adopts a new substantive rule that applies retroactively on collateral review to people condemned as juveniles to die in prison.” Pet. for Cert. i. In addition, the Court directed the parties to address the following question: “Do we have jurisdiction to decide whether the Supreme Court of Louisiana correctly refused to give retroactive effect in this case to our decision in Miller?” 575 U. S. ___ (2015). II The parties agree that the Court has jurisdiction to decide this case. To ensure this conclusion is correct, the Court appointed Richard D. Bernstein as amicus curiae to brief and argue the position that the Court lacks jurisdiction. He has ably discharged his assigned responsibilities. Amicus argues that a State is under no obligation to give a new rule of constitutional law retroactive effect in its own collateral review proceedings. As those proceedings are created by state law and under the State’s plenary control, amicus contends, it is for state courts to define applicable principles of retroactivity. Under this view, the Louisiana Supreme Court’s decision does not implicate a federal right; it only determines the scope of relief avail-able in a particular type of state proceeding—a question of state law beyond this Court’s power to review. If, however, the Constitution establishes a rule and requires that the rule have retroactive application, then a state court’s refusal to give the rule retroactive effect is reviewable by this Court. Cf. Griffith v. Kentucky, 479 U. S. 314, 328 (1987) (holding that on direct review, a new constitutional rule must be applied retroactively “to all cases, state or federal”). States may not disregard a controlling, constitutional command in their own courts. See Martin v. Hunter’s Lessee, 1 Wheat. 304, 340–341, 344 (1816); see also Yates v. Aiken, 484 U. S. 211, 218 (1988) (when a State has not “placed any limit on the issues that it will entertain in collateral proceedings . . . it has a duty to grant the relief that federal law requires”). Amicus’ argument therefore hinges on the premise that this Court’s retroactivity precedents are not a constitutional mandate. Justice O’Connor’s plurality opinion in Teague v. Lane, 489 U. S. 288 (1989) , set forth a framework for retroactiv-ity in cases on federal collateral review. Under Teague, a new constitutional rule of criminal procedure does not apply, as a general matter, to convictions that were final when the new rule was announced. Teague recognized, however, two categories of rules that are not subject to its general retroactivity bar. First, courts must give retroactive effect to new substantive rules of constitutional law. Substantive rules include “rules forbidding criminal punishment of certain primary conduct,” as well as “rules prohibiting a certain category of punishment for a class of defendants because of their status or offense.” Penry v. Lynaugh, 492 U. S. 302, 330 (1989) ; see also Teague, supra, at 307. Although Teague describes new substantive rules as an exception to the bar on retroactive application of procedural rules, this Court has recognized that substantive rules “are more accurately characterized as . . . not subject to the bar.” Schriro v. Summerlin, 542 U. S. 348, 352, n. 4 (2004). Second, courts must give retroactive effect to new “ ‘ “watershed rules of criminal procedure” implicating the fundamental fairness and accuracy of the criminal proceeding.’ ” Id., at 352; see also Teague, 489 U. S., at 312–313. It is undisputed, then, that Teague requires the retroactive application of new substantive and watershed procedural rules in federal habeas proceedings. Amicus, however, contends that Teague was an interpretation of the federal habeas statute, not a constitutional command; and so, the argument proceeds, Teague’s retroactivity holding simply has no application in a State’s own collateral review proceedings. To support this claim, amicus points to language in Teague that characterized the Court’s task as “ ‘defin[ing] the scope of the writ.’ ” Id., at 308 (quoting Kuhlmann v. Wilson, 477 U. S. 436, 447 (1986) (plurality opinion)); see also 489 U. S., at 317 (White, J., concurring in part and concurring in judgment) (“If we are wrong in construing the reach of the habeas corpus statutes, Congress can of course correct us . . . ”); id., at 332 (Brennan, J., dissenting) (“No new facts or arguments have come to light suggesting that our [past] reading of the federal habeas statute . . . was plainly mistaken”). In addition, amicus directs us to Danforth v. Minnesota, 552 U. S. 264 (2008) , in which a majority of the Court held that Teague does not preclude state courts from giving retroactive effect to a broader set of new constitutional rules than Teague itself required. 552 U. S., at 266. The Danforth majority concluded that Teague’s general rule of nonretroactivity for new constitutional rules of criminal procedure “was an exercise of this Court’s power to interpret the federal habeas statute.” 552 U. S., at 278. Since Teague’s retroactivity bar “limit[s] only the scope of federal habeas relief,” the Danforth majority reasoned, States are free to make new procedural rules retroactive on state collateral review. 552 U. S., at 281–282. Amicus, however, reads too much into these statements. Neither Teague nor Danforth had reason to address whether States are required as a constitutional matter to give retroactive effect to new substantive or watershed procedural rules. Teague originated in a federal, not state, habeas proceeding; so it had no particular reason to discuss whether any part of its holding was required by the Constitution in addition to the federal habeas statute. And Danforth held only that Teague’s general rule of nonretroactivity was an interpretation of the federal habeas statute and does not prevent States from providing greater relief in their own collateral review courts. The Danforth majority limited its analysis to Teague’s general retroactivity bar, leaving open the question whether Teague’s two exceptions are binding on the States as a matter of constitutional law. 552 U. S., at 278; see also id., at 277 (“[T]he case before us now does not involve either of the ‘Teague exceptions’ ”). In this case, the Court must address part of the question left open in Danforth. The Court now holds that when a new substantive rule of constitutional law controls the outcome of a case, the Constitution requires state collateral review courts to give retroactive effect to that rule. Teague’s conclusion establishing the retroactivity of new substantive rules is best understood as resting upon constitutional premises. That constitutional command is, like all federal law, binding on state courts. This holding is limited to Teague’s first exception for substantive rules; the constitutional status of Teague’s exception for watershed rules of procedure need not be addressed here. This Court’s precedents addressing the nature of substantive rules, their differences from procedural rules, and their history of retroactive application establish that the Constitution requires substantive rules to have retroactive effect regardless of when a conviction became final. The category of substantive rules discussed in Teague originated in Justice Harlan’s approach to retroactivity. Teague adopted that reasoning. See 489 U. S., at 292, 312 (discussing Mackey v. United States, 401 U. S. 667, 692 (1971) (opinion concurring in judgments in part and dissenting in part); and Desist v. United States, 394 U. S. 244 , n. 2 (1969) (Harlan, J., dissenting)). Justice Harlan defined substantive constitutional rules as “those that place, as a matter of constitutional interpretation, certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe.” Mackey, supra, at 692. In Penry v. Lynaugh, decided four months after Teague, the Court recognized that “the first exception set forth in Teague should be understood to cover not only rules forbidding criminal punishment of certain primary conduct but also rules prohibiting a certain category of punishment for a class of defendants because of their status or offense.” 492 U. S., at 330. Penry explained that Justice Harlan’s first exception spoke “in terms of substantive categorical guarantees accorded by the Constitution, regardless of the procedures followed.” Id., at 329. Whether a new rule bars States from proscribing certain conduct or from inflicting a certain punishment, “[i]n both cases, the Constitution itself deprives the State of the power to impose a certain pen-alty.” Id., at 330. Substantive rules, then, set forth categorical constitutional guarantees that place certain criminal laws and punishments altogether beyond the State’s power to impose. It follows that when a State enforces a proscription or penalty barred by the Constitution, the resulting conviction or sentence is, by definition, unlawful. Procedural rules, in contrast, are designed to enhance the accuracy of a conviction or sentence by regulating “the manner of determining the defendant’s culpability.” Schriro, 542 U. S., at 353; Teague, supra, at 313. Those rules “merely raise the possibility that someone convicted with use of the invalidated procedure might have been acquitted otherwise.” Schriro, supra, at 352. Even where proce-dural error has infected a trial, the resulting conviction or sentence may still be accurate; and, by extension, the defendant’s continued confinement may still be lawful. For this reason, a trial conducted under a procedure found to be unconstitutional in a later case does not, as a general matter, have the automatic consequence of invalidating a defendant’s conviction or sentence. The same possibility of a valid result does not exist where a substantive rule has eliminated a State’s power to proscribe the defendant’s conduct or impose a given punishment. “[E]ven the use of impeccable factfinding procedures could not legitimate a verdict” where “the conduct being penalized is constitutionally immune from punishment.” United States v. United States Coin & Currency, 401 U. S. 715, 724 (1971) . Nor could the use of flawless sentencing procedures legitimate a punishment where the Constitution immunizes the defendant from the sentence imposed. “No circumstances call more for the invocation of a rule of complete retroactivity.” Ibid. By holding that new substantive rules are, indeed, retroactive, Teague continued a long tradition of giving retroactive effect to constitutional rights that go beyond procedural guarantees. See Mackey, supra, at 692–693 (opinion of Harlan, J.) (“[T]he writ has historically been available for attacking convictions on [substantive] grounds”). Before Brown v. Allen, 344 U. S. 443 (1953) , “federal courts would never consider the merits of a constitutional claim if the habeas petitioner had a fair opportunity to raise his arguments in the original proceeding.” Desist, 394 U. S., at 261 (Harlan, J., dissenting). Even in the pre-1953 era of restricted federal habeas, however, an exception was made “when the habeas petitioner attacked the constitutionality of the state statute under which he had been convicted. Since, in this situation, the State had no power to proscribe the conduct for which the petitioner was imprisoned, it could not constitutionally insist that he remain in jail.” Id., at 261, n. 2 (Harlan, J., dissenting) (citation omitted). In Ex parte Siebold, 100 U. S. 371 (1880) , the Court addressed why substantive rules must have retroactive effect regardless of when the defendant’s conviction became final. At the time of that decision, “[m]ere error in the judgment or proceedings, under and by virtue of which a party is imprisoned, constitute[d] no ground for the issue of the writ.” Id., at 375. Before Siebold, the law might have been thought to establish that so long as the conviction and sentence were imposed by a court of competent jurisdiction, no habeas relief could issue. In Siebold, however, the petitioners attacked the judgments on the ground that they had been convicted under unconstitutional statutes. The Court explained that if “this position is well taken, it affects the foundation of the whole proceedings.” Id., at 376. A conviction under an unconstitutional law “is not merely erroneous, but is illegal and void, and cannot be a legal cause of imprisonment. It is true, if no writ of error lies, the judgment may be final, in the sense that there may be no means of reversing it. But . . . if the laws are unconstitutional and void, the Circuit Court acquired no jurisdiction of the causes.” Id., at 376–377. As discussed, the Court has concluded that the same logic governs a challenge to a punishment that the Constitution deprives States of authority to impose. Penry, supra, at 330; see also Friendly, Is Innocence Irrelevant? Collateral Attack on Criminal Judgments, 38 U. Chi. L. Rev. 142, 151 (1970) (“Broadly speaking, the original sphere for collateral attack on a conviction was where the tribunal lacked jurisdiction either in the usual sense or because the statute under which the defendant had been prosecuted was unconstitutional or because the sentence was one the court could not lawfully impose” (footnotes omitted)). A conviction or sentence imposed in violation of a substantive rule is not just erroneous but contrary to law and, as a result, void. See Siebold, 100 U. S., at 376. It follows, as a general principle, that a court has no authority to leave in place a conviction or sentence that violates a substantive rule, regardless of whether the conviction or sentence became final before the rule was announced. Siebold and the other cases discussed in this opinion, of course, do not directly control the question the Court now answers for the first time. These precedents did not involve a state court’s postconviction review of a conviction or sentence and so did not address whether the Constitution requires new substantive rules to have retroactive effect in cases on state collateral review. These decisions, however, have important bearing on the analysis necessary in this case. In support of its holding that a conviction obtained under an unconstitutional law warrants habeas relief, the Siebold Court explained that “[a]n unconstitutional law is void, and is as no law.” Ibid. A penalty imposed pursuant to an unconstitutional law is no less void because the prisoner’s sentence became final before the law was held unconstitutional. There is no grandfather clause that permits States to enforce punishments the Constitution forbids. To conclude otherwise would undercut the Constitution’s substantive guarantees. Writing for the Court in United States Coin & Currency, Justice Harlan made this point when he declared that “[n]o circumstances call more for the invocation of a rule of complete retroactivity” than when “the conduct being penalized is constitutionally immune from punishment.” 401 U. S., at 724. United States Coin & Currency involved a case on direct review; yet, for the reasons explained in this opinion, the same principle should govern the application of substantive rules on collateral review. As Justice Harlan explained, where a State lacked the power to proscribe the habeas petitioner’s conduct, “it could not constitutionally insist that he remain in jail.” Desist, supra, at 261, n. 2 (dissenting opinion). If a State may not constitutionally insist that a prisoner remain in jail on federal habeas review, it may not constitutionally insist on the same result in its own postconviction proceedings. Under the Supremacy Clause of the Constitution, state collateral review courts have no greater power than federal habeas courts to mandate that aprisoner continue to suffer punishment barred by the Constitution. If a state collateral proceeding is open to a claim controlled by federal law, the state court “has a duty to grant the relief that federal law requires.” Yates, 484 U. S., at 218. Where state collateral review proceedings permit prisoners to challenge the lawfulness of their confinement, States cannot refuse to give retroactive effect to a substantive constitutional right that determines the outcome of that challenge. As a final point, it must be noted that the retroactive application of substantive rules does not implicate a State’s weighty interests in ensuring the finality of convictions and sentences. Teague warned against the intrusiveness of “continually forc[ing] the States to marshal resources in order to keep in prison defendants whose trials and appeals conformed to then-existing constitutional standards.” 489 U. S., at 310. This concern has no application in the realm of substantive rules, for no resources marshaled by a State could preserve a conviction or sentence that the Constitution deprives the State of power to impose. See Mackey, 401 U. S., at 693 (opinion of Harlan, J.) (“There is little societal interest in permitting the criminal process to rest at a point where it ought properly never to repose”). In adjudicating claims under its collateral review procedures a State may not deny a controlling right asserted under the Constitution, assuming the claim is properly presented in the case. Louisiana follows these basic Supremacy Clause principles in its postconviction proceedings for challenging the legality of a sentence. The State’s collateral review procedures are open to claims that a decision of this Court has rendered certain sentences illegal, as a substantive matter, under the Eighth Amendment. See, e.g., State v. Dyer, 2011–1758, pp. 1–2 (La. 11/23/11), 77 So. 3d 928, 928–929 (per curiam) (considering claim on collateral review that this Court’s decision in Graham v. Florida, 560 U. S. 48 , rendered petitioner’s life-without-parole sentence illegal). Montgomery alleges that Miller announced a substantive constitutional rule and that the Louisiana Supreme Court erred by failing to recognize its retroactive effect. This Court has jurisdiction to review that determination. III This leads to the question whether Miller’s prohibition on mandatory life without parole for juvenile offenders indeed did announce a new substantive rule that, under the Constitution, must be retroactive. As stated above, a procedural rule “regulate[s] only the manner of determining the defendant’s culpability.” Schriro, 542 U. S., at 353. A substantive rule, in contrast, forbids “criminal punishment of certain primary conduct” or prohibits “a certain category of punishment for a class of defendants because of their status or offense.” Penry, 492 U. S., at 330; see also Schriro, supra, at 353 (A substantive rule “alters the range of conduct or the class of persons that the law punishes”). Under this standard, and for the reasons explained below, Miller announced a substantive rule that is retroactive in cases on collateral review. The “foundation stone” for Miller’s analysis was this Court’s line of precedent holding certain punishments disproportionate when applied to juveniles. 567 U. S., at ___, n. 4 (slip op., at 8, n. 4). Those cases include Graham v. Florida, supra, which held that the Eighth Amendment bars life without parole for juvenile nonhomicide offenders, and Roper v. Simmons, 543 U. S. 551 , which held that the Eighth Amendment prohibits capital punishment for those under the age of 18 at the time of their crimes. Protection against disproportionate punishment is the central substantive guarantee of the Eighth Amendment and goes far beyond the manner of determining a defendant’s sentence. See Graham, supra, at 59 (“The concept of proportionality is central to the Eighth Amendment”);see also Weems v. United States, 217 U. S. 349 (1910); Harmelin v. Michigan, 501 U. S. 957 –998 (1991) (Kennedy, J., concurring in part and concurring in judgment). Miller took as its starting premise the principle established in Roper and Graham that “children are constitutionally different from adults for purposes of sentencing.” 567 U. S., at ___ (slip op., at 8) (citing Roper, supra, at 569–570; and Graham, supra, at 68). These differences result from children’s “diminished culpability and greater prospects for reform,” and are apparent in three primary ways: “First, children have a ‘lack of maturity and an underdeveloped sense of responsibility,’ leading to recklessness, impulsivity, and heedless risk-taking. Second, children ‘are more vulnerable to negative influences and outside pressures,’ including from their family and peers; they have limited ‘control over their own environment’ and lack the ability to extricate themselves from horrific, crime-producing settings. And third, a child’s character is not as ‘well formed’ as an adult’s; his traits are ‘less fixed’ and his actions less likely to be ‘evidence of irretrievable depravity.’ ” 567 U. S., at ___ (slip op., at 8) (quoting Roper, supra, at 569–570; alterations, citations, and some internal quotation marks omitted). As a corollary to a child’s lesser culpability, Miller recognized that “the distinctive attributes of youth diminish the penological justifications” for imposing life without parole on juvenile offenders. 567 U. S., at ___ (slip op., at 9). Because retribution “relates to an offender’s blameworthiness, the case for retribution is not as strong with a minor as with an adult.” Ibid. (quoting Graham, supra, at 71; internal quotation marks omitted). The deterrence rationale likewise does not suffice, since “the same characteristics that render juveniles less culpable than adults—their immaturity, recklessness, and impetuosity—make them less likely to consider potential punishment.” 567 U. S., at ___–___ (slip op., at 9–10) (internal quotation marks omitted). The need for incapacitation is lessened, too, because ordinary adolescent development diminishes the likelihood that a juvenile offender “ ‘forever will be a danger to society.’ ” Id., at ___ (slip op., at 10) (quoting Graham, 560 U. S., at 72). Rehabilitation is not a satisfactory rationale, either. Rehabilitation cannot justify the sentence, as life without parole “forswears altogether the rehabilitative ideal.” 567 U. S., at ___ (slip op., at 10) (quoting Graham, supra, at 74). These considerations underlay the Court’s holding in Miller that mandatory life-without-parole sentences for children “pos[e] too great a risk of disproportionate punishment.” 567 U. S., at ___ (slip op., at 17). Miller requires that before sentencing a juvenile to life without parole, the sentencing judge take into account “how children are different, and how those differences counsel against irrevocably sentencing them to a lifetime in prison.” Ibid. The Court recognized that a sentencer mightencounter the rare juvenile offender who exhibits such irretrievable depravity that rehabilitation is impossible and life without parole is justified. But in light of “children’s diminished culpability and heightened capacity for change,” Miller made clear that “appropriate occasions for sentencing juveniles to this harshest possible penalty will be uncommon.” Ibid. Miller, then, did more than require a sentencer to consider a juvenile offender’s youth before imposing life without parole; it established that the penological justifications for life without parole collapse in light of “the distinctive attributes of youth.” Id., at ___ (slip op., at 9). Even if a court considers a child’s age before sentencing him or her to a lifetime in prison, that sentence still violates the Eighth Amendment for a child whose crime reflects “ ‘unfortunate yet transient immaturity.’ ” Id., at ___ (slip op., at 17) (quoting Roper, 543 U. S., at 573). Because Miller determined that sentencing a child to life without parole is excessive for all but “ ‘the rare juvenile offender whose crime reflects irreparable corruption,’ ” 567 U. S., at ___ (slip op., at 17) (quoting Roper, supra, at 573), it rendered life without parole an unconstitutional penalty for “a class of defendants because of their status”—that is, juvenile offenders whose crimes reflect the transient immaturity of youth. Penry, 492 U. S., at 330. As a result, Miller announced a substantive rule of constitutional law. Like other substantive rules, Miller is retroactive because it “ ‘necessarily carr[ies] a significant risk that a defendant’ ”—here, the vast majority of juvenile offenders—“ ‘faces a punishment that the law cannot impose upon him.’ ” Schriro, 542 U. S., at 352 (quoting Bousley v. United States, 523 U. S. 614, 620 (1998) ). Louisiana nonetheless argues that Miller is procedural because it did not place any punishment beyond the State’s power to impose; it instead required sentencing courts to take children’s age into account before condemning them to die in prison. In support of this argument, Louisiana points to Miller’s statement that the decision “does not categorically bar a penalty for a class of offenders or type of crime—as, for example, we did in Roper or Graham. Instead, it mandates only that a sentencer follow a certain process—considering an offender’s youth and attendant characteristics—before imposing a particular penalty.” Miller, supra, at ___ (slip op., at 20). Miller, it is true, did not bar a punishment for all juvenile offenders, as the Court did in Roper or Graham. Miller did bar life without parole, however, for all but the rarest of juvenile offenders, those whose crimes reflect permanent incorrigibility. For that reason, Miller is no less substantive than are Roper and Graham. Before Miller, every juvenile convicted of a homicide offense could be sentenced to life without parole. After Miller, it will be the rare juvenile offender who can receive that same sentence. The only difference between Roper and Graham, on the one hand, and Miller, on the other hand, is that Miller drew a line between children whose crimes reflect transient immaturity and those rare children whose crimes reflect irreparable corruption. The fact that life without parole could be a proportionate sentence for the latter kind of juvenile offender does not mean that all other children imprisoned under a disproportionate sentence have not suffered the deprivation of a substantive right. To be sure, Miller’s holding has a procedural component. Miller requires a sentencer to consider a juvenile offender’s youth and attendant characteristics before determining that life without parole is a proportionate sentence. See 567 U. S., at ___ (slip op., at 20). Louisiana contends that because Miller requires this process, it must have set forth a procedural rule. This argument, however, conflates a procedural requirement necessary to implement a substantive guarantee with a rule that “regulate[s] only the manner of determining the defendant’s culpability.” Schriro, supra, at 353. There are instances in which a substantive change in the law must be attended by a procedure that enables a prisoner to show that he falls within the category of persons whom the law may no longer punish. See Mackey, 401 U. S., at 692, n. 7 (opinion of Harlan, J.) (“Some rules may have both procedural and substantive ramifications, as I have used those terms here”). For example, when an element of a criminal offense is deemed unconstitutional, a prisoner convicted under that offense receives a new trial where the government must prove the prisoner’s conduct still fits within the modified definition of the crime. In a similar vein, when the Constitution prohibits a particular form of punishment for a class of persons, an affected prisoner receives a procedure through which he can show that he belongs to the protected class. See, e.g., Atkins v. Virginia, 536 U. S. 304, 317 (2002) (requiring a procedure to determine whether a particular individual with an intellectual disability “fall[s] within the range of [intellectually disabled] offenders about whom there is a national consensus” that execution is impermissible). Those procedural requirements do not, of course, transform substantive rules into procedural ones. The procedure Miller prescribes is no different. A hearing where “youth and its attendant characteristics” are considered as sentencing factors is necessary to separate those juveniles who may be sentenced to life without parole from those who may not. 567 U. S., at ___ (slip op., at 1). The hearing does not replace but rather gives effect to Miller’s substantive holding that life without parole is an excessive sentence for children whose crimes reflect transient immaturity. Louisiana suggests that Miller cannot have made a constitutional distinction between children whose crimes reflect transient immaturity and those whose crimes reflect irreparable corruption because Miller did not require trial courts to make a finding of fact regarding a child’s incorrigibility. That this finding is not required, however, speaks only to the degree of procedure Miller mandated in order to implement its substantive guarantee. When a new substantive rule of constitutional law is established, this Court is careful to limit the scope of any attendant procedural requirement to avoid intruding more than necessary upon the States’ sovereign administration of their criminal justice systems. See Ford v. Wainwright, 477 U. S. 399 –417 (1986) (“[W]e leave to the State[s] the task of developing appropriate ways to enforce the constitutional restriction upon [their] execution of sentences”). Fidelity to this important principle of federalism, however, should not be construed to demean the substantive character of the federal right at issue. That Miller did not impose a formal factfinding requirement does not leave States free to sentence a child whose crime reflects transient immaturity to life without parole. To the contrary, Miller established that this punishment is disproportionate under the Eighth Amendment. For this reason, the death penalty cases Louisiana cites in support of its position are inapposite. See, e.g., Beard v. Banks, 542 U. S. 406, 408 (2004) (holding nonretroactive the rule that forbids instructing a jury to disregard mitigating factors not found by a unanimous vote); O’Dell v. Netherland, 521 U. S. 151, 153 (1997) (holding nonretroactive the rule providing that, if the prosecutor cites future dangerousness, the defendant may inform the jury of his ineligibility for parole); Sawyer v. Smith, 497 U. S. 227, 229 (1990) (holding nonretroactive the rule that forbids suggesting to a capital jury that it is not responsible for a death sentence). Those decisions altered the processes in which States must engage before sentencing a person to death. The processes may have had some effect on the likelihood that capital punishment would be imposed, but none of those decisions rendered a certain penalty unconstitutionally excessive for a category of offenders. The Court now holds that Miller announced a substantive rule of constitutional law. The conclusion that Miller states a substantive rule comports with the principles that informed Teague. Teague sought to balance the important goals of finality and comity with the liberty interests of those imprisoned pursuant to rules later deemed unconstitutional. Miller’s conclusion that the sentence of life without parole is disproportionate for the vast majority of juvenile offenders raises a grave risk that many are being held in violation of the Constitution. Giving Miller retroactive effect, moreover, does not require States to relitigate sentences, let alone convictions, in every case where a juvenile offender received mandatory life without parole. A State may remedy a Miller violation by permitting juvenile homicide offenders to be considered for parole, rather than by resentencing them. See, e.g., Wyo. Stat. Ann. §6–10–301(c) (2013) (juvenile homicide offenders eligible for parole after 25 years). Allowing those offenders to be considered for parole ensures that juveniles whose crimes reflected only transient immaturity—and who have since matured—will not be forced to serve a disproportionate sentence in violation of the Eighth Amendment. Extending parole eligibility to juvenile offenders does not impose an onerous burden on the States, nor does it disturb the finality of state convictions. Those prisoners who have shown an inability to reform will continue to serve life sentences. The opportunity for release will be afforded to those who demonstrate the truth of Miller’s central intuition—that children who commit even heinous crimes are capable of change. Petitioner has discussed in his submissions to this Court his evolution from a troubled, misguided youth to a model member of the prison community. Petitioner states that he helped establish an inmate boxing team, of which he later became a trainer and coach. He alleges that he has contributed his time and labor to the prison’s silkscreen department and that he strives to offer advice and serve as a role model to other inmates. These claims have not been tested or even addressed by the State, so the Court does not confirm their accuracy. The petitioner’s sub-missions are relevant, however, as an example of onekind of evidence that prisoners might use to demonstrate rehabilitation. * * * Henry Montgomery has spent each day of the past 46 years knowing he was condemned to die in prison. Perhaps it can be established that, due to exceptional circumstances, this fate was a just and proportionate punishment for the crime he committed as a 17-year-old boy. In light of what this Court has said in Roper, Graham, and Miller about how children are constitutionally different from adults in their level of culpability, however, prisoners like Montgomery must be given the opportunity to show their crime did not reflect irreparable corruption; and, if it did not, their hope for some years of life outside prison walls must be restored. The judgment of the Supreme Court of Louisiana is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.

577.US.2015_14-1095

Petitioner Musacchio resigned as president of Exel Transportation Services (ETS) in 2004, but with help from the former head of ETS’s information-technology department, he accessed ETS’s computer system without ETS’s authorization through early 2006. In November 2010, Musacchio was indicted under 18 U. S. C. §1030(a)(2)(C), which makes it a crime if a person “intentionally accesses a computer without authorization or exceeds authorized access” and thereby “obtains . . . information from any protected computer.” (Emphasis added.) He was charged in count 1 with conspiring to commit both types of improper access and in count 23 with making unauthorized access “[o]n or about” November 24, 2005. In a 2012 superseding indictment, count 1 dropped the charge of conspiracy to exceed authorized access, and count 2 changed count 23’s date to “[o]n or about” November 23–25, 2005. Musacchio never argued in the trial court that his prosecution violated the 5-year statute of limitations applicable to count 2. See §3282(a). At trial, the Government did not object when the District Court instructed the jury that §1030(a)(2)(C) “makes it a crime . . . to intentionally access a computer without authorization and exceed authorized access” (emphasis added), even though the conjunction “and” added an additional element. The jury found Musacchio guilty on counts 1 and 2. On appeal, he challenged the sufficiency of the evidence supporting his conspiracy conviction and argued, for the first time, that his prosecution on count 2 was barred by §3282(a)’s statute of limitations. In affirming his conviction, the Fifth Circuit assessed Musacchio’s sufficiency challenge against the charged elements of the conspiracy count rather than against the heightened jury instruction, and it concluded that he had waived his statute-of-limitations defense by failing to raise it at trial. Held: 1. A sufficiency challenge should be assessed against the elements of the charged crime, not against the elements set forth in an erroneous jury instruction. Sufficiency review essentially addresses whether the Government’s case was strong enough to reach the jury. A reviewing court conducts a limited inquiry tailored to ensuring that a defendant receives the minimum required by due process: a “meaningful opportunity to defend” against the charge against him and a jury finding of guilt “beyond a reasonable doubt.” Jackson v. Virginia, 443 U. S. 307 –315. It does this by considering only the “legal” question “whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” Id., at 319. A reviewing court’s determination thus does not rest on how the jury was instructed. The Government’s failure to introduce evidence of an additional element does not implicate these principles, and its failure to object to a heightened jury instruction does not affect sufficiency review. Because Musacchio does not dispute that he was properly charged with conspiracy to obtain unauthorized access or that the evidence was sufficient to convict him of the charged crime, the Fifth Circuit correctly rejected his sufficiency challenge. Pp. 5–8. 2. A defendant cannot successfully raise §3282(a)’s statute-of-limitations bar for the first time on appeal. Pp. 8–11. (a) A time bar is jurisdictional only if Congress has “clearly state[d]” that it is. Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___. Here, the “text, context, and relevant historical treatment” of §3282(a), Reed Elsevier, Inc. v. Muchnick, 559 U. S. 154 , establish that it imposes a nonjurisdictional defense that becomes part of a case only if a defendant raises it in the district court. The provision does not expressly refer to subject-matter jurisdiction or speak in jurisdictional terms. It thus stands in marked contrast to §3231, which speaks squarely to federal courts’ general criminal subject-matter “jurisdiction” and does not “conditio[n] its jurisdictional grant on” compliance with §3282(a)’s statute of limitations. Id., at 165. The history of §3282(a)’s limitations bar further confirms that the provision does not impose a jurisdictional limit. See United States v. Cook, 17 Wall. 168, 181; Smith v. United States, 568 U. S. ___, ___. Pp. 8–10. (b) Because §3282(a) does not impose a jurisdictional limit, the failure to raise the defense at or before trial is reviewable on appeal—if at all—only for plain error. A district court’s failure to enforce an unraised limitations defense under §3282(a) cannot be a plain error, however, because if a defendant fails to press the defense, it does not become part of the case and, thus, there is no error for an appellate court to correct. Pp. 10–11. 590 Fed. Appx. 359, affirmed. Thomas, J., delivered the opinion for a unanimous Court.

In this case, the Government failed to object to a jury instruction that erroneously added an element that it had to prove, and petitioner failed to press a statute-of-limitations defense until his appeal. We address two questions arising from the parties’ failures to raise timely challenges. We first consider how a court should assess a challenge to the sufficiency of the evidence in a criminal case when a jury instruction adds an element to the charged crime and the Government fails to object. We conclude that the sufficiency of the evidence should be assessed against the elements of the charged crime. We next consider whether the statute-of-limitations defense contained in 18 U. S. C. §3282(a) (the general federal criminal statute of limitations) may be successfully raised for the first time on appeal. We conclude that it may not be. I Petitioner Michael Musacchio served as president of a logistics company, Exel Transportation Services (ETS), until his resignation in 2004. In 2005, he formed a rival company, Total Transportation Services (TTS). Musacchio was soon joined there by Roy Brown, who previouslyheaded ETS’s information-technology department. At TTS, Brown, using a password, continued to access ETS’s computer system without ETS’s authorization. Brown also gave Musacchio access to ETS’s system. This improper access of ETS’s system kept on until early 2006. In November 2010, a grand jury indicted Musacchio under 18 U. S. C. §1030(a)(2)(C). Under that provision, a person commits a crime when he “intentionally accesses a computer without authorization or exceeds authorized access,” and in doing so “obtains . . . information from any protected computer.” (Emphasis added.) The statute thus provides two ways of committing the crime of improperly accessing a protected computer: (1) obtaining access without authorization; and (2) obtaining access with authorization but then using that access improperly. See ibid.; §1030(e)(6) (defining “exceeds authorized access”). Count 1 of the indictment charged Musacchio with conspiring to commit both types of improper access. Count 23 charged him with making unauthorized access to ETS’s e-mail server “[o]n or about” November 24, 2005. App. 70–71.[1] In 2012, the Government filed a superseding indictment amending those charges. Count 1 dropped the charge of conspiracy to exceed authorized access, limiting that charge to conspiracy to make unauthorized access. Count 2 amended the allegations originally contained in count 23 by alleging that Musacchio accessed specific ETS e-mail accounts “[o]n or about” November 23–25, 2005. Id., at 83–84. The Government later filed a second superseding indictment that made no changes relevant here. Musacchio proceeded to a jury trial. At no time before or during trial did he argue that his prosecution violated the 5-year statute of limitations applicable to count 2. See 18 U. S. C. §3282(a) (providing general 5-year statute of limitations). For the Government’s part, it submitted proposed jury instructions on the conspiracy count before and during the trial. Each set of proposed instructions identified that count as involving “Unauthorized Access to Protected Computer[s],” and none required the jury additionally to find that Musacchio conspired to exceed authorized access to protected computers. Musacchio did not propose instructions on the conspiracy count. Diverging from the indictment and the proposed instructions, the District Court instructed the jury on count 1 that §1030(a)(2)(C) “makes it a crime for a person to intentionally access a computer without authorization and exceed authorized access.” App. 168 (emphasis added). The parties agree that this instruction was erroneous: By using the conjunction “and” when referring to both ways of violating §1030(a)(2)(C), the instruction required the Government to prove an additional element. Yet the Government did not object to this error in the instructions. The jury found Musacchio guilty on both counts 1 and 2. The District Court sentenced him to 60 months’ imprisonment. Musacchio appealed, making the two challenges that he again advances in this Court. First, he challenged the sufficiency of the evidence supporting his conspiracy conviction on count 1. He maintained, moreover, that the sufficiency of the evidence should be assessed against the erroneous jury instruction that included the additional element. Second, he argued, for the first time, that his prosecution on count 2—for unauthorized access—was barred by the 5-year statute of limitations because the superseding indictment was filed seven years after the crime and did not relate back to the timely originalindictment. The Fifth Circuit rejected both challenges and affirmed Musacchio’s conviction. 590 Fed. Appx. 359 (2014) ( per curiam). First, the Court of Appeals concluded that it should assess Musacchio’s sufficiency challenge against the charged elements of the conspiracy count, not against the erroneous jury instruction. See id., at 362–363. Under Fifth Circuit precedent, the court explained, erroneously heightened jury instructions generally become the binding “law of the case” on appeal. Id., at 362 (internal quotation marks omitted). Circuit precedent supplies an exception, however, when (1) the jury instruction is “ ‘patently erroneous,’ ” and (2) “ ‘the issue is not misstated in the indictment.’ ” Ibid. (quoting United States v. Guevara, 408 F. 3d 252, 258 (CA5 2005)). The Fifth Circuit concluded that those conditions for applying the exception were satisfied. See 590 Fed. Appx., at 362–363. The court explained that the instruction’s requirement of an additional element was “an obvious clerical error,” and that the indictment correctly charged Musacchio only with “Conspiracy To Make Unauthorized Access to [a] Protected Computer.” Id., at 362. Therefore, the Fifth Circuit did not assess Musacchio’s sufficiency challenge under the heightened jury instruction. Id., at 362–363. Because Musacchio did not dispute that the evidence was sufficient to support a conviction under the elements set out in the indictment, the Fifth Circuit rejected his challenge. Id., at 363. Second, the Fifth Circuit rejected Musacchio’s statute-of-limitations defense, concluding that he had “waived” the defense by failing to raise it at trial. Id., at 363, 364. We granted certiorari to resolve two questions that have divided the lower courts. 576 U. S. ___ (2015). The first question is whether the sufficiency of the evidence in a criminal case should be measured against the elements described in the jury instructions where those instructions, without objection, require the Government to prove more elements than do the statute and indictment. Compare, e.g., United States v. Romero, 136 F. 3d 1268, 1272–1273 (CA10 1998) (explaining that sufficiency is measured against heightened jury instructions), with Guevara, supra, at 258 (CA5) (adopting an exception to that rule). The second question is whether a statute-of-limitations defense not raised at or before trial is reviewable on appeal. Compare, e.g., United States v. Franco-Santiago, 681 F. 3d 1, 12, and n. 18 (CA1 2012) (limitations defense not raised and preserved before or at trial is reviewable on appeal for plain error), with United States v. Walsh, 700 F. 2d 846, 855–856 (CA2 1983) (limitations defense not properly raised below is not reviewable on appeal). II We first address how a court should assess a sufficiency challenge when a jury instruction adds an element to the charged crime and the Government fails to object. We hold that, when a jury instruction sets forth all the elements of the charged crime but incorrectly adds one more element, a sufficiency challenge should be assessed against the elements of the charged crime, not against the erroneously heightened command in the jury instruction. That conclusion flows from the nature of a court’s task in evaluating a sufficiency-of-the-evidence challenge. Sufficiency review essentially addresses whether “the government’s case was so lacking that it should not have even been submitted to the jury.” Burks v. United States, 437 U. S. 1, 16 (1978) (emphasis deleted). On sufficiency review, a reviewing court makes a limited inquiry tailored to ensure that a defendant receives the minimum that due process requires: a “meaningful opportunity to defend” against the charge against him and a jury finding of guilt “beyond a reasonable doubt.” Jackson v. Virginia, 443 U. S. 307 –315 (1979). The reviewing court considers only the “legal” question “whether, after viewing the evidence in the light most favorable to the prosecution, any rational trier of fact could have found the essential elements of the crime beyond a reasonable doubt.” Id., at 319 (emphasis in original). That limited review does not intrude on the jury’s role “to resolve conflicts in the testimony, to weigh the evidence, and to draw reasonable inferences from basic facts to ultimate facts.” Ibid. A reviewing court’s limited determination on sufficiency review thus does not rest on how the jury was instructed. When a jury finds guilt after being instructed on all elements of the charged crime plus one more element, the jury has made all the findings that due process requires. If a jury instruction requires the jury to find guilt on the elements of the charged crime, a defendant will have had a “meaningful opportunity to defend” against the charge. Id., at 314. And if the jury instruction requires the jury to find those elements “beyond a reasonable doubt,” the defendant has been accorded the procedure that this Court has required to protect the presumption of innocence. Id., at 314–315. The Government’s failure to introduce evidence of an additional element does not implicate the principles that sufficiency review protects. All that a defendant is entitled to on a sufficiency challenge is for the court to make a “legal” determination whether the evidence was strong enough to reach a jury at all. Id., at 319. The Government’s failure to object to the heightened jury instruction thus does not affect the court’s review for sufficiency of the evidence.[2] Musacchio does not contest that the indictment here properly charged him with the statutory elements for conspiracy to obtain unauthorized access. The jury instructions required the jury to find all of the elements of that charged offense beyond a reasonable doubt. Nor does he dispute that the evidence was sufficient to convict him of the crime charged in the indictment—of conspiring to make unauthorized access. Accordingly, the Fifth Circuit correctly rejected his sufficiency challenge. The Fifth Circuit erred, however, in basing that conclusion on the law-of-the-case doctrine. See 590 Fed. Appx., at 362–363. That doctrine does not apply here. The law-of-the-case doctrine generally provides that “ ‘when a court decides upon a rule of law, that decision should continue to govern the same issues in subsequent stages in the same case.’ ” Pepper v. United States, 562 U. S. 476, 506 (2011) (quoting Arizona v. California, 460 U. S. 605, 618 (1983) ). The doctrine “expresses the practice of courts generally to refuse to reopen what has been decided,” but it does not “limit [courts’] power.” Messenger v. Anderson, 225 U. S. 436, 444 (1912) . Thus, the doctrine may describe an appellate court’s decision not to depart from a ruling that it made in a prior appeal in the same case. See C. Wright et al., 18B Federal Practice and Procedure §4478, p. 646, and n. 16 (2d ed. 2002) (collecting cases). But the doctrine is “something of a misnomer” when used to describe how an appellate court assesses a lower court’s rulings. United States v. Wells, 519 U. S. 482 , n. 4 (1997). An appellate court’s function is to revisit matters decided in the trial court. When an appellate court reviews a matter on which a party failed to object below, its review may well be constrained by other doctrines such as waiver, forfeiture, and estoppel, as well as by the type of challenge that it is evaluating. But it is not bound by district court rulings under the law-of-the-case doctrine. That doctrine does not bear on how to assess a sufficiency challenge when a jury convicts a defendant after being instructed—without an objection by the Government—on all charged elements of a crime plus an additional element. III We now consider whether a defendant may successfully raise the statute-of-limitations bar in 18 U. S. C. §3282(a) for the first time on appeal. Musacchio argues that he may do so, either because §3282(a) imposes a nonwaivable limit on federal courts’ subject-matter jurisdiction or because a previously unraised limitations claim may constitute plain error that can be noticed on appeal. We disagree with both points, and hold that a defendant cannot successfully raise this statute-of-limitations bar for the first time on appeal. A Statutes of limitations and other filing deadlines “ordinarily are not jurisdictional.” Sebelius v. Auburn Regional Medical Center, 568 U. S. ___, ___ (2013) (slip op., at 8). We treat a time bar as jurisdictional only if Congress has “clearly stated” that it is. Id., at ___ (slip op., at 6–7); (brackets and internal quotation marks omitted); see, e.g., Henderson v. Shinseki, 562 U. S. 428, 436, 439 (2011) (requiring a “clear indication” that a statute is jurisdictional (internal quotation marks omitted)). To determine whether Congress has made the necessary clear statement, we examine the “text, context, and relevant historical treatment” of the provision at issue. Reed Elsevier, Inc. v. Muchnick, 559 U. S. 154, 166 (2010) . Congress has not made such a clear statement here. Rather, the statutory text, context, and history establish that §3282(a) imposes a nonjurisdictional defense that becomes part of a case only if a defendant raises it in the district court. The statutory text suggests that §3282(a) does not impose a jurisdictional limit. Section 3282(a) provides: “Except as otherwise expressly provided by law, no person shall be prosecuted, tried, or punished for any offense, not capital, unless the indictment is found or the information is instituted within five years next after such offense shall have been committed.” Although §3282(a) uses mandatory language, it does not expressly refer to subject-matter jurisdiction or speak in jurisdictional terms. The text of §3282(a) does not, therefore, provide a “clear indication that Congress wanted that provision to be treated as having jurisdictional attributes.” Henderson, supra, at 439. Context confirms that §3282(a) does not impose a jurisdictional limit. Federal courts’ general criminal subject-matter jurisdiction comes from 18 U. S. C. §3231, which states: “The district courts . . . shall have original jurisdiction . . . of all offenses against the laws of the United States.” Section 3231 speaks squarely to federal courts’ “jurisdiction,” in marked contrast to §3282(a), which does not mention “jurisdiction” or a variant of that term. And, nothing in §3231 “conditions its jurisdictional grant on” compliance with §3282(a)’s statute of limitations. Reed Elsevier, supra, at 165. This context supports the conclusion that §3282(a) is not jurisdictional. The history of the limitations bar in §3282(a) demonstrates that it is a defense that becomes part of a case only if the defendant presses it in the district court. This Court held in United States v. Cook, 17 Wall. 168 (1872), that a statute of limitations—identical in all relevant respects to §3282(a)—was “a matter of defence and must be pleaded or given in evidence by the accused.” Id., at 181; see §32, 1Stat. 119 (statute of limitations); see also Cook, supra, at 173, and n. * (citing and describing statute of limitations). When a defendant introduces the limitations defense into the case, the Government then has “the right to reply or give evidence” on the limitations claim. 17 Wall., at 179. Cook was decided more than 140 years ago, and we have adhered to its holding. Just three Terms ago, we reaffirmed that “[c]ommission of [a federal] crime within the statute-of-limitations period is not an element of the . . . offense,” and “it is up to the defendant to raise the limitations defense.” Smith v. United States, 568 U. S. ___, ___ (2013) (slip op., at 6) (citing Cook; emphasis deleted); see also Biddinger v. Commissioner of Police of City of New York, 245 U. S. 128, 135 (1917) (“The statute of limitations is a defense and must be asserted on the trial by the defendant in criminal cases . . . ” (citing Cook)). There is, in sum, a long history of treating the operative language in §3282(a) as providing a nonjurisdictional defense that a defendant must press at trial to insert into the case. In keeping with §3282(a)’s text, context, and history, we conclude that §3282(a) provides a nonjurisdictional defense, not a jurisdictional limit. B Because §3282(a) does not impose a jurisdictional limit, the failure to raise it at or before trial means that it is reviewable on appeal—if at all—only for plain error. See Fed. Rule Crim. Proc. 52(b) (providing for consideration of “[a] plain error that affects substantial rights” even though the error “was not brought to the court’s attention”). We conclude, however, that a district court’s failure to enforce an unraised limitations defense under §3282(a) cannot be a plain error.[3] As explained above, a statute-of-limitations defense becomes part of a case only if the defendant puts the defense in issue. When a defendant presses a limitations defense, the Government then bears the burden of establishing compliance with the statute of limitations by presenting evidence that the crime was committed within the limitations period or by establishing an exception to the limitations period. See Cook, supra, at 179. When a defendant fails to press a limitations defense, the defense does not become part of the case and the Government does not otherwise have the burden of proving that it filed a timely indictment. When a defendant does not press the defense, then, there is no error for an appellate court to correct—and certainly no plain error. A defendant thus cannot successfully raise the statute-of-limitations defense in §3282(a) for the first time on appeal. The Fifth Circuit correctly refused to consider Musacchio’s limitations defense here. * * * For the foregoing reasons, we affirm the judgment of the Fifth Circuit. It is so ordered.Notes 1 Counts 2 through 22 charged other defendants with exceeding authorized access to specific e-mail accounts. App. 68–70. Those defendants pleaded guilty, and later indictments dropped those counts. 2 In resolving the first question presented, we leave open several matters. First, we express no view on the question whether sufficiency of the evidence at trial must be judged by reference to the elements charged in the indictment, even if the indictment charges one or more elements not required by statute. Second, we do not suggest that the Government adds an element to a crime for purposes of sufficiency review when the indictment charges different means of committing a crime in the conjunctive. Third, we also do not suggest that an erroneous jury instruction cannot result in reversible error just because the evidence was sufficient to support a conviction. 3 Because we conclude that the failure to enforce §3282(a)’s limitations defense cannot be plain error, we do not resolve whether the failure to raise that defense in the District Court amounts to waiver (which some courts have held to preclude all appellate review of the defense) or forfeiture (which some courts have held to allow at least plain-error review). See United States v. Franco-Santiago, 681 F. 3d 1, 12, n. 18 (CA1 2012) (collecting cases).

577.US.2015_14-1406

In 1854, the Omaha Tribe entered into a treaty with the United States agreeing to establish a 300,000-acre reservation and to “cede” and “forever relinquish all right and title to” its remaining land in present-day Nebraska for a fixed sum of money. In 1865, the Omaha Tribe again entered into a treaty with the United States agreeing to “cede, sell, and convey” land for a fixed sum. When, in 1872, the Tribe sought to sell more of its land to the United States, Congress took a different tack. In lieu of a fixed-sum purchase, Congress authorized the Secretary of the Interior to survey, appraise, and sell tracts of reservation land to western settlers and to deposit any proceeds from the land sales in the U. S. Treasury for the Tribe’s benefit. Congress took the same approach in 1882 when it passed the Act in question. That Act authorized the Secretary of the Interior to survey, appraise, and sell roughly 50,000 acres of reservation land lying west of a railroad right-of-way. W. E. Peebles purchased a tract under the terms of the 1882 Act and established the village of Pender. In 2006, the Tribe amended its Beverage Control Ordinance and sought to subject Pender retailers to the amended ordinance. See 18 U. S. C. §1161 (permitting tribes to regulate liquor sales on reservation land and in “Indian country”). Pender and its retailers brought a suit against the Tribe in Federal District Court to challenge the ordinance, and the State intervened on their behalf. They alleged that they were not within the reservation boundaries or in Indian country and therefore could not be subject to the ordinance. They sought declaratory relief and a permanent injunction prohibiting the Tribe from asserting its jurisdiction over the disputed land. Concluding that the 1882 Act did not diminish the Omaha Reservation, the District Court denied relief, and the Eighth Circuit affirmed. Held: The 1882 Act did not diminish the Omaha Indian Reservation. Pp. 5–12. (a) Only Congress may diminish the boundaries of an Indian reservation, and its intent to do so must be clear. Solem v. Bartlett, 465 U. S. 463 . This Court’s framework for determining whether an Indian reservation has been diminished is well settled and starts with the statutory text. Hagen v. Utah, 510 U. S. 399 . Here, the 1882 Act bears none of the common textual indications that express such clear intent, e.g., “[e]xplicit reference to cession or other language evidencing the present and total surrender of all tribal interests” or “an unconditional commitment from Congress to compensate the Indian tribe for its opened land,” Solem, supra, at 470. The Act’s language opening the land “for settlement under such rules and regulations as [the Secretary] may prescribe,” 22Stat. 341, falls into a category of surplus land acts that “merely opened reservation land to settlement,” DeCoteau v. District County Court for Tenth Judicial Dist., 420 U. S. 425 . A comparison of the text of the 1854 and 1865 treaties, which unequivocally terminated the Tribe’s jurisdiction over its land, with the 1882 Act confirms this conclusion. Pp. 5–8. (b) In diminishment cases, this Court has also examined “all the circumstances surrounding the opening of a reservation,” Hagen, supra, at 412, including the contemporaneous understanding of the Act’s effect on the reservation. Here, such historical evidence cannot overcome the text of the 1882 Act, which lacks any indication that Congress intended to diminish the reservation. Dueling remarks by legislators about the 1882 Act are far from the unequivocal evidence required in diminishment cases. Pp. 8–10. (c) Finally, and to a lesser extent, the Court may look to subsequent demographic history and subsequent treatment of the land by government officials. See Solem, supra, at 471–472. This Court has never relied solely on this third consideration to find diminishment, and the mixed record of subsequent treatment of the disputed land in this case cannot overcome the statutory text. Petitioners point to the Tribe’s absence from the disputed territory for more than 120 years, but this subsequent demographic history is the “least compelling” evidence in the diminishment analysis. South Dakota v. Yankton Sioux Tribe, 522 U. S. 329 . Likewise, evidence of the subsequent treatment of the disputed land by government officials has similarly limited value. And, while compelling, the justifiable expectations of the non-Indians living on the land cannot alone diminish reservation boundaries. Pp. 10–12. (d) Because the parties have raised only the single question of diminishment, the Court expresses no view about whether equitable considerations of laches and acquiescence may curtail the Tribe’s power to tax the retailers of Pender. Cf. City of Sherrill v. Oneida Indian Nation of N. Y., 544 U. S. 197 –221. P. 12. 774 F. 3d 1166, affirmed. Thomas, J., delivered the opinion for a unanimous Court.

The village of Pender, Nebraska sits a few miles west of an abandoned right-of-way once used by the Sioux City and Nebraska Railroad Company. We must decide whether Pender and surrounding Thurston County, Nebraska,are within the boundaries of the Omaha Indian Reservation or whether the passage of an 1882 Act empowering the United States Secretary of the Interior to sell the Tribe’s land west of the right-of-way “diminished” the reservation’s boundaries, thereby “free[ing]” the disputed land of “its reservation status.” Solem v. Bartlett, 465 U. S. 463, 467 (1984) . We hold that Congress did not diminish the reservation in 1882 and that the disputed land is within the reservation’s boundaries. I A Centuries ago, the Omaha Tribe settled in present-day eastern Nebraska. By the mid-19th century, the Tribe was destitute and, in exchange for much-needed revenue, agreed to sell a large swath of its land to the United States. In 1854, the Tribe entered into a treaty with the United States to create a 300,000-acre reservation. Treaty with the Omahas (1854 Treaty), Mar. 16, 1854, 10Stat. 1043. The Tribe agreed to “cede” and “forever relinquish all right and title to” its land west of the Mississippi River, excepting the reservation, in exchange for $840,000, to be paid over 40 years. Id., at 1043–1044. In 1865, after the displaced Wisconsin Winnebago Tribe moved west, the Omaha Tribe agreed to “cede, sell, and convey” an additional 98,000 acres on the north side of the reservation to the United States for the purpose of creating a reservation for the Winnebagoes. Treaty with the Omaha Indians (1865 Treaty), Mar. 6, 1865, 14Stat. 667–668. The Tribe sold the land for a fixed sum of $50,000. Id., at 667. In 1872, the Tribe again expressed its wish to sell portions of the reservation, but Congress took a different tack than it had in the 1854 and 1865 Treaties. Instead of purchasing a portion of the reservation for a fixed sum, Congress authorized the Secretary of the Interior to survey, appraise, and sell up to 50,000 acres on the western side of the reservation “to be separated from the remaining portion of said reservation” by a north-south line agreed to by the Tribe and Congress. Act of June 10, 1872 (1872 Act), ch. 436, §1, 17Stat. 391. Under the 1872 Act, a nonmember could purchase “tracts not exceeding one hundred and sixty acres each” or “the entire body offered.” Ibid. Proceeds from any sales would be “placed to the credit of said Indians on the books of the treasury of the United States.” Ibid. But the proceeds were meager. The 1872 Act resulted in only two sales totaling 300.72 acres. Then came the 1882 Act, central to the dispute between petitioners and respondents. In that Act, Congress again empowered the Secretary of the Interior “to cause to be surveyed, if necessary, and sold” more than 50,000 acres lying west of a right-of-way granted by the Tribe and approved by the Secretary of the Interior in 1880 for use by the Sioux City and Nebraska Railroad Company. Act of Aug. 7, 1882 (1882 Act), 22Stat. 341. The land for sale under the terms of the 1882 Act overlapped substantially with the land Congress tried, but failed, to sell in 1872. Once the land was appraised “in tracts of forty acres each,” the Secretary was “to issue [a] proclamation” that the “lands are open for settlement under such rules and regulations as he may prescribe.” §§1, 2, id., at 341. Within one year of that proclamation, a nonmember could purchase up to 160 acres of land (for no less than $2.50 per acre) in cash paid to the United States, so long as the settler “occup[ied]” it, made “valuable improvements thereon,” and was “a citizen of the United States, or . . . declared his intention to become such.” §2, id., at 341. The proceeds from any land sales, “after paying all expenses incident to and necessary for carrying out the provisions of th[e] act,” were to “be placed to the credit of said Indians in the Treasury of the United States.” §3, id., at 341. Interest earned on the proceeds was to be “annu-ally expended for the benefit of said Indians, under the direction of the Secretary of the Interior.” Ibid. The 1882 Act also included a provision, common in the late 19th century, that enabled members of the Tribe to select individual allotments, §§5–8, id., at 342–343, as a means of encouraging them to depart from the communal lifestyle of the reservation. See Solem, supra, at 467. The 1882 Act provided that the United States would convey the land to a member or his heirs in fee simple after holding it in trust on behalf of the member and his heirs for 25 years. §6, 22Stat. 342. Members could select allotments on any part of the reservation, either east or west of the right-of-way. §8, id., at 343. After the members selected their allotments—only 10 to 15 of which were located west of the right-of-way—the Secretary proclaimed that the remaining 50,157 acres west of the right-of-way were open for settlement by nonmembers in April 1884. One of those settlers was W. E. Peebles, who “purchased a tract of 160 acres, on which he platted the townsite for Pender.” Smith v. Parker, 996 F. Supp. 2d 815, 828 (Neb. 2014). B The village of Pender today numbers 1,300 residents. Most are not associated with the Omaha Tribe. Less than 2% of Omaha tribal members have lived west of the right-of-way since the early 20th century. Despite its longstanding absence, the Tribe sought to assert jurisdiction over Pender in 2006 by subjecting Pender retailers to its newly amended Beverage Control Ordinance. The ordinance requires those retailers to obtain a liquor license (costing $500, $1,000, or $1,500 depending upon the class of license) and imposes a 10% sales tax on liquor sales. Nonmembers who violate the ordinance are subject to a $10,000 fine. The village of Pender and Pender retailers, including bars, a bowling alley, and social clubs, brought a federal suit against members of the Omaha Tribal Council in their official capacities to challenge the Tribe’s power to impose the requirements of the Beverage Control Ordinance on nonmembers. Federal law permits the Tribe to regulate liquor sales on its reservation and in “Indian country” so long as the Tribe’s regulations are (as they were here) “certified by the Secretary of the Interior, and published in the Federal Register.” 18 U. S. C. §1161. The challengers alleged that they were neither within the boundaries of the Omaha Indian Reservation nor in Indian country and, consequently, were not bound by the ordinance. The State of Nebraska intervened on behalf of the plaintiffs, and the United States intervened on behalf of the Omaha Tribal Council members. The State’s intervention was prompted, in part, by the Omaha Tribe’s demand that Nebraska share with the Tribe revenue that the State received from fuel taxes imposed west of the right-of-way. In addition to the relief sought by Pender and the Pender retailers, Nebraska sought a permanent injunction prohibiting the Tribe from asserting tribal jurisdiction over the 50,157 acres west of the abandoned right-of-way. After examining the text of the 1882 Act, as well as the contemporaneous and subsequent understanding of the 1882 Act’s effect on the reservation boundaries, the District Court concluded that Congress did not diminish the Omaha Reservation in 1882. 996 F. Supp. 2d, at 844. Accordingly, the District Court denied the plaintiffs’ request for injunctive and declaratory relief barring the Tribe’s enforcement of the Beverage Control Ordinance. The Eighth Circuit affirmed. Smith v. Parker, 774 F. 3d 1166, 1168–1169 (2014). We granted certiorari to resolve whether the 1882 Act diminished the Omaha Reservation. 576 U. S. ___ (2015). II We must determine whether Congress “diminished” the Omaha Indian Reservation in 1882. If it did so, the State now has jurisdiction over the disputed land. Solem, 465 U. S., at 467. If Congress, on the other hand, did not diminish the reservation and instead only enabled nonmembers to purchase land within the reservation, then federal, state, and tribal authorities share jurisdiction over these “opened” but undiminished reservation lands. Ibid. The framework we employ to determine whether an Indian reservation has been diminished is well settled. Id., at 470–472. “[O]nly Congress can divest a reservation of its land and diminish its boundaries,” and its intent to do so must be clear. Id., at 470. To assess whether an Act of Congress diminished a reservation, we start with the statutory text, for “[t]he most probative evidence of diminishment is, of course, the statutory language used to open the Indian lands.” Hagen v. Utah, 510 U. S. 399, 411 (1994) . Under our precedents, we also “examine all the circumstances surrounding the opening of a reservation.” Id., at 412. Because of “the turn-of-the-century assumption that Indian reservations were a thing of the past,” many surplus land Acts did not clearly convey “whether opened lands retained reservation status or were divested of all Indian interests.” Solem, supra, at 468. For that reason, our precedents also look to any “unequivocal evidence” of the contemporaneous and subsequent understanding of the status of the reservation by members and nonmembers, as well as the United States and the State of Nebraska. South Dakota v. Yankton Sioux Tribe, 522 U. S. 329, 351 (1998) . A As with any other question of statutory interpretation, we begin with the text of the 1882 Act, the most “probative evidence” of diminishment. Solem, supra, at 470; see, e.g., United States v. Ron Pair Enterprises, Inc., 489 U. S. 235, 241 (1989) (“The task of resolving the dispute over the meaning of [a statutory text] begins where all such inqui-ries must begin: with the language of the statute itself ”). Common textual indications of Congress’ intent to diminish reservation boundaries include “[e]xplicit reference to cession or other language evidencing the present and total surrender of all tribal interests” or “an unconditional commitment from Congress to compensate the Indian tribe for its opened land.” Solem, supra, at 470. Such language “providing for the total surrender of tribal claims in exchange for a fixed payment” evinces Congress’ intent to diminish a reservation, Yankton Sioux, supra, at 345, and creates “an almost insurmountable presumption that Congress meant for the tribe’s reservation to be diminished,” Solem, supra, at 470–471. Similarly, a statutory provision restoring portions of a reservation to “the public domain” signifies diminishment. Hagen, 510 U. S., at 414. In the 19th century, to restore land to the public domain was to extinguish the land’s prior use—its use, for example, as an Indian reservation—and to return it to the United States either to be sold or set aside for other public purposes. Id., at 412–413. The 1882 Act bore none of these hallmarks of diminishment. The 1882 Act empowered the Secretary to survey and appraise the disputed land, which then could be purchased in 160-acre tracts by nonmembers. 22Stat. 341. The 1882 Act states that the disputed lands would be “open for settlement under such rules and regulations as [the Secretary of the Interior] may prescribe.” Ibid. And the parcels would be sold piecemeal in 160-acre tracts. Ibid. So rather than the Tribe’s receiving a fixed sum for all of the disputed lands, the Tribe’s profits were entirely dependent upon how many nonmembers purchased the appraised tracts of land. From this text, it is clear that the 1882 Act falls into another category of surplus land Acts: those that “merely opened reservation land to settlement and provided that the uncertain future proceeds of settler purchases should be applied to the Indians’ benefit.” DeCoteau v. District County Court for Tenth Judicial Dist., 420 U. S. 425 ,448 (1975). Such schemes allow “non-Indian settlers to own land on the reservation.” Seymour v. Superintendent of Wash. State Penitentiary, 368 U. S. 351, 356 (1962) . But in doing so, they do not diminish the reservation’s boundaries. Our conclusion that Congress did not intend to diminish the reservation in 1882 is confirmed by the text of earlier treaties between the United States and the Tribe. See Mattz v. Arnett, 412 U. S. 481, 504 (1973) (comparing statutory text to earlier bills). In drafting the 1882 Act, Congress legislated against the backdrop of the 1854 and 1865 Treaties—both of which terminated the Tribe’s jurisdiction over their land “in unequivocal terms.” Ibid. Those treaties “ced[ed]” the lands and “reliquish[ed]” any claims to them in exchange for a fixed sum. 10Stat. 1043–1044; see also 14Stat. 667 (“The Omaha tribe of Indians do hereby cede, sell, and convey to the United States a tract of land from the north side of their present reservation . . . ” (emphasis added)). The 1882 Act speaks in much different terms, both in describing the way the individual parcels were to be sold to nonmembers and the way in which the Tribe would profit from those sales. That 1882 Act also closely tracks the 1872 Act, which petitioners do not contend diminished the reservation. The change in language in the 1882 Act undermines petitioners’ claim that Congress intended to do the same with the reservation’s boundaries in 1882 as it did in 1854 and 1865. Petitioners have failed at the first and most important step. They cannot establish that the text of the 1882 Act evinced an intent to diminish the reservation. B We now turn to the history surrounding the passage of the 1882 Act. The mixed historical evidence relied upon by the parties cannot overcome the lack of clear textual signal that Congress intended to diminish the reservation. That historical evidence in no way “unequivocally reveal[s] a widely held, contemporaneous understanding that the affected reservation would shrink as a result of the proposed legislation.” Solem, 465 U. S., at 471 (emphasis added); see also Exxon Mobil Corp. v. Allapattah Services, Inc., 545 U. S. 546, 568 (2005) (describing the “often murky, ambiguous, and contradictory” nature of extratextual evidence of congressional intent). Petitioners rely largely on isolated statements that some legislators made about the 1882 Act. Senator Henry Dawes of Massachusetts, for example, noted that he had been “assured that [the 1882 Act] would leave an ample reservation” for the Tribe. 13 Cong. Rec. 3032 (1882) (emphasis added). And Senator John Ingalls of Kansas observed “that this bill practically breaks up that portion at least of the reservation which is to be sold, and provides that it shall be disposed of to private purchasers.” Id., at 3028. Whatever value these contemporaneous floor statements might have, other such statements support the opposite conclusion—that Congress never intended to diminish the reservation. Senator Charles Jones of Flor-ida, for example, spoke of “white men purchas[ing] titles to land within this reservation and settl[ing] down with the Indians on it.” Id., at 3078 (emphasis added). Such dueling remarks by individual legislators are far from the “clear and plain” evidence of diminishment required under this Court’s precedent. Yankton Sioux, 522 U. S., at 343 (internal quotation marks omitted); see also Solem, 465 U. S., at 478 (noting that it was unclear whether statements referring to a “ ‘reduced reservation’ ” alluded to the “reduction in Indian-owned lands that would occur once some of the opened lands were sold to settlers or to the reduction that a complete cession of tribal interests in the opened area would precipitate”). More illuminating than cherry-picked statements by individual legislators would be historical evidence of “the manner in which the transaction was negotiated” with the Omaha Tribe. Id., at 471.[1] In Yankton Sioux, for example, recorded negotiations between the Commissioner of Indian Affairs and leaders of the Yankton Sioux Tribe unambiguously “signaled [the Tribe’s] understanding that the cession of the surplus lands dissolved tribal governance of the 1858 reservation.” 522 U. S., at 353. No such unambiguous evidence exists in the record of these negotiations. In particular, petitioners’ reliance on the remarks of Representative Edward Valentine of Nebraska, who stated, “You cannot find one of those Indians that does not want the western portion sold,” and that the Tribe wished to sell the land to those who would “ ‘reside upon it and cultivate it’ ” so that the Tribe members could “benefit of these improvements,” 13 Cong. Rec. 6541, falls short. Nothing about this statement or other similar statements unequivocally supports a finding that the existing boundaries of the reservation would be diminished. C Finally, we consider both the subsequent demographic history of opened lands, which serves as “one additional clue as to what Congress expected would happen once land on a particular reservation was opened to non-Indian settlers,” Solem, 465 U. S., at 472, as well as the United States’ “treatment of the affected areas, particularly in the years immediately following the opening,” which has “some evidentiary value,” id., at 471. Our cases suggest that such evidence might “reinforc[e]” a finding of diminishment or nondiminishment based on the text. Mattz, 412 U. S., at 505; see also, e.g., Rosebud Sioux Tribe v. Kneip, 430 U. S. 584 –605 (1977) (invoking subsequent history to reject a petitioner’s “strained” textual reading of a congressional Act). But this Court has never relied solely on this third consideration to find diminishment. As petitioners have discussed at length, the Tribe was almost entirely absent from the disputed territory for more than 120 years. Brief for Petitioners 24–30. The Omaha Tribe does not enforce any of its regulations—including those governing businesses, fire protection, animal control, fireworks, and wildlife and parks—in Pender or in other locales west of the right-of-way. 996 F. Supp. 2d, at 832. Nor does it maintain an office, provide social services, or host tribal celebrations or ceremonies west of the right-of-way. Ibid. This subsequent demographic history cannot overcome our conclusion that Congress did not intend to diminish the reservation in 1882. And it is not our role to “rewrite” the 1882 Act in light of this subsequent demographic history. DeCoteau, 420 U. S., at 447. After all, evidence of the changing demographics of disputed land is “the least compelling” evidence in our diminishment analysis, for “[e]very surplus land Act necessarily resulted in a surge of non-Indian settlement and degraded the ‘Indian character’ of the reservation, yet we have repeatedly stated that not every surplus land Act diminished the affected reservation.” Yankton Sioux, 522 U. S., at 356. Evidence of the subsequent treatment of the disputed land by Government officials likewise has “limited interpretive value.” Id., at 355. Petitioners highlight that, for more than a century and with few exceptions, reports from the Office of Indian Affairs and in opinion letters from Government officials treated the disputed land as Nebraska’s. Brief for Petitioners 24–38; see also 996 F. Supp. 2d, at 828, 830. It was not until this litigation commenced that the Department of the Interior definitively changed its position, concluding that the reservation boundaries were in fact not diminished in 1882. See id., at 830–831. For their part, respondents discuss late-19th-century statutes referring to the disputed land as part of the reservation, as well as inconsistencies in maps and statements by Government officials. Brief for Respondent Omaha Tribal Council et al. 45–52; Brief for United States 38–52; see also 996 F. Supp. 2d, at 827, 832–833. This “mixed record” of subsequent treatment of the disputed land cannot overcome the statutory text, which is devoid of any language indicative of Congress’ intent to diminish. Yankton Sioux, supra, at 356. Petitioners’ concerns about upsetting the “justifiable expectations” of the almost exclusively non-Indian settlers who live on the land are compelling, Rosebud Sioux, supra, at 605, but these expectations alone, resulting from the Tribe’s failure to assert jurisdiction, cannot diminish reservation boundaries. Only Congress has the power to diminish a reservation. DeCoteau, 420 U. S., at 449. And though petitioners wish that Congress would have “spoken differently” in 1882, “we cannot remake history.” Ibid. * * * In light of the statutory text, we hold that the 1882 Act did not diminish the Omaha Indian Reservation. Because petitioners have raised only the single question of diminishment,[2] we express no view about whether equitable considerations of laches and acquiescence may curtail the Tribe’s power to tax the retailers of Pender in light of the Tribe’s century-long absence from the disputed lands. Cf. City of Sherrill v. Oneida Indian Nation of N. Y., 544 U. S. 197 –221 (2005). The judgment of the Court of Appeals for the Eighth Circuit is affirmed. It is so ordered.Notes 1 Until this Court’s 1903 decision in Lone Wolf v. Hitchcock, 187 U. S. 553 –568, the question whether Congress could unilaterally abrogate treaties with tribes and divest them of their reservation lands was unsettled. Thus, what the tribe agreed to has been significant in the Court’s diminishment analysis. See, e.g., South Dakota v. Yankton Sioux Tribe, 522 U. S. 329 –353 (1998). Historical evidence of how pre-Lone Wolf sales of lands were negotiated has been deemed compelling, whereas historical evidence of negotiations post-Lone Wolf might be less so. See, e.g., Hagen v. Utah, 510 U. S. 399 –417 (1994). 2 See, e.g., Plaintiff’s Brief in Support of Motion for Summary Judgment in No. 4:07–cv–03101 (D Neb.), pp. 31, 38 (defendants cannot “impose an alcohol tax and licensing scheme outside the boundaries of the Omaha Reservation”); Plaintiff Intervenor’s Brief in Support of Plaintiff’s Motion for Summary Judgment in No. 4:07–cv–03101 (D Neb.), pp. 1–2; see also Smith v. Parker, 996 F. Supp. 2d 815, 834 (Neb. 2014) (“In this case, I must decide whether Congress’s Act of August 7, 1882 . . . diminished the boundaries of the Omaha Indian Reservation, or whether the Act simply permitted non-Indians to settle within existing Omaha Reservation boundaries”); Smith v. Parker, 774 F. 3d 1166, 1167 (CA8 2014) (“Appellants challenge the district court’s determination that the Omaha Indian Reservation was not diminished by an 1882 act of Congress”).

578.US.2015_15-5238

The Sex Offender Registration and Notification Act (SORNA) makes it a federal crime for certain sex offenders to “knowingly fai[l] to register or update a registration,” 18 U. S. C. §2250(a)(3), and requires that offenders who move to a different State “shall, not later than 3 business days after each change of name, residence, employment, or student status,” inform in person “at least 1 jurisdiction involved pursuant to [42 U. S. C. §16913(a)] . . . of all changes” to required information, §16913(c). A §16913(a) jurisdiction is “each jurisdiction where the offender resides, . . . is an employee, and . . . is a student.” Petitioner Nichols, a registered sex offender who moved from Kansas to the Philippines without updating his registration, was arrested, escorted to the United States, and charged with violating SORNA. After conditionally pleading guilty, Nichols argued on appeal that SORNA did not require him to update his registration in Kansas. The Tenth Circuit affirmed his conviction, holding that though Nichols left Kansas, the State remained a “jurisdiction involved” for SORNA purposes. Held: SORNA did not require Nichols to update his registration in Kansas once he departed the State. Pp. 4–8. (a) SORNA’s plain text dictates this holding. Critical here is §16913(a)’s use of the present tense. Nichols once resided in Kansas, but after moving, he “resides” in the Philippines. It follows that once Nichols moved, he was no longer required to appear in Kansas because it was no longer a “jurisdiction involved.” Nor was he required to appear in the Philippines, which is not a SORNA “jurisdiction.” §16911(10). Section 16913(c)’s requirements point to the same conclusion: Nichols could not have appeared in person in Kansas “after” leaving the State. SORNA’s drafters could have required sex offenders to deregister in their departure jurisdiction before leaving the country had that been their intent. Pp. 4–6. (b) The Government resists this straightforward reading. It argues that a jurisdiction where an offender registers remains “involved” even after the offender leaves, but that would require adding the extra clause “where the offender appears on a registry” to §16913(a). Also unconvincing is the claim that §16914(a)(3)’s requiring the offender to provide each address where he “will reside” shows that SORNA contemplates the possibility of an offender’s updating his registration before he actually moves. That provision merely lists the pieces of information to be updated; it says nothing about an obligation to update in the first place. Finally, the Government’s argument that Nichols actually experienced two “changes” of residence—first, when he turned in his apartment keys in Kansas, and second, when he checked into his Manila hotel—is inconsistent with ordinary English usage. Pp. 6–7. (c) Although “the most formidable argument concerning the statute’s purposes [cannot] overcome the clarity [found] in the statute’s text,” Kloeckner v. Solis, 568 U. S. ___, ___, n. 4, the Court is mindful of those purposes and notes that its interpretation is not likely to create deficiencies in SORNA’s scheme. Recent legislation by Congress, as well as existing state-law registration requirements, offers reassurance that sex offenders will not be able to escape punishment for leaving the United States without notifying their departure jurisdictions. Pp. 7–8. 775 F. 3d 1225, reversed. Alito, J., delivered the opinion for a unanimous Court.

Lester Ray Nichols, a registered sex offender living in the Kansas City area, moved to the Philippines without notifying Kansas authorities of his change in residence. For that omission Nichols was convicted of failing to update his sex-offender registration, in violation of 18 U. S. C. §2250(a). We must decide whether federal law required Nichols to update his registration in Kansas to reflect his departure from the State. I A Following the high-profile and horrific rape and murder of 7-year-old Megan Kanka by her neighbor, States in the early 1990’s began enacting registry and community-notification laws to monitor the whereabouts of individuals previously convicted of sex crimes. See Smith v. Doe, 538 U. S. 84, 89 (2003) ; Filler, Making the Case for Megan’s Law, 76 Ind. L. J. 315, 315–317 (2001). Congress followed suit in 1994 with the Jacob Wetterling Crimes Against Children and Sexually Violent Offender Registration Act, 108Stat. 2038, 42 U. S. C. §14071 et seq. (1994 ed.). Named after an 11-year-old who was kidnapped at gunpoint in 1989 (and who remains missing today), the Wetterling Act conditioned federal funds on States’ enacting sex-offender registry laws meeting certain minimum standards. Smith, 538 U. S., at 89–90. “By 1996, every State, the District of Columbia, and the Federal Government had enacted some variation of” a sex-offender registry. Id., at 90. In 2006, Congress replaced the Wetterling Act with the Sex Offender Registration and Notification Act (SORNA), 120Stat. 590, 42 U. S. C. §16901 et seq. Two changes are pertinent here. First, Congress made it a federal crime for a sex offender who meets certain requirements to “knowingly fai[l] to register or update a registration as required by [SORNA].” 18 U. S. C. §2250(a)(3); see Carr v. United States, 560 U. S. 438 –442 (2010). Second, Congress amended the provisions governing the registration requirements when an offender moves to a different State. The original Wetterling Act had directed States to require a sex offender to “register the new address with a designated law enforcement agency in another State to which the person moves not later than 10 days after such person establishes residence in the new State, if the new State has a registration requirement.” 42 U. S. C. §14071(b)(5) (1994 ed.) (emphasis added). Congress later amended this provision to direct States to require a sex offender to “report the change of address to the responsible agency in the State the person is leaving, and [to] comply with any registration requirement in the new State of residence.” 42 U. S. C. §14071(b)(5) (2000 ed.) (emphasis added). SORNA repealed this provision of the Wetterling Act. 120Stat. 600. In its place, federal law now provides: “A sex offender shall, not later than 3 business days after each change of name, residence, employment, or student status, appear in person in at least 1 jurisdiction involved pursuant to subsection (a) and inform that jurisdiction of all changes in the information required for that offender in the sex offender registry.” 42 U. S. C. §16913(c) (emphasis added). Subsection (a), in turn, provides: “A sex offender shall register, and keep the registration current, in each jurisdiction where the offender resides, where the offender is an employee, and where the offender is a student.” §16913(a). A sex offender is required to notify only one “jurisdiction involved”; that jurisdiction must then notify a list of interested parties, including the other jurisdictions. §§16921(b)(1)–(7). The question presented in this case is whether the State a sex offender leaves—that is, the State where he formerly resided—qualifies as an “involved” jurisdiction under §16913. B In 2003, Nichols was convicted of traveling with intent to engage in illicit sexual conduct with a minor, in violation of 18 U. S. C. §2423(b). Although his offense predated SORNA’s enactment, Nichols was nevertheless required upon his eventual release in December 2011 to register as a sex offender in Kansas, where he chose to settle. 28 CFR 72.3 (2015). Nichols complied with SORNA’s registration requirements—until November 9, 2012, when he abruptly disconnected all of his telephone lines, deposited his apartment keys in his landlord’s drop-box, and boarded a flight to Manila. When Nichols was a no-show at mandatory sex-offender treatment, a warrant was issued revoking his supervised release. With the assistance of American security forces, local police in Manila arrested Nichols in December 2012, and federal marshals then escorted him back to the United States, where he was charged with one count of “knowingly fail[ing] to register or update a registration as required by [SORNA],” 18 U. S. C. §2250(a)(3). After unsuccessfully moving to dismiss the indictment on the ground that SORNA did not require him to update his registration in Kansas, Nichols conditionally pleaded guilty, reserving his right to appeal the denial of his motion. The Tenth Circuit affirmed. 775 F. 3d 1225 (2014). Following its own precedent in United States v. Murphy, 664 F. 3d 798 (2011), the panel held that when a sex offender “ ‘leaves a residence in a state, and then leaves the state entirely, that state remains a jurisdiction involved’ ” under §16913. 775 F. 3d, at 1229. Over four dissenting votes, the court denied Nichols’s petition for rehearing en banc. 784 F. 3d 666 (2015). In adhering to Murphy, the Tenth Circuit reentrenched a split created by the Eighth Circuit’s decision in United States v. Lunsford, 725 F. 3d 859 (2013). Remarkably, Lunsford also involved a sex offender who moved from the Kansas City area—on the Missouri side—to the Philippines. Contra the Tenth Circuit’s decision below, Lunsford held that that defendant had no obligation to update his registration in Missouri because a sex offender is required “to ‘keep the registration current’ in the jurisdiction where he ‘resides,’ not a jurisdiction where he ‘resided.’ ” Id., at 861 (citation omitted). We granted certiorari to resolve the split. 577 U. S. ___ (2015). II As noted, Nichols was required to “appear in person in at least 1 jurisdiction involved pursuant to subsection (a) and inform that jurisdiction of” his change of residence. 42 U. S. C. §16913(c). Subsection (a) mentions three possible jurisdictions: “where the offender resides, where the offender is an employee, and where the offender is a student.” §16913(a). The Philippines is not a “jurisdiction” under SORNA; no foreign country is. See §16911(10). Putting these provisions together, SORNA therefore requires a sex offender who changes his residence to appear, within three business days of the change, in person in at least one jurisdiction (but not a foreign country) where he resides, works, or studies, and to inform that jurisdiction of the address change. Critically, §16913(a) uses only the present tense: “resides,” “is an employee,” “is a student.” A person who moves from Leavenworth to Manila no longer “resides” (present tense) in Kansas; although he once resided in Kansas, after his move he “resides” in the Philippines. It follows that once Nichols moved to Manila, he was no longer required to appear in person in Kansas to update his registration, for Kansas was no longer a “jurisdiction involved pursuant to subsection (a)” of §16913. The requirement in §16913(c) to appear in person and register “not later than 3 business days after each change of . . . residence” points to the same conclusion. Nichols could not have appeared in person in Kansas “after” leaving the State. To be sure, one may argue that the day before his departure was “not later than 3 business days after” his departure, but no one in ordinary speech uses language in such a strained and hypertechnical way. If the drafters of SORNA had thought about the problem of sex offenders who leave the country and had sought to require them to (de)register in the departure jurisdiction, they could easily have said so; indeed, that is exactly what the amended Wetterling Act had required. 42 U. S. C. §14071(b)(5) (2000 ed.) (“report the change of address to the responsible agency in the State the person is leaving”). It is also what Kansas state law requires: Nichols had a duty to notify, among other entities, “the registering law enforcement agency or agencies where last registered.” Kan. Stat. Ann. §22–4905(g) (2014 Cum. Supp.) (emphasis added). Congress could have chosen to retain the language in the amended Wetterling Act, or to adopt locution similar to that of the Kansas statute (and echoed in the statutes of many other States, cf. Brief for Petitioner 6, n. 1). It did neither. SORNA’s plain text—in particular, §16913(a)’s consistent use of the present tense—therefore did not require Nichols to update his registration in Kansas once he no longer resided there. III The Government resists this straightforward reading of the statutory text, arguing instead that once an offender registers in a jurisdiction, “that jurisdiction necessarily remains ‘involved pursuant to subsection (a),’ because the offender continues to appear on its registry as a current resident.” Brief for United States 24. But §16913(a) lists only three possibilities for an “involved” jurisdiction: “where the offender resides, where the offender is an employee, and where the offender is a student.” Notably absent is “where the offender appears on a registry.” We decline the Government’s invitation to add an extra clause to the text of §16913(a). As we long ago remarked in another context, “[w]hat the government asks is not a construction of a statute, but, in effect, an enlargement of it by the court, so that what was omitted, presumably by inadvertence, may be included within its scope. To supply omissions transcends the judicial function.” Iselin v. United States, 270 U. S. 245, 251 (1926) . Just so here. Relatedly, the Government points out that among the pieces of information a sex offender must provide as part of his registration is “[t]he address of each residence at which the sex offender resides or will reside.” §16914(a)(3) (emphasis added). The use of the future tense, says the Government, shows that SORNA contemplates the possibility of an offender’s updating his registration before actually moving. But §16914(a) merely lists the pieces of information that a sex offender must provide if and when he updates his registration; it says nothing about whether the offender has an obligation to update his registration in the first place. Finally, the Government argues that Nichols actually experienced not one but two “changes” of residence—the first when he “abandoned” his apartment in Leavenworth by turning in his keys, and the second when he checked into his hotel in Manila. On the Government’s view, a sex offender’s “residence information will change when he leaves the place where he has been residing, and it will change again when he arrives at his new residence. He must report both of those changes in a timely fashion.” Brief for United States 21. We think this argument too clever by half; when someone moves from, say, Kansas City, Kansas, to Kansas City, Missouri, we ordinarily would not say he moved twice: once from Kansas City, Kansas, to a state of homelessness, and then again from homelessness to Kansas City, Missouri. Nor, were he to drive an RV between the cities, would we say that he changed his residence four times (from the house on the Kansas side of the Missouri River to a state of homelessness when he locks the door behind him; then to the RV when he climbs into the vehicle; then back to homelessness when he alights in the new house’s driveway; and then, finally, to the new house in Missouri). And what if he were to move from Kansas to California and spend several nights in hotels along the way? Such ponderings cannot be the basis for imposing criminal punishment. “We interpret criminal statutes, like other statutes, in a manner consistent with ordinary English usage.” Abramski v. United States, 573 U. S. ___, ___ (2014) (Scalia, J., dissenting) (slip op., at 4); Flores-Figueroa v. United States, 556 U. S. 646, 652 (2009) . In ordinary English, Nichols changed his residence just once: from Kansas to the Philippines. We are mindful that SORNA’s purpose was to “make more uniform what had remained ‘a patchwork of federal and 50 individual state registration systems,’ with ‘loopholes and deficiencies’ that had resulted in an estimated 100,000 sex offenders becoming ‘missing’ or ‘lost.’ ” United States v. Kebodeaux, 570 U. S. ___, ___–___ (2013) (slip op., at 11–12) (citation omitted). Yet “even the most formidable argument concerning the statute’s purposes couldnot overcome the clarity we find in the statute’s text.” Kloeckner v. Solis, 568 U. S. ___, ___, n. 4 (2012) (slip op., at 14, n. 4). Our interpretation of the SORNA provisions at issue in this case in no way means that sex offenders will be able to escape punishment for leaving the United States without notifying the jurisdictions in which they lived while in this country. Congress has recently criminalized the “knowin[g] fail[ure] to provide information required by [SORNA] relating to intended travel in foreign commerce.” International Megan’s Law to Prevent Child Exploitation and Other Sexual Crimes Through Advanced Notification of Traveling Sex Offenders, Pub. L. 114–119, §6(b)(2), 130Stat. 23, to be codified at 18 U. S. C. §2250(b). Such information includes “anticipated dates and places of departure, arrival, or return[;] carrier and flight numbers for air travel[;] destination country and address or other contact information therein,” et cetera. §6(a)(1)(B), 130Stat. 22, to be codified at 42 U. S. C. §16914(a)(7). Both parties agree that the new law captures Nichols’s conduct. Supp. Brief for United States 3; Reply Brief 10; Tr. of Oral Arg. 18, 35. And, of course, Nichols’s failure to update his registration in Kansas violated state law. Kan. Stat. Ann. §22–4905(g). We are thus reassured that our holding today is not likely to create “loopholes and deficiencies” in SORNA’s nationwide sex-offender registration scheme. * * * The judgment of the Court of Appeals for the Tenth Circuit is reversed. It is so ordered.

577.US.2015_13-1067

Respondent Carol Sachs, a California resident, purchased a Eurail pass over the Internet from a Massachusetts-based travel agent. While using that pass to board a train in Austria operated by petitioner OBB Personenverkehr AG (OBB), the Austrian state-owned railway, Sachs fell to the tracks and suffered traumatic personal injuries. She sued OBB in Federal District Court. OBB moved to dismiss, claiming that her suit was barred by the Foreign Sovereign Immunities Act, which shields foreign states and their agencies and instrumentalities from suit in United States courts, unless a specified exception applies. Sachs countered that her suit fell within the Act’s commercial activity exception, which abrogates sovereign immunity for suits “based upon a commercial activity carried on in the United States by [a] foreign state,” 28 U. S. C. §1605(a)(2), reasoning that her suit was “based upon” the Massachusetts-based travel agent’s sale of the Eurail pass in the United States, and that the travel agent’s sale of that pass could be attributed to OBB through common law principles of agency. The District Court held that Sachs’s suit did not fall within §1605(a)(2) and dismissed the suit, but the en banc Ninth Circuit reversed. The court first concluded that the Eurail pass sale by the travel agent could be attributed to OBB through common law principles of agency, and then determined that Sachs’s suit was “based upon” that Eurail pass sale because the sale established a single element necessary to recover under each cause of action brought by Sachs. Held: Sachs’s suit falls outside the commercial activity exception and is therefore barred by sovereign immunity. Pp. 5–11. (a) Sachs’s suit is not “based upon” the sale of the Eurail pass for purposes of §1605(a)(2). Therefore, the Court has no need to address whether the Act allows the travel agent’s sale of the Eurail pass to be attributed to OBB through common law principles of agency. Pp. 5–9. (1) Although the Act does not elaborate on the phrase “based upon,” Saudi Arabia v. Nelson, 507 U. S. 349 , provides sufficient guidance to resolve this case. There, the Court held that the “based upon” inquiry requires a court to determine the “particular conduct on which the action is ‘based,’ ” id., at 356, and identified that conduct by looking to “the ‘gravamen of the complaint,’ ” id., at 357. Pp. 5–6. (2) The Ninth Circuit used a flawed approach when it found that the “based upon” inquiry would be satisfied if the sale of the Eurail pass provided “an element” of each of Sachs’s claims. This Court’s approach in Nelson is flatly incompatible with such a one-element approach, which necessarily requires a court to identify all the elements of each claim before finding that the claim falls outside §1605(a)(2). The Nelson Court did not undertake such an exhaustive claim-by-claim, element-by-element analysis or engage in the choice-of-law analysis necessary to such an undertaking. See id., at 356–358. P. 7. (3) As opposed to adopting a one-element test, the Nelson Court zeroed in on the core of the plaintiffs’ suit—the conduct that actually injured the plaintiffs—to identify the conduct that the suit was “based upon.” See id., at 358. All of Sachs’s claims turn on the same tragic episode in Austria, allegedly caused by wrongful conduct and dangerous conditions in Austria, which led to injuries suffered in Austria. However Sachs frames her suit, the incident in Innsbruck, Austria, remains at its foundation. Any other approach would allow plaintiffs to evade the Act’s restrictions through artful pleading. See id., at 363. Pp. 7–9. (b) Sachs now contends that her claims are “based upon” OBB’s entire railway enterprise. Because that argument was never presented to any lower court, it is forfeited. See Taylor v. Freeland & Kronz, 503 U. S. 638 –646. Pp. 9–10. 737 F. 3d 584, reversed. Roberts, C. J., delivered the opinion for a unanimous Court.

The Foreign Sovereign Immunities Act shields foreign states and their agencies from suit in United States courts unless the suit falls within one of the Act’s specifically enumerated exceptions. This case concerns the scope of the commercial activity exception, which withdraws sovereign immunity in any case “in which the action is based upon a commercial activity carried on in the United States by [a] foreign state.” 28 U. S. C. §1605(a)(2). Respondent Carol Sachs is a resident of Californiawho purchased in the United States a Eurail pass forrail travel in Europe. She suffered traumatic personal in-juries when she fell onto the tracks at the Innsbruck, Austria, train station while attempting to board a train operated by the Austrian state-owned railway. She sued the railway in Federal District Court, arguing that her suit was not barred by sovereign immunity because it is “based upon” the railway’s sale of the pass to her in the United States. We disagree and conclude that her action is instead “based upon” the railway’s conduct in Innsbruck. We therefore hold that her suit falls outside the commercial activity exception and is barred by sovereign immunity. I A Petitioner OBB Personenverkehr AG (OBB) operates a railway that carries nearly 235 million passengers each year on routes within Austria and to and from points beyond Austria’s frontiers. OBB is wholly owned by OBB Holding Group, a joint-stock company created by the Republic of Austria. OBB Holding Group in turn is wholly owned by the Austrian Federal Ministry of Transport, Innovation, and Technology. Sachs v. Republic of Austria, 737 F. 3d 584, 587 (CA9 2013). OBB—along with 29 other railways throughout Europe—is a member of the Eurail Group, an association responsible for the marketing and management of the Eurail pass program. Brief for International Rail Transport Committee as Amicus Curiae 12; 737 F. 3d, at 587. Eurail passes allow their holders unlimited passage for a set period of time on participating Eurail Group railways. They are available only to non-Europeans, who may purchase them both directly from the Eurail Group and indirectly through a worldwide network of travel agents. Brief for International Rail Transport Committee as Amicus Curiae 12–13, and n. 3; Brief for Respondent4–5. Carol Sachs is a resident of Berkeley, California. In March 2007, she purchased a Eurail pass over the Internet from The Rail Pass Experts, a Massachusetts-based travel agent. The following month, Sachs arrived at the Innsbruck train station, planning to use her Eurail pass to ride an OBB train to Prague. As she attempted to board the train, Sachs fell from the platform onto the tracks. OBB’s moving train crushed her legs, both of which had to be amputated above the knee. 737 F. 3d, at 587–588. Sachs sued OBB in the United States District Court for the Northern District of California, asserting five causes of action: (1) negligence; (2) strict liability for design defects in the train and platform; (3) strict liability for failure to warn of those design defects; (4) breach of an implied warranty of merchantability for providing a train and platform unsafe for their intended uses; and (5) breach of an implied warranty of fitness for providing a train and platform unfit for their intended uses. App. 14–18. OBB claimed sovereign immunity and moved to dismiss the suit for lack of subject matter jurisdiction. 737 F. 3d, at 588. B The Foreign Sovereign Immunities Act “provides the sole basis for obtaining jurisdiction over a foreign statein the courts of this country.” Argentine Republic v. Amerada Hess Shipping Corp., 488 U. S. 428, 443 (1989) . The Act defines “foreign state” to include a state “agency or instrumentality,” 28 U. S. C. §1603(a), and both parties agree that OBB qualifies as a “foreign state” for purposes of the Act. OBB is therefore “presumptively immune from the jurisdiction of United States courts” unless one of the Act’s express exceptions to sovereign immunity applies. Saudi Arabia v. Nelson, 507 U. S. 349, 355 (1993) . Sachs argues that her suit falls within the Act’s commercial activity exception, which provides in part that a foreign state does not enjoy immunity when “the action is based upon a commercial activity carried on in the United States by the foreign state.” §1605(a)(2).[1] The District Court concluded that Sachs’s suit did not fall within §1605(a)(2) and therefore granted OBB’s motion to dismiss. 2011 WL 816854, *1, *4 (ND Cal., Jan. 28, 2011). A divided panel of the United States Court of Appeals for the Ninth Circuit affirmed. 695 F. 3d 1021 (2012). The full court ordered rehearing en banc and, with three judges dissenting, reversed the panel decision. 737 F. 3d 584. The en banc majority first observed that, “based on the agreement of the parties,” “the only relevant commercial activity within the United States was [Sachs’s] March 2007 purchase of a Eurail pass from the Rail Pass Experts,” a Massachusetts company. Id., at 591, n. 4 (internal quotation marks omitted). The court concluded that The Rail Pass Experts had acted as OBB’s agent and, using common law principles of agency, attributed that Eurail pass sale to OBB. Id., at 591–598. The court next asked whether Sachs’s claims were “based upon” the sale of the Eurail pass within the meaning of §1605(a)(2). The “based upon” determination, the court explained, requires that the commercial activity within the United States be “connected with the conduct that gives rise to the plaintiff’s cause of action.” Id., at 590. But, the court continued, “it is not necessary that the entire claim be based upon the commercial activity of OBB.” Id., at 599. Rather, in the court’s view, Sachs would satisfy the “based upon” requirement for a particular claim “if an element of [that] claim consists in conduct that occurred in commercial activity carried on in the United States.” Ibid. (internal quotation marks omitted). Applying California law, see id., at 600, n. 14, the court analyzed Sachs’s causes of action individually and concluded that the sale of the Eurail pass established a necessary element of each of her claims. Turning first to the negligence claim, the court found that Sachs was required to show that OBB owed her a duty of care as a passenger as one element of that claim. The court concluded that such a duty arose from the sale of the Eurail pass. Id., at 600–602. Turning next to the other claims, the court determined that the existence of a “transaction between a seller and a consumer” was a necessary element of Sachs’s strict liability and breach of implied warranty claims. Id., at 602. The sale of the Eurail pass, the court noted, provided proof of such a transaction. Ibid. Having found that “the sale of the Eurail pass in the United States forms an essential element of each of Sachs’s claims,” the court concluded that each claim was “based upon a commercial activity carried on in the United States” by OBB. Ibid. We granted certiorari. 574 U. S. ___ (2015). II OBB contends that the sale of the Eurail pass is not attributable to the railway, reasoning that the Foreign Sovereign Immunities Act does not allow attribution through principles found in the common law of agency. OBB also argues that even if such attribution were allowed under the Act, Sachs’s suit is not “based upon” the sale of the Eurail pass for purposes of §1605(a)(2). We agree with OBB on the second point and therefore do not reach the first. A The Act itself does not elaborate on the phrase “based upon.” Our decision in Saudi Arabia v. Nelson, 507 U. S. 349 , however, provides sufficient guidance to resolve this case. In Nelson, a husband and wife brought suit against Saudi Arabia and its state-owned hospital, seeking damages for intentional and negligent torts stemming from the husband’s allegedly wrongful arrest, imprisonment, and torture by Saudi police while he was employed at a hospital in Saudi Arabia. Id., at 351, 353–354. The Saudi defendants claimed sovereign immunity under the Act, arguing, inter alia, that §1605(a)(2) was inapplicable because the suit was “based upon” sovereign acts—the exercise of Saudi police authority—and not upon commercial activity. See Brief for Petitioners in Saudi Arabia v. Nelson, O. T. 1992, No. 91–552, pp. 12–14. The Nelsons countered that their suit was “based upon” the defendants’ commercial activities in “recruit[ing] Scott Nelson for work at the hospital, sign[ing] an employment contract with him, and subsequently employ[ing] him.” 507 U. S., at 358. We rejected the Nelsons’ arguments. The Act’s “based upon” inquiry, we reasoned, first requires a court to “identify[ ] the particular conduct on which the [plaintiff’s] action is ‘based.’ ” Id., at 356. Considering dictionary definitions and lower court decisions, we explained that a court should identify that “particular conduct” by looking to the “basis” or “foundation” for a claim, id., at 357 (citing dictionary definitions), “those elements . . . that, if proven, would entitle a plaintiff to relief,” ibid., and “the ‘gravamen of the complaint,’ ” ibid. (quoting Callejo v. Bancomer, S. A., 764 F. 2d 1101, 1109 (CA5 1985)). Under that analysis, we found that the commercial activities, while they “led to the conduct that eventually injured the Nelsons,” were not the particular conduct upon which their suit was based. The suit was instead based upon the Saudi sovereign acts that actually injured them. 507 U. S., at 358. The Nelsons’ suit therefore did not fit within §1605(a)(2). Id., at 361–362. B The Ninth Circuit held that Sachs’s claims were “based upon” the sale of the Eurail pass because the sale of the pass provided “an element” of each of her claims. 737 F. 3d, at 599. Under Nelson, however, the mere fact that the sale of the Eurail pass would establish a single element of a claim is insufficient to demonstrate that the claim is “based upon” that sale for purposes of §1605(a)(2). The Ninth Circuit apparently derived its one-element test from an overreading of one part of one sentence in Nelson, in which we observed that “the phrase [‘based upon’] is read most naturally to mean those elements of a claim that, if proven, would entitle a plaintiff to relief under his theory of the case.” 507 U. S., at 357. We do not see how that mention of elements—plural—could be considered an endorsement of a one-element test, nor how the particular element the Ninth Circuit singled out for each of Sachs’s claims could be construed to entitle her to relief. Be that as it may, our analysis in Nelson is flatly incompatible with a one-element approach. A one-element test necessarily requires a court to identify all the elementsof each claim in a complaint before that court may re-ject those claims for falling outside §1605(a)(2). But we did not undertake such an exhaustive claim-by-claim, element-by-element analysis of the Nelsons’ 16 causes of action, nor did we engage in the choice-of-law analysis that would have been a necessary prelude to such an undertaking. Compare id., at 356–358, with 737 F. 3d, at 600, n. 14 (noting disagreement over whether state or federal common law principles govern suits under the Foreign Sovereign Immunities Act). Nelson instead teaches that an action is “based upon” the “particular conduct” that constitutes the “gravamen” of the suit. Rather than individually analyzing each of the Nelsons’ causes of action, we zeroed in on the core of their suit: the Saudi sovereign acts that actually injured them. As the Court explained: “Even taking each of the Nelsons’ allegations about Scott Nelson’s recruitment and employment as true, those facts alone entitle the Nelsons to nothing under their theory of the case. The Nelsons have . . . alleged . . . personal injuries caused by [the defendants’] intentional wrongs and by [the defendants’] negligent failure to warn Scott Nelson that they might commit those wrongs. Those torts, and not the arguably commercial activities that preceded their commission, form the basis for the Nelsons’ suit.” 507 U. S., at 358. Under this analysis, the conduct constituting the gravamen of Sachs’s suit plainly occurred abroad. All of her claims turn on the same tragic episode in Austria, alleg-edly caused by wrongful conduct and dangerous conditions in Austria, which led to injuries suffered in Austria. Sachs maintains that some of those claims are not limited to negligent conduct or unsafe conditions in Austria, but rather involve at least some wrongful action in the United States. Her strict liability claim for failure to warn, for example, alleges that OBB should have alerted her to the dangerous conditions at the Innsbruck train station when OBB sold the Eurail pass to her in the United States. Under any theory of the case that Sachs pre-sents, however, there is nothing wrongful about the sale of the Eurail pass standing alone. Without the existence of the unsafe boarding conditions in Innsbruck, there would have been nothing to warn Sachs about when she bought the Eurail pass. However Sachs frames her suit, the incident in Innsbruck remains at its foundation. As we explained in Nelson, any other approach would allow plaintiffs to evade the Act’s restrictions through artful pleading. For example, any plaintiff “could recast virtually any claim of intentional tort . . . as a claim of failure to warn, simply by charging the defendant with an obligation to announce its own tortious propensity before indulging it.” Id., at 363. To allow such “recast[ing]” of a complaint, we reasoned, would “give jurisdictional significance to [a] feint of language,” thereby “effectively thwart[ing] the Act’s manifest purpose.” Ibid. A century ago, in a letter to then-Professor Frankfurter, Justice Holmes wrote that the “essentials” of a personal injury narrative will be found at the “point of contact”—“the place where the boy got his fingers pinched.” Letter (Dec. 19, 1915), in Holmes and Frankfurter: Their Correspondence, 1912–1934, p. 40 (R. Mennel & C. Compston eds. 1996). At least in this case, that insight holds true. Regardless of whether Sachs seeks relief under claims for negligence, strict liability for failure to warn, or breach of implied warranty, the “essentials” of her suit for purposes of §1605(a)(2) are found in Austria.[2] III Sachs raises a new argument in this Court in an attempt to fit her claims within §1605(a)(2). In addition to arguing that her claims are “based upon” the sale of the Eurail pass, she now contends that her suit is “based upon” “OBB’s overall commercial railway enterprise.” Brief for Respondent 24; see also Tr. of Oral Arg. 38. “[C]ommercial activity carried on in the United States by the foreign state,” as used in §1605(a)(2), is defined to mean “commercial activity carried on by such state and having substantial contact with the United States.” §1603(e). Sachs’s new theory is that OBB’s entire railway enterprise constitutes the “commercial activity” that has the requisite “substantial contact with the United States,” because OBB reaches out to American customers by marketing and selling Eurail passes in the United States. That argument was never presented to any lower court and is therefore forfeited. Sachs argued in the courts below only that her claims were “based upon” the sale of the Eurail pass, and the lower courts resolved the case on that understanding. See, e.g., 737 F. 3d, at 591, n. 4 (“The district court concluded, based on the agreement of the parties, that ‘the only relevant commercial activity within the United States was plaintiff’s March 2007 purchase of a Eurail Pass from the Rail Pass Experts.’ We consider only the relevant conduct as defined by the district court.”).[3] Indeed, when we granted certiorari, the relevant question presented for our review was whether Sachs’s claims were “based upon” the “sale of the ticket in the United States.” Pet. for Cert. i; accord, Brief for Respondent i. We have answered that question in the negative. Absent unusual circumstances—none of which is present here—we will not entertain arguments not made below. Taylor v. Freeland & Kronz, 503 U. S. 638 –646 (1992). We therefore conclude that Sachs has failed to demonstrate that her suit falls within the commercial activity exception in §1605(a)(2). OBB has sovereign immunity under the Act, and accordingly the courts of the United States lack jurisdiction over the suit. The judgment of the United States Court of Appeals for the Ninth Circuit is reversed. It is so ordered.Notes 1 Section 1605(a)(2) contains three separate clauses. In full, the section provides: 2 We cautioned in Nelson that the reach of our decision was limited, see Saudi Arabia v. Nelson, 507 U. S. 349 , n. 4 (1993), and similar caution is warranted here. Domestic conduct with respect to different types of commercial activity may play a more significant role in other suits under the first clause of §1605(a)(2). In addition, we consider here only a case in which the gravamen of each claim is found in the same place. 3 See also Points and Authorities in Opposition to OBB Personenverkehr AG’s Motion to Dismiss in No. 08–01840 (ND Cal.), p. 8 (“The claims herein are based on the purchase of the Eurail pass.”); Appellant’s Opening Brief in No. 11–15458 (CA9), p. 10 (“[T]he claims are ‘based upon’ the purchase of the ticket which occurred in the United States.”); Appellant’s Reply Brief in No. 11–15458 (CA9), p. 8 (“[H]er claim was based on the purchase/sale of the ticket.”). The District Court decided the case on that understanding of Sachs’s argument. See 2011 WL 816854, *2 (ND Cal., Jan. 28, 2011); see also 2010 WL 4916394, *1 (ND Cal., Nov. 22, 2010). As did the Ninth Circuit panel, see 695 F. 3d 1021, 1024 (2012), and, as noted, the Ninth Circuit en banc. When OBB petitioned this Court for writ of certiorari, Sachs’s brief in opposition repeated her earlier arguments. See Brief in Opposition 2; see also this Court’s Rule 15.2.

578.US.2015_14-361

Petitioner Samuel Ocasio, a former police officer, participated in a kickback scheme in which he and other officers routed damaged vehicles from accident scenes to an auto repair shop in exchange for payments from the shopowners. Petitioner was charged with obtaining money from the shopowners under color of official right, in violation of the Hobbs Act, 18 U. S. C. §1951, and of conspiring to violate the Hobbs Act, in violation of 18 U. S. C. §371. At trial, the District Court rejected petitioner’s argument that—because the Hobbs Act prohibits the obtaining of property “from another”—a Hobbs Act conspiracy requires proof that the alleged conspirators agreed to obtain property from someone outside the conspiracy. Petitioner was convicted on all counts, and the Fourth Circuit affirmed. Petitioner now challenges his conspiracy conviction, contending that he cannot be convicted of conspiring with the shopowners to obtain money from them under color of official right. Held: A defendant may be convicted of conspiring to violate the Hobbs Act based on proof that he reached an agreement with the owner of the property in question to obtain that property under color of official right. Pp. 5–18. (a) The general federal conspiracy statute, under which petitioner was convicted, makes it a crime to “conspire . . . to commit any offense against the United States.” 18 U. S. C. §371. Section 371’s use of the term “conspire” incorporates age-old principles of conspiracy law. And under established case law, the fundamental characteristic of a conspiracy is a joint commitment to an “endeavor which, if completed, would satisfy all of the elements of [the underlying substantive] criminal offense.” Salinas v. United States, 522 U. S. 52 . A conspirator need not agree to commit the substantive offense—or even be capable of committing it—in order to be convicted. It is sufficient that the conspirator agreed that the underlying crime be committed by a member of the conspiracy capable of committing it. See id., at 63–65; United States v. Holte, 236 U. S. 140 ; Gebardi v. United States, 287 U. S. 112 . Pp. 5–10. (b) These basic principles of conspiracy law resolve this case. To establish the alleged Hobbs Act conspiracy, the Government only needed to prove an agreement that some conspirator commit each element of the substantive offense. Petitioner and the shopowners reached just such an agreement: They shared a common purpose that petitioner and other police officers would obtain property “from another”—that is, from the shopowners—under color of official right. Pp. 10–14. (c) Contrary to petitioner’s claims, this decision does not dissolve the distinction between extortion and conspiracy to commit extortion. Nor does it transform every bribe of a public official into a conspiracy to commit extortion. And while petitioner exaggerates the impact of this decision, his argument would create serious practical problems. Under his approach, the validity of a charge of Hobbs Act conspiracy would often depend on difficult property-law questions having little to do with culpability. Pp. 14–18. 750 F. 3d 399, affirmed. Alito, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined. Breyer, J., filed a concurring opinion. Thomas, J., filed a dissenting opinion. Sotomayor, J., filed a dissenting opinion, in which Roberts, C. J., joined.

Petitioner Samuel Ocasio, a former officer in the Baltimore Police Department, participated in a kickback scheme with the owners of a local auto repair shop. When petitioner and other Baltimore officers reported to the scene of an auto accident, they persuaded the owners of damaged cars to have their vehicles towed to the repair shop, and in exchange for this service the officers received payments from the shopowners. Petitioner was convicted of obtaining money from the shopowners under color of official right, in violation of the Hobbs Act, 18 U. S. C. §1951, and of conspiring to violate the Hobbs Act, in violation of 18 U. S. C. §371. He now challenges his conspiracy conviction, contending that, as a matter of law, he cannot be convicted of conspiring with the shopowners to obtain money from them under color of official right. We reject this argument because it is contrary to age-old principles of conspiracy law. I Hernan Alexis Moreno Mejia (known as Moreno) and Edwin Javier Mejia (known as Mejia) are brothers who co-owned and operated the Majestic Auto Repair Shop (Majestic). In 2008, Majestic was struggling to attract customers, so Moreno and Mejia made a deal with a Baltimore police officer, Jhonn Corona. In exchange for kickbacks, Officer Corona would refer motorists whose cars were damaged in accidents to Majestic for towing and repairs. Officer Corona then spread the word to other members of the force, and eventually as many as 60 other officers sent damaged cars to Majestic in exchange for payments of $150 to $300 per referral. Petitioner began to participate in this scheme in 2009. On several occasions from 2009 to 2011, he convinced accident victims to have their cars towed to Majestic. Often, before sending a car to Majestic, petitioner called Moreno from the scene of an accident to ensure that the make and model of the car, the extent of the damage, and the car’s insurance coverage would allow the shopowners to turn a profit on the repairs. After directing a vehicle to Majestic, petitioner would call Moreno and request his payment. Because police are often among the first to arrive at the scene of an accident, the Baltimore officers were well positioned to route damaged vehicles to Majestic. As a result, the kickback scheme was highly successful: It substantially increased Majestic’s volume of business and profits, and by early 2011 it provided Majestic with at least 90% of its customers. Moreno, Mejia, petitioner, and nine other Baltimore officers were indicted in 2011. The shopowners and most of the other officers eventually pleaded guilty pursuant to plea deals, but petitioner did not. In a superseding indictment, petitioner was charged with three counts of violating the Hobbs Act, 18 U. S. C. §1951, by extorting money from Moreno with his consent and under color of official right. As all parties agree, the type of extortion for which petitioner was convicted—obtaining property from another with his consent and under color of official right—is the “rough equivalent of what we would now describe as ‘taking a bribe.’ ” Evans v. United States, 504 U. S. 255, 260 (1992) . To prove this offense, the Government “need only show that a public official has obtained a payment to which he was not entitled, knowing that the payment was made in return for official acts.” Id., at 268. Petitioner and another Baltimore officer, Kelvin Quade Manrich, were also charged with violating the general federal conspiracy statute, 18 U. S. C. §371. The indictment alleged that petitioner and Manrich conspired with Moreno, Mejia, and other Baltimore officers to bring about the same sort of substantive violations with which petitioner was charged. Before trial, petitioner began to raise a variant of the legal argument that has brought his case to this Court. He sought a jury instruction stating that “[i]n order to convict a defendant of conspiracy to commit extortion under color of official right, the government must prove beyond a reasonable doubt that the conspiracy was to obtain money or property from some person who was not a member of the conspiracy.” App. 53. In support of this instruction, petitioner relied on the Sixth Circuit’s decision in United States v. Brock, 501 F. 3d 762 (2007), which concerned two bail bondsmen who made payments to a court clerk in exchange for the alteration of court records. The Sixth Circuit held that “[t]o be covered by the [Hobbs Act], the alleged conspirators . . . must have formed an agreement to obtain ‘property from another,’ which is to say, formed an agreement to obtain property from someone outside the conspiracy.” Id., at 767. The District Court did not rule on this request prior to trial. Petitioner’s codefendant, Manrich, pleaded guilty during the trial, and at the close of the prosecution’s case and again at the close of all evidence, petitioner moved for a judgment of acquittal on the conspiracy count based on Brock. The District Court denied these motions, concluding that the Fourth Circuit had already rejected Brock’s holding in United States v. Spitler, 800 F. 2d 1267 (1986). The District Court also refused to give petitioner’s proposed instruction. Instead, the court adopted the sort of standard instructions that are typically used in conspiracy cases. See generally L. Sand et al., Modern Federal Jury Instructions: Criminal §19.01 (2015). In order to convict petitioner of the conspiracy charge, the jury was told, the prosecution was required to prove (1) that two or more persons entered into an unlawful agreement; (2) that petitioner knowingly and willfully became a member of the conspiracy; (3) that at least one member of the conspiracy knowingly committed at least one overt act; and (4) that the overt act was committed to further an objective of the conspiracy. The court “caution[ed]” “that mere knowledge or acquiescence, without participation in the unlawful plan, is not sufficient” to demonstrate membership in the conspiracy. App. 195. Rather, the court explained, the conspirators must have had “a mutual understanding . . . to cooperate with each other to accomplish an unlawful act,” and petitioner must have joined the conspiracy “with the intention of aiding in the accomplishment of those unlawful ends.” Id., at 192, 195. The jury found petitioner guilty on both the conspiracy count and the three substantive extortion counts, and the District Court sentenced him to concurrent terms of 18 months in prison on all four counts. On appeal to the Fourth Circuit, petitioner’s primary argument was the same one he had pressed before the District Court: that his conspiracy conviction was fatally flawed because the conspirators had not agreed to obtain money from a person who was not a member of the conspiracy. The Fourth Circuit rejected petitioner’s argument and affirmed his convictions. 750 F. 3d 399 (2014). We then granted certiorari, 574 U. S. ___ (2015), and we now affirm. II Under longstanding principles of conspiracy law, a defendant may be convicted of conspiring to violate the Hobbs Act based on proof that he entered into a conspiracy that had as its objective the obtaining of property from another conspirator with his consent and under color of official right. A In analyzing petitioner’s arguments, we begin with the text of the statute under which he was convicted, namely, the general federal conspiracy statute, which makes it a crime to “conspire . . . to commit any offense against the United States.” 18 U. S. C. §371 (emphasis added). Section 371’s use of the term “conspire” incorporates long-recognized principles of conspiracy law. And under established case law, the fundamental characteristic of aconspiracy is a joint commitment to an “endeavor which,if completed, would satisfy all of the elements of [the underlying substantive] criminal offense.” Salinas v. United States, 522 U. S. 52, 65 (1997) ; see 2 J. Bishop, Commentaries on the Criminal Law §175, p. 100 (rev. 7th ed. 1882) (“Conspiracy, in the modern law, is generally defined as a confederacy of two or more persons to accomplish some unlawful purpose”); J. Hawley & M. McGregor, The Criminal Law 99–100 (3d ed. 1899) (similar); W. LaFave, Criminal Law 672 (5th ed. 2010) (similar). Although conspirators must “pursue the same criminal objective,” “a conspirator [need] not agree to commit or facilitate each and every part of the substantive offense.” Salinas, supra, at 63. A defendant must merely reach an agreement with the “specific intent that the underlying crime be committed” by some member of the conspiracy. 2 K. O’Malley, J. Grenig, & W. Lee, Federal Jury Practice and Instructions: Criminal §31:03, p. 225 (6th ed. 2008) (emphasis added); see also id., §31:02, at 220 (explaining that a defendant must “intend to agree and must intend that the substantive offense be committed” (emphasis added)). “The government does not have to prove that the defendant intended to commit the underlying offense himself/herself.” Id., §31:03, at 226. Instead, “[i]f conspirators have a plan which calls for some conspirators to perpetrate the crime and others to provide support, the supporters are as guilty as the perpetrators.” Salinas, supra, at 64; see Sand, supra, §19.01, at 19–54 (“[W]hen people enter into a conspiracy to accomplish an unlawful end, each and every member becomes an agent for the other conspirators in carrying out the conspiracy”). A few simple examples illustrate this important point. Entering a dwelling is historically an element of burglary, see, e.g., LaFave, supra, at 1069, but a person may conspire to commit burglary without agreeing to set foot inside the targeted home. It is enough if the conspirator agrees to help the person who will actually enter the dwelling, perhaps by serving as a lookout or driving the getaway car. Likewise, “[a] specific intent to distribute drugs oneself is not required to secure a conviction for participating in a drug-trafficking conspiracy.” United States v. Piper, 35 F. 3d 611, 614 (CA1 1994). Agreeing to store drugs at one’s house in support of the conspiracy may be sufficient. Ibid. Not only is it unnecessary for each member of a conspiracy to agree to commit each element of the substantive offense, but also a conspirator may be convicted “even though he was incapable of committing the substantive offense” himself. Salinas, supra, at 64; see United States v. Rabinowich, 238 U. S. 78, 86 (1915) (“A person may be guilty of conspiring although incapable of committing the objective offense”); Sand, supra, §19.01, at 19–3 (“[ Y ]ou may find the defendant guilty of conspiracy despite the fact that he himself was incapable of committing the substantive crime”). The Court applied these principles in two cases involving the Mann Act. See Act of June 25, 1910, ch. 395, 36Stat. 825. Section 2 of the Mann Act made it a crime to transport a woman or cause her to be transported across state lines for an immoral purpose.[1] In United States v. Holte, 236 U. S. 140 (1915) , a federal grand jury charged a woman, Clara Holte, with conspiring with a man named Chester Laudenschleger to violate this provision. The District Court dismissed the charge against Holte, holding that because a woman such as Holte could not be con-victed for the substantive offense of transporting herselfor causing herself to be transported across state lines, she also could not be convicted of conspiring to commit that offense. In a succinct opinion by Justice Holmes, the Court rejected this argument, stating that “plainly a person may conspire for the commission of a crime by a third person,” even if “she could not commit the substantive crime” herself. Id., at 144–145.[2] The dissent argued that this holding effectively turned every woman who acquiesced in a covered interstate trip into a conspirator, see id., at 148 (opinion of Lamar, J.), but the Court disagreed. The Court acknowledged that “there may be a degree of coöperation” insufficient to make a woman a conspirator, but it refused to rule out the possibility that a woman could conspire to cause herself to be transported. Id., at 144. To illustrate this point, the Court provided the example of a woman who played an active role in planning and carrying out the trip.[3] The Court expanded on these points in Gebardi v.United States, 287 U. S. 112 (1932) , another Mann Act conspiracy case. A man and a woman were convicted for conspiring to transport the woman from one state to another for an immoral purpose. Id., at 115–116. In deciding the case, the Gebardi Court explicitly reaffirmed the longstanding principle that “[i]ncapacity of one to commit the substantive offense does not necessarily imply that he may with impunity conspire with others who are able to commit it.” Id., at 120. Moreover, the Court fully accepted Holte’s holding that a woman could be convicted of conspiring to cause herself to be transported across state lines. See 287 U. S., at 116–117. But the Court held that the evidence before it was insufficient to support the conspiracy convictions because it “show[ed] no more than that [the woman] went willingly upon the journeys for the purposes alleged.” Id., at 117. Noting that there was no evidence that the woman was “the active or moving spirit in conceiving or carrying out the transportation,” the Court held that the evidence of her “mere consent” or “acquiescence” was not enough. Id., at 117, 123.[4] Holte and Gebardi make perfectly clear that a person may be convicted of conspiring to commit a substantive offense that he or she cannot personally commit. They also show that when that person’s consent or acquiescence is inherent in the underlying substantive offense, something more than bare consent or acquiescence may be needed to prove that the person was a conspirator. B These basic principles of conspiracy law resolve this case. In order to establish the existence of a conspiracy to violate the Hobbs Act, the Government has no obligation to demonstrate that each conspirator agreed personally to commit—or was even capable of committing—the substantive offense of Hobbs Act extortion. It is sufficient to prove that the conspirators agreed that the underlying crime be committed by a member of the conspiracy who was capable of committing it. In other words, each conspirator must have specifically intended that some conspirator commit each element of the substantive offense.[5] That is exactly what happened here: Petitioner, Moreno, and Mejia “share[d] a common purpose,” namely, that petitioner and other police officers would commit every element of the substantive extortion offense. Salinas, 522 U. S., at 63–64. Petitioner and other officers would obtain property “under color of official right,” something that Moreno and Mejia were incapable of doing because they were not public officials. And petitioner and other officers would obtain that money from “another,” i.e., from Mo-reno, Mejia, or Majestic. Although Moreno and Mejia were incapable of committing the underlying substantive offense as principals,[6] they could, under the reasoning of Holte and Gebardi, conspire to commit Hobbs Act extortion by agreeing to help petitioner and other officers commit the substantive offense. See Holte, 236 U. S., at 145 (“[A] conspiracy with an officer or employé of the government or any other for an offence that only he could commit has been held for many years to fall within the conspiracy section . . . of the penal code”); see also Salinas, supra, at 63–64; Gebardi, supra, at 120–121; Rabinowich, 238 U. S., at 86. For these reasons, it is clear that petitioner could be convicted of conspiring to obtain property from the shopowners with their consent and under color of official right. C In an effort to escape this conclusion, petitioner argues that the usual rules do not apply to the type of Hobbs Act conspiracy charged in this case. His basic argument, as ultimately clarified,[7] is as follows. All members of a conspiracy must share the same criminal objective. The objective of the conspiracy charged in this case was to obtain money “from another, with his consent . . . under color of official right.” But Moreno and Mejia did not have the objective of obtaining money “from another” because the money in question was their own. Accordingly, they were incapable of being members of the conspiracy charged in this case. And since there is insufficient evidence in the record to show that petitioner conspired with anyone other than Moreno and Mejia, he must be acquitted. See Reply Brief 3–11, 17–20. This argument fails for a very simple reason: Contrary to petitioner’s claim, he and the shopowners did have a common criminal objective. The objective was not that each conspirator, including Moreno and Mejia, would obtain money from “another” but rather that petitioner and other Baltimore officers would do so. See App. 36–37, Superseding Indictment ¶11 (“It was a purpose of the conspiracy for Moreno and Mejia to enrich over 50 BPD [Baltimore Police Department] Officers . . . in exchange for the BPD Officers’ exercise of their official positions and influence to cause vehicles to be towed or otherwise delivered to Majestic”). Petitioner does not dispute that he was properly convicted for three substantive Hobbs Act violations based on proof that he obtained money “from another.” The criminal objective on which petitioner, Moreno, and Mejia agreed was that petitioner and other Baltimore officers would commit substantive violations of this nature. Thus, under well-established rules of conspiracy law, petitioner was properly charged with and convicted of conspiring with the shopowners. Nothing in the text of the Hobbs Act even remotely undermines this conclusion, and petitioner’s invocation of the rule of lenity[8] and principles of federalism[9] is unavailing. 1 Petitioner argues that our interpretation makes the Hobbs Act sweep too broadly, creating a national antibribery law and displacing a carefully crafted network of state and federal statutes. He contends that a charge of conspiring to obtain money from a conspirator with his consent and under color of official right is tantamount to a charge of soliciting or accepting a bribe and that allowing such a charge undermines 18 U. S. C. §666 (a federal bribery statute applicable to state and local officials) and state bribery laws. He also argues that extortion conspiracies of this sort were not known prior to the enactment of the Hobbs Act and that there is no evidence that Congress meant for that Act to plow this new ground. The subtext of these arguments is that it seems unnatural to prosecute bribery on the basis of a statute prohibiting “extortion,” but this Court held in Evans that Hobbs Act extortion “under color of official right” includes the “rough equivalent of what we would now describe as ‘taking a bribe.’ ” 504 U. S., at 260. Petitioner does not ask us to overturn Evans, see, e.g., Brief for Petitioner 1; Tr. of Oral Arg. 4–5, 12–13, and we have no occasion to do so. Having already held that §1951 prohibits the “rough equivalent” of bribery, we have no principled basis for precluding the prosecution of conspiracies to commit that same offense.[10] Petitioner also exaggerates the reach of our decision. It does not, as he claims, dissolve the distinction between extortion and conspiracy to commit extortion. Because every act of extortion under the Hobbs Act requires property to be obtained with “consent,” petitioner argues, proof of that consent will always or nearly always establish the existence of a conspiratorial agreement and thus allow the Government to turn virtually every such extortion case into a conspiracy case. But there are plenty of instances in which the “consent” required under the Hobbs Act will not be enough to constitute the sort of agreement needed under the law of conspiracy. As used in the Hobbs Act, the phrase “with his consent” is designed to distinguish extortion (“obtaining of property from another, with his consent,” 18 U. S. C. §1951(b)(2) (emphasis added)) from robbery (“obtaining of personal property from the person or in the presence of another, against his will,” §1951(b)(1) (emphasis added)). Thus, “consent” simply signifies the taking of property under circumstances falling short of robbery, and such “consent” is quite different from the mens rea necessary for aconspiracy. This conclusion is clear from the language of §1951 prohibiting the obtaining of property “from another, with his consent, induced by wrongful use of actual or threatened force, violence, or fear.” §1951(b)(2) (emphasisadded). This language applies when, for example, a store owner makes periodic protection payments to gang members out of fear that they will otherwise trash the store. While these payments are obtained with the store owner’s grudging consent, the store owner, simply by making the demanded payments, does not enter into a conspiratorial agreement with the gang members conducting the shakedown. See Salinas, 522 U. S., at 63–65 (conspirators must pursue “the same criminal objective”); United States v. Bailey, 444 U. S. 394, 405 (1980) (conspiracy requires “a heightened mental state”); Anderson v. United States, 417 U. S. 211, 223 (1974) (“the prosecution must show that the offender acted with a specific intent”). Just as mere acquiescence in a Mann Act violation is insufficient to create a conspiracy, see Gebardi, 287 U. S., at 121–123; Holte, 236 U. S., at 145, the minimal “consent” required to trigger §1951 is insufficient to form a conspiratorial agreement. Our interpretation thus does not turn virtually every act of extortion into a conspiracy. Nor does our reading transform every bribe of a public official into a conspiracy to commit extortion. The “consent” required to pay a bribe does not necessarily create a conspiratorial agreement. In cases where the bribe payor is merely complying with an official demand, the payor lacks the mens rea necessary for a conspiracy. See Sa-linas, supra, at 63–65; Bailey, supra, at 405; Anderson, supra, at 223; Gebardi, supra, at 121–123. For example, imagine that a health inspector demands a bribe from a restaurant owner, threatening to close down the restaurant if the owner does not pay. If the owner reluctantly pays the bribe in order to keep the business open, the owner has “consented” to the inspector’s demand, butthis mere acquiescence in the demand does not form a conspiracy.[11] 2 While petitioner exaggerates the impact of our decision, his argument would create serious practical problems. The validity of a charge of Hobbs Act conspiracy would often depend on difficult property-law questions having little to do with criminal culpability. In this case, for example, ownership of the money obtained by petitioner is far from clear. It appears that the funds came from Majestic’s account, App. 97–98, 149, and there is evidence that during the period of petitioner’s membership in the conspiracy, Majestic was converted from a limited liability company to a regular business corporation, id., at 145; App. in No. 12–4462 (CA4), pp. 655–656, 736. After that transformation, the money obtained by petitioner may have come from corporate funds. A corporation is an entity distinct from its shareholders, and therefore, even under petitioner’s interpretation of the applicable law, Moreno and Mejia would have agreed that petitioner would obtain money “from another,” not from them. Suppose that Moreno or Mejia had made the payments by taking money from a personal bank account. Would that dictate a different outcome? Or suppose that Majestic was a partnership and the payments came from a com-pany account. Would that mean that Moreno agreed that officers would obtain money “from another” insofar as they would obtain Mejia’s share of the partnership funds and that Mejia similarly agreed that officers would obtain money “from another” insofar as they would obtain the share belonging to Moreno? Or consider this example. Suppose that the owner and manager of a nightclub reach an agreement with a public official under which the owner will bribe the official to approve the club’s liquor license application. Under petitioner’s approach, the public official and the club manager may be guilty of conspiring to commit extortion, because they agreed that the official would obtain property “from another”—that is, the owner. But as “the ‘another’ from whom the property is obtained,” Reply Brief 10, the owner could not be prosecuted. There is no apparent reason, however, why the manager but not the owner should be culpable in this situation. III A defendant may be convicted of conspiring to violate the Hobbs Act based on proof that he reached an agreement with the owner of the property in question to obtain that property under color of official right. Because petitioner joined such an agreement, his conspiracy conviction must stand. The judgment of the United States Court of Appeals for the Fourth Circuit is affirmed. It is so ordered.Notes 1 In full, §2 provided as follows: 2 The Court assumed that Holte could not be convicted as a principal for the substantive offense of causing herself to be transported across state lines. But the Court noted that it might be possible for a woman to violate §2 of the Mann Act in a different way: by “aiding in procuring any form of transportation for” a covered interstate trip. Holte, 236 U. S., at 144; see 36Stat. 825 (“aid or assist in obtaining transportation”). If a woman could commit that substantive §2 violation, the Court explained, there is no reason why she could not also be convicted of conspiring to commit that offense. See 236 U. S., at 145. The Court, however, refused to hold that this was the only ground on which a woman like Holte could be convicted for conspiring to violate §2. Id., at 144–145. It thus addressed the broader question of whether it was possible for a woman in Holte’s position to commit the offense of conspiring “that Laudenschleger should procure transportation and should cause [Holte] to be transported.” Id., at 144. 3 The Court wrote: 4 The path of reasoning by which the Gebardi Court reached these conclusions was essentially as follows: 5 Section 371 also requires that one of the conspirators commit an overt act in furtherance of the offense. Petitioner does not dispute that this element was satisfied. 6 The Government argues that the lower courts have long held that a private person may be guilty of this type of Hobbs Act extortion as an aider and abettor. See Brief for United States 36–37. We have no occasion to reach that question here. 7 Petitioner’s position has evolved over the course of this litigation. As noted, petitioner requested a jury instruction stating that “[i]n order to convict a defendant of conspiracy to commit extortion under color of official right, the government must prove beyond a reasonable doubt that the conspiracy was to obtain money or property from some person who was not a member of the conspiracy.” App. 53. Under this instruction, as long as the shopowners were named as conspirators, petitioner could not have been convicted even if there was ample evidence to prove that he conspired with other Baltimore officers to obtain money from the shopowners. (And, indeed, when he first raised his Brock argument, see United States v. Brock, 501 F. 3d 762 (CA6 2007), another officer, Manrich, was still in the case and was charged with the same conspiracy.) 8 That rule applies only when a criminal statute contains a “grievous ambiguity or uncertainty,” and “only if, after seizing everything from which aid can be derived,” the Court “can make no more than a guess as to what Congress intended.” Muscarello v. United States, 524 U. S. 125 –139 (1998) (internal quotation marks omitted). 9 We are not unmindful of the federalism concerns implicated by this case, but those same concerns were raised—and rejected—in Evans v. United States, 504 U. S. 255 (1992) , see id., at 290 (Thomas, J., dissenting) (“The Court’s construction of the Hobbs Act is repugnant . . . to basic tenets of federalism”), which we accept as controlling here, see Part II–C–1, infra. 10 Justice Thomas argues that Evans was wrongly decided, and his position makes sense to the extent that he simply refuses to accept that case. But it founders insofar as it suggests that even if Evans is accepted in relation to substantive Hobbs Act charges, it should not be extended to conspiracy cases. See post, at 1 (dissenting opinion) (“I would not extend Evans’ errors further”); post, at 3 (“[The Court’s] holding . . . needlessly extends Evans’ error to the conspiracy context”); post, at 4 (“The Court today takes another step away from the common-law understanding of extortion that the Hobbs Act adopted”). It would be very strange if a provision of the criminal code meant one thing with respect to charges of a substantive violation but something very different in cases involving a conspiracy to commit the same offense. 11 Petitioner also claims that naming Moreno and Mejia as conspirators opened the door for prosecutors to employ the potent party-joinder and evidentiary rules that conspiracy charges make available. See Brief for Petitioner 10–11, 18, 26–27, 37. But the naming of the shopowners had no effect on joinder. The only other defendant named in the superseding indictment, Manrich, could have been joined even if the shopowners had not been named. Nor did naming Moreno and Mejia have any effect on the admissibility of evidence of overt acts committed by the Baltimore officers named as petitioner’sco-conspirators.

579.US.2015_15-233

In response to an ongoing fiscal crisis, petitioner Puerto Rico enacted the Puerto Rico Public Corporation Debt Enforcement and Recovery Act. Portions of the Recovery Act mirror Chapters 9 and 11 of the Federal Bankruptcy Code and enable Puerto Rico’s public utility corporations to restructure their climbing debt. Respondents, a group of investment funds and utility bondholders, sought to enjoin the Act. They contended, among other things, that a Bankruptcy Code provision explicitly pre-empts the Recovery Act, see 11 U. S. C. §903(1). The District Court enjoined the Act’s enforcement, and the First Circuit affirmed, concluding that the Bankruptcy Code’s definition of “State” to include Puerto Rico, except for purposes of defining who may be a debtor under Chapter 9, §101(52), did not remove Puerto Rico from the scope of the pre-emption provision. Held: Section 903(1) of the Bankruptcy Code pre-empts Puerto Rico’s Recovery Act. Pp. 5–15. (a) Three federal municipal bankruptcy provisions are relevant here. First, the “gateway” provision, §109(c), requires a Chapter 9 debtor to be an insolvent municipality that is “specifically authorized” by a State “to be a debtor.” Second, the pre-emption provision, §903(1), expressly bars States from enacting municipal bankruptcy laws. Third, the definition of “State,” §101(52), as amended in 1984, “includes . . . Puerto Rico, except for the purpose of defining who may be a debtor under chapter 9.” Pp. 5–8. (b) If petitioners are correct that the amended definition of “State” excludes Puerto Rico altogether from Chapter 9, then the pre-emption provision does not apply. But if respondents’ narrower reading is correct and the definition only precludes Puerto Rico from authorizing its municipalities to seek Chapter 9 relief, then Puerto Rico is barred from implementing its Recovery Act. Pp. 8–14. (1) The Bankruptcy Code’s plain text supports respondents’ reading. The unambiguous language of the pre-emption provision “contains an express pre-emption clause,” the plain wording of which “necessarily contains the best evidence of Congress’ pre-emptive intent.” Chamber of Commerce of United States of America v. Whiting, 563 U. S. 582 . The definition provision excludes Puerto Rico for the single purpose of defining who may be a Chapter 9 debtor, an unmistakable reference to the §109 gateway provision. This conclusion is reinforced by the definition’s use of the phrase “defining who may be a debtor under chapter 9,” §101(52), which is tantamount to barring Puerto Rico from “specifically authorizing” which municipalities may file Chapter 9 petitions under the gateway provision, §903(1). The text of the exclusion thus extends no further. Had Congress intended to exclude Puerto Rico from Chapter 9 altogether, including Chapter 9’s pre-emption provision, Congress would have said so. Pp. 9–11. (2) The amended definition of “State” does not exclude Puerto Rico from all of Chapter 9’s provisions. First, Puerto Rico’s exclusion as a “State” for purposes of the gateway provision does not also remove Puerto Rico from Chapter 9’s separate pre-emption provision. A State that chooses under the gateway provision not to authorize a municipality to file is still bound by the pre-emption provision. Likewise, Puerto Rico is bound by the pre-emption provision, even though Congress has removed its authority under the gateway provision to authorize its municipalities to seek Chapter 9 relief. Second, because Puerto Rico was not “by definition” excluded from Chapter 9, both §903’s introductory clause and its proviso, the pre-emption provision, continue to apply in Puerto Rico. Finally, the argument that the Recovery Act is not a “State law” that can be pre-empted is based on technical amendments to the terms “creditor” and “debtor” that are too “subtle” to support such a “[f]undamental chang[e] in the scope” of Chapter 9’s pre-emption provision. Kellogg Brown & Root Services, Inc. v. United States ex rel. Carter, 575 U. S. ___, ___. Pp. 11–14. 805 F. 3d 322, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Kagan, JJ., joined. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, J., joined. Alito, J., took no part in the consideration or decision of these cases. Notes 1 Together with No. 15–255, Acosta-Febo et al. v. Franklin California Tax-Free Trust et al., also on certiorari to the same court.

The Federal Bankruptcy Code pre-empts state bankruptcy laws that enable insolvent municipalities to restructure their debts over the objections of creditors and instead requires municipalities to restructure such debts under Chapter 9 of the Code. 11 U. S. C. §903(1). We must decide whether Puerto Rico is a “State” for purposes of this pre-emption provision. We hold that it is. The Bankruptcy Code has long included Puerto Rico as a “State,” but in 1984 Congress amended the definition of “State” to exclude Puerto Rico “for the purpose of defining who may be a debtor under chapter 9.” Bankruptcy Amendments and Federal Judgeship Act, §421(j)(6), 98Stat. 368, now codified at 11 U. S. C. §101(52). Puerto Rico interprets this amended definition to mean that Chapter 9 no longer applies to it, so it is no longer a “State” for purposes of Chapter 9’s pre-emption provision. We hold that Congress’ exclusion of Puerto Rico from the definition of a “State” in the amended definition does not sweep so broadly. By excluding Puerto Rico “for the purpose of defining who may be a debtor under chapter 9,” §101(52) (emphasis added), the Code prevents Puerto Rico from authorizing its municipalities to seek Chapter 9 relief. Without that authorization, Puerto Rico’s municipalities cannot qualify as Chapter 9 debtors. §109(c)(2). But Puerto Rico remains a “State” for other purposes related to Chapter 9, including that chapter’s pre-emption provision. That provision bars Puerto Rico from enacting its own municipal bankruptcy scheme to restructure the debt of its insolvent public utilities companies. I A Puerto Rico and its instrumentalities are in the midst of a fiscal crisis. More than $20 billion of Puerto Rico’s climbing debt is shared by three government-owned public utilities companies: the Puerto Rico Electric Power Authority, the Puerto Rico Aqueduct and Sewer Authority, and the Puerto Rico Highways and Transportation Authority. For the fiscal year ending in 2013, the three public utilities operated with a combined deficit of $800 million. The Government Development Bank for Puerto Rico (Bank)—the Commonwealth’s government-owned bank and fiscal agent—has previously provided financing to enable the utilities to continue operating without defaulting on their debt obligations. But the Bank now faces a fiscal crisis of its own. As of fiscal year 2013, it had loaned nearly half of its assets to Puerto Rico and its public utilities. Puerto Rico’s access to capital markets has also been severely compromised since ratings agencies downgraded Puerto Rican bonds, including the utilities’, to noninvestment grade in 2014. Puerto Rico responded to the fiscal crisis by enacting the Puerto Rico Corporation Debt Enforcement and Recovery Act (Recovery Act) in 2014, which enables the Commonwealth’s public utilities to implement a recovery or restructuring plan for their debt. 2014 Laws P. R. p. 371. See generally McGowen, Puerto Rico Adopts A Debt Recovery Act For Its Public Corporations, 10 Pratt’s J. Bkrtcy. Law 453 (2014). Chapter 2 of the Recovery Act createsa “consensual” debt modification procedure that permits the public utilities to propose changes to the terms of the outstanding debt instruments, for example, changing the interest rate or the maturity date of the debt. 2014 Laws P. R., at 428–429. In conjunction with the debt modification, the public utility must also propose a Bank-approved recovery plan to bring it back to financial self-sufficiency. Ibid. The debt modification binds all creditors so long as those holding at least 50% of affected debt participate in (or consent to) a vote regarding the modifications, and the participating creditors holding at least 75% of affected debt approve the modifications. Id., at 430. Chapter 3 of the Recovery Act, on the other hand, mirrors Chapters 9 and 11 of the Federal Bankruptcy Code by creating a court-supervised restructuring process intended to offer the best solution for the broadest group of creditors. See id., at 448–449. Creditors holding two-thirds of an affected class of debt must participate in the vote to approve the restructuring plan, and half of those participants must agree to the plan. Id., at 449. B A group of investment funds, including the Franklin California Tax-Free Trust, and BlueMountain Capital Management, LLC, brought separate suits against Puerto Rico and various government officials, including agents of the Bank, to enjoin the enforcement of the Recovery Act. Collectively, the plaintiffs hold nearly $2 billion in bonds issued by the Electric Power Authority, one of the distressed utilities. The complaints alleged, among other claims, that the Federal Bankruptcy Code prohibited Puerto Rico from implementing its own municipal bankruptcy scheme. The District Court consolidated the suits and ruled in the plaintiffs’ favor on their pre-emption claim. 85 F. Supp. 3d 577 (PR 2015). The court concluded that the pre-emption provision in Chapter 9 of the Federal Bankruptcy Code, 11 U. S. C. §903(1), precluded Puerto Rico from implementing the Recovery Act and enjoined its enforcement. 85 F. Supp. 3d, at 601, 614. The First Circuit affirmed. 805 F. 3d 322 (2015). The court examined the 1984 amendment to the definition of “State” in the Federal Bankruptcy Code, which includes Puerto Rico as a “State” for purposes of the Code “ ‘except for the purpose of defining who may be a debtor under chapter 9.’ ” Id., at 330–331 (quoting §101(52); emphasis added). The court concluded that the amendment did not remove Puerto Rico from the scope of the pre-emption provision and held that the pre-emption provision barred the Recovery Act. Id., at 336–337. The court opined that it was up to Congress, not Puerto Rico, to decide when the government-owned companies could seek bankruptcy relief. Id., at 345. We granted the Commonwealth’s petitions for writs of certiorari. 577 U. S. ___ (2015).[1] II These cases require us to parse three provisions of the Bankruptcy Code: the “who may be a debtor” provision requiring States to authorize municipalities to seek Chapter 9 relief, §109(c), the pre-emption provision barring States from enacting their own municipal bankruptcy schemes, §903(1), and the definition of “State,” §101(52). We first explain the text and history of these provisions. We then conclude that Puerto Rico is still a “State” for purposes of the pre-emption provision and hold that this provision pre-empts the Recovery Act. A The Constitution empowers Congress to establish “uniform Laws on the subject of Bankruptcies throughout the United States.” Art. I, §8, cl. 4. Congress first exercised that power by enacting a series of temporary bankruptcy Acts beginning in 1800, which gave way to a permanent federal bankruptcy scheme in 1898. See An Act To Establish a Uniform System of Bankruptcy Throughout the United States, 30Stat. 544; Hanover Nat. Bank v. Moyses, 186 U. S. 181, 184 (1902) . But Congress did not enter the field of municipal bankruptcy until 1933 when it enacted the precursor to Chapter 9, a chapter of the Code enabling an insolvent “municipality,” meaning a “political subdivision or public agency or instrumentality of a State,” 11 U. S. C. §101(40), to restructure municipal debts. See McConnell & Picker, When Cities Go Broke: A Conceptual Introduction to Municipal Bankruptcy, 60 U. Chi. L. Rev. 425, 427, 450–451 (1993). Congress has tailored the federal municipal bankruptcy laws to preserve the States’ reserved powers over their municipalities. This Court struck down Congress’ first attempt to enable the States’ political subdivisions to file for federal bankruptcy relief after concluding that it infringed the States’ powers “to manage their own affairs.” Ashton v. Cameron County Water Improvement Dist. No. One, 298 U. S. 513, 531 (1936) . Congress tried anew in 1937, and the Court upheld the amended statute as an appropriate balance of federal and state power. See United States v. Bekins, 304 U. S. 27 –53 (1938). Critical to the Court’s constitutional analysis was that the State had first authorized its instrumentality to seek relief under the federal bankruptcy laws. See id., at 47–49, 53–54. Still today, the provision of the Bankruptcy Code defining who may be a debtor under Chapter 9, which we refer to here as the “gateway” provision, requires the States to authorize their municipalities to seek relief under Chapter 9 before the municipalities may file a Chapter 9 petition: “§109. Who may be a debtor . . . . . “(c) An entity may be a debtor under chapter 9 of this title if and only if such entity — “(1) is a municipality; “(2) is specifically authorized, in its capacity as a municipality or by name, to be a debtor under such chapter by State law, or by a governmental officer or organization empowered by State law to authorize such entity to be a debtor under such chapter . . . .” The States’ powers are not unlimited, however. The federal bankruptcy laws changed again in 1946 to bar the States from enacting their own municipal bankruptcy schemes. The amendment overturned this Court’s holding in Faitoute Iron & Steel Co. v. Asbury Park, 316 U. S. 502 –509 (1942) (rejecting contention that Congress occupied the field of municipal bankruptcy law). In Faitoute, the Court held that federal bankruptcy laws did not pre-empt New Jersey’s municipal bankruptcy scheme, which required municipalities to seek relief under state law before resorting to the federal municipal bankruptcy scheme. Ibid. To override Faitoute, Congress enacted a provision expressly pre-empting state municipal bankruptcy laws. Act of July 1, 1946, 60Stat. 415. The express pre-emption provision, central to these cases, is now codified with some stylistic changes in §903(1): “§903. Reservation of State power to control municipalities “This chapter does not limit or impair the power of a State to control, by legislation or otherwise, a municipality of or in such State in the exercise of the political or governmental powers of such municipality, including expenditures for such exercise, but— “(1) a State law prescribing a method of composition of indebtedness of such municipality may not bind any creditor that does not consent to such composition; and “(2) a judgment entered under such a law maynot bind a creditor that does not consent to suchcomposition.” The third provision of the Bankruptcy Code at issue is the definition of “State,” which has included Puerto Rico since it became a Territory of the United States in 1898. The first Federal Bankruptcy Act, also enacted in 1898, defined “States” to include “the Territories, the Indian Territory, Alaska, and the District of Columbia.” 30Stat. 545. When Congress recodified the bankruptcy laws to form the Federal Bankruptcy Code in 1978, the definition of “State” dropped out of the definitional section. See generally Bankruptcy Reform Act, 92Stat. 2549–2554. Congress then amended the Code to reincorporate the definition of “State” in 1984. §421, 98Stat. 368–369, now codified at §101(52). The amended definition includes Puerto Rico as a State for purposes of the Code with one exception: “§101. Definitions . . . . . “(52) The term ‘State’ includes the District ofColumbia and Puerto Rico, except for the purpose of defining who may be a debtor under chapter 9 of this title.” B It is our task to determine the effect of the amended definition of “State” on the Code’s other provisions governing Chapter 9 proceedings. We must decide whether, in light of the amended definition, Puerto Rico is no longer a “State” only for purposes of the gateway provision, which requires States to authorize their municipalities to seek Chapter 9 relief, or whether Puerto Rico is also no longer a “State” for purposes of the pre-emption provision. The parties do not dispute that, before 1984, Puerto Rico was a “State” for purposes of Chapter 9’s pre-emption provision. Accordingly, before 1984, federal law would have pre-empted the Recovery Act because it is a “State law prescribing a method of composition of indebtedness” for Puerto Rico’s instrumentalities that would bind nonconsenting creditors, §903(1). The parties part ways, however, in deciphering how the 1984 amendment to the definition of “State” affected the pre-emption provision. Petitioners interpret the amended definition of “State” to exclude Puerto Rico altogether from Chapter 9. If petitioners are correct, then the pre-emption provision does not apply to them. Puerto Rico, in other words, may enact its own municipal bankruptcy scheme without running afoul of the Code. Respondents, on the other hand, read the amended definition narrowly. They contend that the definition precludes Puerto Rico from “specifically authoriz[ing]” its municipalities to seek relief, as required by the gateway provision, §109(c)(2), but that Puerto Rico is no less a “State” for purposes of the pre-emption provision than the other “State[s],” as that term is defined in the Code. If respondents are correct, then the pre-emption provision applies to Puerto Rico and bars it from enacting the Recovery Act. Respondents have the better reading. We hold that Puerto Rico is still a “State” for purposes of the pre-emption provision. The 1984 amendment precludes Puerto Rico from authorizing its municipalities to seek relief under Chapter 9, but it does not remove Puerto Rico from the reach of Chapter 9’s pre-emption provision. 1 The plain text of the Bankruptcy Code begins and ends our analysis. Resolving whether Puerto Rico is a “State” for purposes of the pre-emption provision begins “with the language of the statute itself,” and that “is also where the inquiry should end,” for “the statute’s language is plain.” United States v. Ron Pair Enterprises, Inc., 489 U. S. 235, 241 (1989) . And because the statute “contains an express pre-emption clause,” we do not invoke any presumption against pre-emption but instead “focus on the plain wording of the clause, which necessarily contains the best evidence of Congress’ pre-emptive intent.” Chamber of Commerce of United States of America v. Whiting, 563 U. S. 582, 594 (2011) (internal quotation marks omitted); see also Gobeille v. Liberty Mut. Ins. Co., 577 U. S. ___, ___ (2016) (slip op., at 12). The amended definition of “State” excludes Puerto Rico for the single “purpose of defining who may be a debtor under chapter 9 of this title.” §101(52) (emphasis added). That exception unmistakably refers to the gateway provision in §109, titled “who may be a debtor.” Section 109(c) begins, “An entity may be a debtor under chapter 9 of this title if and only if . . . .” §109(c). We interpret Congress’ use of the “who may be a debtor” language in the amended definition of “State” to mean that Congress intended to exclude Puerto Rico from this gateway provision delineating who may be a debtor under Chapter 9. See, e.g., Sullivan v. Stroop, 496 U. S. 478, 484 (1990) (reading same term used in different parts of the same Act to have the same meaning); see also Northcross v. Board of Ed. of Memphis City Schools, 412 U. S. 427, 428 (1973) (per curiam) (“[S]imilarity of language . . . is . . . a strong indication that the two statutes should be interpreted pari passu”). Puerto Rico, therefore, is not a “State” for purposes of the gateway provision, so it cannot perform the single function of the “State[s]” under that provision: to “specifically authoriz[e]” municipalities to seek Chapter 9 relief. §109(c). As a result, Puerto Rico’s municipalities cannot satisfy the requirements of Chapter 9’s gateway provision until Congress intervenes. The amended definition’s use of the term “defining” also confirms our conclusion that the amended definition excludes Puerto Rico as a “State” for purposes of the gateway provision. The definition specifies that Puerto Rico is not a “ ‘State . . . for the purpose of defining who may be a debtor under Chapter 9.” §101(52) (emphasis added). To “define” is “to decide upon,” 4 Oxford English Dictionary 383 (2d ed. 1989), or “to settle” or “to establish or prescribe authoritatively,” Black’s Law Dictionary 380 (5th ed. 1979). As discussed, a State’s role under the gateway provision is to do just that: The State must define (or “decide upon”) which entities may seek Chapter 9 relief. Barring Puerto Rico from “defining who may be a debtor under chapter 9” is tantamount to barring Puerto Rico from “specifically authorizing” which municipalities may file Chapter 9 petitions under the gateway provision.The amended definition of “State” unequivocally ex-cludes Puerto Rico as a “State” for purposes of the gateway provision. The text of the definition extends no further. The exception excludes Puerto Rico only for purposes of the gateway provision. Puerto Rico is no less a “State” for purposes of the pre-emption provision than it was before Congress amended the definition. The Code’s pre-emption provision has prohibited States and Territories defined as “States” from enacting their own municipal bankruptcy schemes for 70 years. See 60Stat. 415 (overturning Faitoute, 316 U. S., at 507–509). Had Congress intended to “alter th[is] fundamental detai[l]” of municipal bankruptcy, we would expect the text of the amended definition to say so. Whitman v. American Trucking Assns., Inc., 531 U. S. 457, 468 (2001) . Congress “does not, one might say, hide elephants in mouseholes.” Ibid. 2 The dissent, adopting many of petitioners’ arguments, reads the amended definition to say what it does not—that “for the purpose of . . . chapter 9,” Puerto Rico is not a State. The arguments in support of that capacious reading are unavailing. First, the dissent agrees with petitioners’ view that the exclusion of Puerto Rico as a “State” for purposes of the gateway provision effectively removed Puerto Rico from all of Chapter 9. See post, at 7–8 (opinion of Sotomayor, J.). To be sure, §109(c) and the surrounding subsections serve an important gatekeeping role. Those provisions “specify who qualifies—and who does not qualify—as a debtor under the various chapters of the Code.” Toibb v. Radloff, 501 U. S. 157, 161 (1991) . For instance, a railroad must file under Chapter 11, not Chapter 7, §§109(b)(1), (d), whereas only “family farmer[s] or family fisherm[e]n” may file under Chapter 12. The provision delineating who may be a debtor under Chapter 9 is no exception. Only municipalities may file under Chapter 9, and only if the State has “specifically authorized” the municipality to do so. §§109(c)(1)–(2); see also McConnell & Picker, 60 Chi. L. Rev., at 455–461 (discussing the gatekeeping requirements for Chapter 9). That Puerto Rico is not a “State” for purposes of the gateway provision, however, says nothing about whether Puerto Rico is a “State” for the other provisions of Chapter 9 involving the States. The States do not “pass through” the gateway provision. Post, at 8. The gateway provision is instead directed at the debtors themselves—the municipalities, in the case of Chapter 9 bankruptcy. A municipality that cannot secure state authorization to file a Chapter 9 petition is excluded from Chapter 9 entirely. But the same cannot be said about the State in which that municipality is located. A State’s only role under the gateway provision is to provide that “authoriz[ation]” to file. §109(c)(2). The pre-emption provision then imposes an additional requirement: The States may not enact their own municipal bankruptcy schemes. A State that chooses not to authorize its municipalities to seek Chapter 9 relief under the gateway provision is no less bound by that pre-emption provision. Here too, Puerto Rico is no less bound by the pre-emption provision even though Congress has removed its authority to provide authorization for its municipalities to file Chapter 9 petitions. Again, if it were Congress’ intent to also exclude Puerto Rico as a “State” for purposes of that pre-emption provision, it would have said so. Second, both petitioners and the dissent place great weight on the introductory clause of §903. Post, at 6–7. The pre-emption provision cannot apply to Puerto Rico, so goes the argument, because it is a proviso to §903’s introductory clause, which they posit is inapplicable to Puerto Rico. The introductory clause affirms that Chapter 9 “does not limit or impair the power of a State to control” its “municipalit[ies].” §903. The dissent surmises that this clause “is irrelevant” and “meaningless” in Puerto Rico. Post, at 7. Because Puerto Rico’s municipalities are ineligible for Chapter 9 relief, Chapter 9 cannot “affec[t] Puerto Rico’s control over its municipalities,” according to the dissent. Ibid. In other words, “there is no power” for the introductory clause to “reserve” for Puerto Rico’s use. Ibid. Petitioners likewise contend that “it would be nonsensical for Congress to provide Puerto Rico with a shield against intrusion by a Chapter that, by definition, can have no effect on Puerto Rico.” Brief for Petitioner Commonwealth of Puerto Rico et al. in No. 15–233, p. 25. So “it follows” that the pre-emption provision, the proviso to that clause, cannot apply either. Ibid. This reading rests on the faulty assumption that Puerto Rico is, “by definition,” excluded from Chapter 9. Ibid. For all of the reasons already explained, see Part II–B–1, supra, it is not. The amended definition of “State” precludes Puerto Rico from authorizing its municipalities to seek Chapter 9 relief. But Puerto Rico is no less a “State” for purposes of §903’s introductory clause and its proviso. Both continue to apply in Puerto Rico. They are neither “irrelevant” nor “meaningless.” Post, at 7. If, for example, Congress created a path for the Puerto Rican municipalities to restructure their debts under Chapter 9, then §903 would assure Puerto Rico, no less a “State” for purposesof this section, of its continued power to “control, bylegislation or otherwise, [its] municipalit[ies] . . . in the exercise of the political or governmental powers of such municipalit[ies].” Third, the Government Development Bank contends that the Recovery Act does not run afoul of the pre-emption provision because the Recovery Act does not bind nonconsenting “creditors,” as the Bankruptcy Code now defines that term. In 1978, Congress redefined “creditor” to mean an “entity that has a claim against the debtor . . . .” 92Stat. 2550, now codified at §101(10) (emphasis added). A “debtor,” in turn, is a “person or municipality concerning which a case under this title has been commenced.” Id., at 2551, now codified at §101(13) (emphasis added). In light of these definitions, the Bank contends that the Puerto Rican municipalities are not “debtor[s]” as the Code defines the term because they cannot “commenc[e]” an action under Chapter 9 without authorization from Puerto Rico. Brief for Petitioner Acosta-Febo et al. 31–33. And because respondents cannot be “creditors” of a nonexistent “debtor,” the Recovery Act is not a “State law” that binds “any creditor.” §903(1). Id., at 31–33. Tellingly, the dissent does not adopt this reading. The Bank’s interpretation would nullify the pre-emption provision. Applying the Bank’s logic, a municipality that fails to meet any one of the requirements of Chapter 9’s gatekeeping provision is not a “debtor” and would have no “creditors.” So a State could refuse to “specifically authoriz[e]” its municipalities to seek relief under Chapter 9, §109(c)(2), required to commence a case under that chapter. That State would be free to enact its own municipal bankruptcy scheme because its municipalities would have no “creditors” under federal law. The technical amendments to the definitions of “creditor” and “debtor” are too “subtle a move” to support such a “[f]undamental chang[e] in the scope” of Chapter 9’s pre-emption provision. Kellogg Brown & Root Services, Inc. v. United States ex rel. Carter, 575 U. S. ___, ___ (2015) (slip op., at 9). * * * The dissent concludes that “the government and people of Puerto Rico should not have to wait for possible congressional action to avert the consequences” of the Commonwealth’s fiscal crisis. Post, at 9. But our constitutional structure does not permit this Court to “rewrite thestatute that Congress has enacted.” Dodd v. United States, 545 U. S. 353, 359 (2005) ; see also Electric Storage Battery Co. v. Shimadzu, 307 U. S. 5, 14 (1939) . That statute precludes Puerto Rico from authorizing its municipalities to seek relief under Chapter 9. But it does not remove Puerto Rico from the scope of Chapter 9’s pre-emption provision. Federal law, therefore, pre-empts the Recovery Act. The judgment of the Court of Appeals for the First Circuit is affirmed. It is so ordered. Justice Alito took no part in the consideration or decision of these cases.Notes 1 After the parties briefed and argued these cases, Members of Congress introduced a bill in the House of Representatives to establish an oversight board to assist Puerto Rico and its instrumentalities. See H. 5278, 114th Cong., 2d Sess. (2016). The bill does not amend the Fed-eral Bankruptcy Code; it instead proposes adding a chapter to Title 48, governing the Territories. Id., §6.

579.US.2015_15-108

Respondents Luis Sánchez Valle and Jaime Gómez Vázquez each sold a gun to an undercover police officer. Puerto Rican prosecutors indicted them for illegally selling firearms in violation of the Puerto Rico Arms Act of 2000. While those charges were pending, federal grand juries also indicted them, based on the same transactions, for violations of analogous U. S. gun trafficking statutes. Both defendants pleaded guilty to the federal charges and moved to dismiss the pending Commonwealth charges on double jeopardy grounds. The trial court in each case dismissed the charges, rejecting prosecutors’ arguments that Puerto Rico and the United States are separate sovereigns for double jeopardy purposes and so could bring successive prosecutions against each defendant. The Puerto Rico Court of Appeals consolidated the cases and reversed. The Supreme Court of Puerto Rico granted review and held, in line with the trial court, that Puerto Rico’s gun sale prosecutions violated the Double Jeopardy Clause. Held: The Double Jeopardy Clause bars Puerto Rico and the United States from successively prosecuting a single person for the same conduct under equivalent criminal laws. Pp. 5–18. (a) Ordinarily, a person cannot be prosecuted twice for the same offense. But under the dual-sovereignty doctrine, the Double Jeopardy Clause does not bar successive prosecutions if they are brought by separate sovereigns. See, e.g., United States v. Lanza, 260 U. S. 377 . Yet “sovereignty” in this context does not bear its ordinary meaning. This Court does not examine the extent of control that one prosecuting entity wields over the other, the degree to which an entity exercises self-governance, or a government’s more particular ability to enact and enforce its own criminal laws. Rather, the test hinges on a single criterion: the “ultimate source” of the power undergird-ing the respective prosecutions. United States v. Wheeler, 435 U. S. 313 . If two entities derive their power to punish from independent sources, then they may bring successive prosecutions. Conversely, if those entities draw their power from the same ultimate source, then they may not. Under that approach, the States are separate sovereigns from the Federal Government and from one another. Because States rely on “authority originally belonging to them before admission to the Union and preserved to them by the Tenth Amendment,” state prosecutions have their roots in an “inherent sovereignty” unconnected to the U. S. Congress. Heath v. Alabama, 474 U. S. 82 . For similar reasons, Indian tribes also count as separate sovereigns. A tribe’s power to punish pre-existed the Union, and so a tribal prosecution, like a State’s, is “attributable in no way to any delegation . . . of federal authority.” Wheeler, 435 U. S., at 328. Conversely, a municipality cannot count as a sovereign distinct from a State, because it receives its power, in the first instance, from the State. See, e.g., Waller v. Florida, 397 U. S. 387 . And most pertinent here, this Court concluded in the early 20th century that U. S. territories—including an earlier incarnation of Puerto Rico itself—are not sovereigns distinct from the United States. Grafton v. United States, 206 U. S. 333 . The Court reasoned that “the territorial and federal laws [were] creations emanating from the same sovereignty,” Puerto Rico v. Shell Co. (P. R.), Ltd., 302 U. S. 253 , and so federal and territorial prosecutors do not derive their powers from independent sources of authority. Pp. 5–11. (b) The Grafton and Shell Co. decisions, in and of themselves, do not control here. In the mid-20th century, Puerto Rico became a new kind of political entity, still closely associated with the United States but governed in accordance with, and exercising self-rule through, a popularly-ratified constitution. The magnitude of that change requires consideration of the dual-sovereignty question anew. Yet the result reached, given the historical test applied, ends up the same. Going back as far as the doctrine demands—to the “ultimate source” of Puerto Rico’s prosecutorial power—reveals, once again, the U. S. Congress. Wheeler, 435 U. S., at 320. Pp. 12–18. (1) In 1950, Congress enacted Public Law 600, which authorized the people of Puerto Rico to organize a government pursuant to a constitution of their own adoption. The Puerto Rican people capitalized on that opportunity, calling a constitutional convention and overwhelmingly approving the charter it drafted. Once Congress approved that proposal—subject to several important conditions accepted by the convention—the Commonwealth of Puerto Rico, a new political entity, came into being. Those constitutional developments were of great significance—and, indeed, made Puerto Rico “sovereign” in one commonly understood sense of that term. At that point, Congress granted Puerto Rico a degree of autonomy comparable to that possessed by the States. If the dual-sovereignty doctrine hinged on measuring an entity’s self-governance, the emergence of the Commonwealth would have resulted as well in the capacity to bring the kind of successive prosecutions attempted here. Pp. 13–14. (2) But the dual-sovereignty test focuses not on the fact of self-rule, but on where it first came from. And in identifying a prosecuting entity’s wellspring of authority, the Court has insisted on going all the way back—beyond the immediate, or even an intermediate, locus of power to what is termed the “ultimate source.” On this settled approach, Puerto Rico cannot benefit from the dual-sovereignty doctrine. True enough, that the Commonwealth’s power to enact and enforce criminal law now proceeds, just as petitioner says, from the Puerto Rico Constitution as “ordain[ed] and establish[ed]” by “the people.” P. R. Const., Preamble. But back of the Puerto Rican people and their Constitution, the “ultimate” source of prosecutorial power remains the U. S. Congress. Congress, in Public Law 600, authorized Puerto Rico’s constitution-making process in the first instance, and Congress, in later legislation, both amended the draft charter and gave it the indispensable stamp of approval. Put simply, Congress conferred the authority to create the Puerto Rico Constitution, which in turn confers the authority to bring criminal charges. That makes Congress the original source of power for Puerto Rico’s prosecutors—as it is for the Federal Government’s. The island’s Constitution, significant though it is, does not break the chain. Pp. 14–18. Affirmed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, and Alito, JJ., joined. Ginsburg, J., filed a concurring opinion, in which Thomas, J., joined. Thomas, J., filed an opinion concurring in part and concurring in the judgment. Breyer, J., filed a dissenting opinion, in which Sotomayor, J., joined.

The Double Jeopardy Clause of the Fifth Amendment prohibits more than one prosecution for the “same offence.” But under what is known as the dual-sovereignty doctrine, a single act gives rise to distinct offenses—and thus may subject a person to successive prosecutions—if it violates the laws of separate sovereigns. To determine whether two prosecuting authorities are different sovereigns for double jeopardy purposes, this Court asks a narrow, historically focused question. The inquiry does not turn, as the term “sovereignty” sometimes suggests, on the degree to which the second entity is autonomous from the first or sets its own political course. Rather, the issue is only whether the prosecutorial powers of the two jurisdictions have independent origins—or, said conversely, whether those powers derive from the same “ultimate source.” United States v. Wheeler, 435 U. S. 313, 320 (1978) . In this case, we must decide if, under that test, Puerto Rico and the United States may successively prosecute a single defendant for the same criminal conduct. We hold they may not, because the oldest roots of Puerto Rico’s power to prosecute lie in federal soil. I A Puerto Rico became a territory of the United States in 1898, as a result of the Spanish-American War. The treaty concluding that conflict ceded the island, then a Spanish colony, to the United States, and tasked Congress with determining “[t]he civil rights and political status” of its inhabitants. Treaty of Paris, Art. 9, Dec. 10, 1898, 30Stat. 1759. In the ensuing hundred-plus years, the United States and Puerto Rico have forged a unique political relationship, built on the island’s evolution into a constitutional democracy exercising local self-rule. Acting pursuant to the U. S. Constitution’s Territory Clause, Congress initially established a “civil government” for Puerto Rico possessing significant authority over internal affairs. Organic Act of 1900, ch. 191, 31Stat. 77; see U. S. Const., Art. IV, §3, cl. 2 (granting Congress the “Power to dispose of and make all needful Rules and Regulations respecting the Territory or other Property belonging to the United States”). The U. S. President, with the advice and consent of the Senate, appointed the governor, supreme court, and upper house of the legislature; the Puerto Rican people elected the lower house themselves. See §§17–35, 31Stat. 81–85. Federal statutes generally applied (as they still do) in Puerto Rico, but the newly constituted legislature could enact local laws in much the same way as the then-45 States. See §§14–15, 32, id., at 80, 83–84; Puerto Rico v. Shell Co. (P. R.), Ltd., 302 U. S. 253, 261 (1937) . Over time, Congress granted Puerto Rico additional autonomy. A federal statute passed in 1917, in addition to giving the island’s inhabitants U. S. citizenship, replaced the upper house of the legislature with a popularly elected senate. See Organic Act of Puerto Rico, ch. 145, §§5, 26, 39Stat. 953, 958. And in 1947, an amendment to that law empowered the Puerto Rican people to elect their own governor, a right never before accorded in a U. S. territory. See Act of Aug. 5, 1947, ch. 490, §1, 61Stat. 770. Three years later, Congress enabled Puerto Rico to embark on the project of constitutional self-governance. Public Law 600, “recognizing the principle of government by consent,” authorized the island’s people to “organize a government pursuant to a constitution of their own adoption.” Act of July 3, 1950, §1, 64Stat. 319. Describing itself as “in the nature of a compact,” the statute submitted its own terms to an up-or-down referendum of Puerto Rico’s voters. Ibid. According to those terms, the eventual constitution had to “provide a republican form of government” and “include a bill of rights”; all else would be hashed out in a constitutional convention. §2, 64Stat. 319. The people of Puerto Rico would be the first to decide, in still another referendum, whether to adopt that convention’s proposed charter. See §3, 64Stat. 319. But Congress would cast the dispositive vote: The constitution, Public Law 600 declared, would become effective only “[u]pon approval by the Congress.” Ibid. Thus began two years of constitution-making for the island. The Puerto Rican people first voted to accept Public Law 600, thereby triggering a constitutional convention. And once that body completed its work, the island’s voters ratified the draft constitution. Congress then took its turn on the document: Before giving its approval, Congress removed a provision recognizing various social welfare rights (including entitlements to food, housing, medical care, and employment); added a sentence prohibiting certain constitutional amendments, including any that would restore the welfare-rights section; and inserted language guaranteeing children’s freedom to attend private schools. See Act of July 3, 1952, 66 Stat. 327; Draft Constitution of the Commonwealth of Puerto Rico (1952), in Documents on the Constitutional Relationship of Puerto Rico and the United States 199 (M. Ramirez Lavandero ed., 3d ed. 1988). Finally, the constitution became law, in the manner Congress had specified, when the convention formally accepted those conditions and the governor “issue[d] a proclamation to that effect.” Ch. 567, 66Stat. 328. The Puerto Rico Constitution created a new political entity, the Commonwealth of Puerto Rico—or, in Spanish, Estado Libre Asociado de Puerto Rico. See P. R. Const., Art. I, §1. Like the U. S. Constitution, it divides political power into three branches—the “legislative, judicial and executive.” Art. I, §2. And again resonant of American founding principles, the Puerto Rico Constitution describes that tripartite government as “republican in form” and “subordinate to the sovereignty of the people of Puerto Rico.” Ibid. The Commonwealth’s power, the Constitution proclaims, “emanates from the people and shall be exercised in accordance with their will, within the terms of the compact agreed upon between the people of Puerto Rico and the United States.” Art. I, §1. B We now leave the lofty sphere of constitutionalism for the grittier precincts of criminal law. Respondents Luis Sánchez Valle and Jaime Gómez Vázquez (on separate occasions) each sold a gun to an undercover police officer. Commonwealth prosecutors indicted them for, among other things, selling a firearm without a permit in violation of the Puerto Rico Arms Act of 2000. See 25 Laws P. R. Ann. §458 (2008). While those charges were pending, federal grand juries indicted Sánchez Valle and Gómez Vázquez, based on the same transactions, for violations of analogous U. S. gun trafficking statutes. See 18 U. S. C. §§922(a)(1)(A), 923(a), 924(a)(1)(D), 924(a)(2). Both defendants pleaded guilty to those federal charges. Following their pleas, Sánchez Valle and Gómez Vázquez moved to dismiss the pending Commonwealth charges on double jeopardy grounds. The prosecutors in both cases opposed those motions, arguing that Puerto Rico and the United States are different sovereigns for double jeopardy purposes, and so could bring successive prosecutions against each of the two defendants. The trial courts rejected that view and dismissed the charges. See App. to Pet. for Cert. 307a–352a. But the Puerto Rico Court of Appeals, after consolidating the two cases, reversed those decisions. See id., at 243a–306a. The Supreme Court of Puerto Rico granted review and held that Puerto Rico’s gun sale prosecutions violated the Double Jeopardy Clause. See id., at 1a–70a. The majority reasoned that, under this Court’s dual-sovereignty doctrine, “what is crucial” is “[t]he ultimate source” of Puerto Rico’s power to prosecute. Id., at 19a; see id., at 20a (“The use of the word ‘sovereignty’ in other contexts and for other purposes is irrelevant”). Because that power originally “derived from the United States Congress”—i.e., the same source on which federal prosecutors rely—the Commonwealth could not retry Sánchez Valle and Gómez Vázquez for unlawfully selling firearms. Id., at 66a. Three justices disagreed, believing that the Commonwealth and the United States are separate sovereigns. See id., at 71a–242a. We granted certiorari, 576 U. S. ___ (2015), to determine whether the Double Jeopardy Clause bars the Federal Government and Puerto Rico from successively prosecuting a defendant on like charges for the same conduct. We hold that it does, and so affirm. II A This case involves the dual-sovereignty carve-out from the Double Jeopardy Clause. The ordinary rule under that Clause is that a person cannot be prosecuted twice for the same offense. See U. S. Const., Amdt. 5 (“nor shall any person be subject for the same offence to be twice put in jeopardy of life or limb”).[1] But two prosecutions, this Court has long held, are not for the same offense if brought by different sovereigns—even when those actions target the identical criminal conduct through equivalent criminal laws. See, e.g., United States v. Lanza, 260 U. S. 377, 382 (1922) . As we have put the point: “[W]hen the same act transgresses the laws of two sovereigns, it cannot be truly averred that the offender has been twice punished for the same offence; but only that by one act he has committed two offences.” Heath v. Alabama, 474 U. S. 82, 88 (1985) (internal quotation marks omitted). The Double Jeopardy Clause thus drops out of the picture when the “entities that seek successively to prosecute a defendant for the same course of conduct [are] separate sovereigns.” Ibid. Truth be told, however, “sovereignty” in this context does not bear its ordinary meaning. For whatever reason, the test we have devised to decide whether two governments are distinct for double jeopardy purposes overtly disregards common indicia of sovereignty. Under that standard, we do not examine the “extent of control” that “one prosecuting authority [wields] over the other.” Wheeler, 435 U. S., at 320. The degree to which an entity exercises self-governance—whether autonomously managing its own affairs or continually submitting to outside direction—plays no role in the analysis. See Shell Co., 302 U. S., at 261–262, 264–266. Nor do we care about a government’s more particular ability to enact and enforce its own criminal laws. See Waller v. Florida, 397 U. S. 387 –395 (1970). In short, the inquiry (despite its label) does not probe whether a government possesses the usual attributes, or acts in the common manner, of a sovereign entity.[2] Rather, as Puerto Rico itself acknowledges, our test hinges on a single criterion: the “ultimate source” of the power undergirding the respective prosecutions. Wheeler, 435 U. S., at 320; see Brief for Petitioner 26. Whether two prosecuting entities are dual sovereigns in the double jeopardy context, we have stated, depends on “whether [they] draw their authority to punish the offender from distinct sources of power.” Heath, 474 U. S., at 88. The inquiry is thus historical, not functional—looking at the deepest wellsprings, not the current exercise, of prosecutorial authority. If two entities derive their power to punish from wholly independent sources (imagine here a pair of parallel lines), then they may bring successive prosecutions. Conversely, if those entities draw their power from the same ultimate source (imagine now two lines emerging from a common point, even if later diverging), then they may not.[3] Under that approach, the States are separate sovereigns from the Federal Government (and from one another). See Abbate v. United States, 359 U. S. 187, 195 (1959) ; Bartkus v. Illinois, 359 U. S. 121 –137 (1959); Heath, 474 U. S., at 88. The States’ “powers to undertake criminal prosecutions,” we have explained, do not “derive[ ] . . . from the Federal Government.” Id., at 89. Instead, the States rely on “authority originally belonging to them before admission to the Union and preserved to them by the Tenth Amendment.” Ibid.; see U. S. Const., Amdt. 10 (“The powers not delegated to the United States by the Constitution . . . are reserved to the States”); Blatchford v. Native Village of Noatak, 501 U. S. 775, 779 (1991) (noting that the States “entered the [Union] with their sovereignty intact”). Said otherwise: Prior to forming the Union, the States possessed “separate and independent sources of power and authority,” which they continue to draw upon in enacting and enforcing criminal laws. Heath, 474 U. S., at 89. State prosecutions therefore have their most ancient roots in an “inherent sovereignty” unconnected to, and indeed pre-existing, the U. S. Congress. Ibid.[4] For similar reasons, Indian tribes also count as separate sovereigns under the Double Jeopardy Clause. Originally, this Court has noted, “the tribes were self-governing sovereign political communities,” possessing (among other capacities) the “inherent power to prescribe laws for their members and to punish infractions of those laws.” Wheeler, 435 U. S., at 322–323. After the formation of the United States, the tribes became “domestic dependent nations,” subject to plenary control by Congress—so hardly “sovereign” in one common sense. United States v.Lara, 541 U. S. 193, 204 (2004) (quoting Cherokee Nation v. Georgia, 5 Pet. 1, 17 (1831)); see Santa Clara Pueblo v. Martinez, 436 U. S. 49, 56 (1978) (“Congress has plenary authority to limit, modify or eliminate the [tribes’] powers of local self-government”). But unless and until Congress withdraws a tribal power—including the power to prosecute—the Indian community retains that authority in its earliest form. See Wheeler, 435 U. S., at 323. The “ultimate source” of a tribe’s “power to punish tribal offenders” thus lies in its “primeval” or, at any rate, “pre-existing” sovereignty: A tribal prosecution, like a State’s, is “attributable in no way to any delegation . . . of federal authority.” Id., at 320, 322, 328; Santa Clara Pueblo, 436 U. S., at 56. And that alone is what matters for the double jeopardy inquiry. Conversely, this Court has held that a municipality cannot qualify as a sovereign distinct from a State—no matter how much autonomy over criminal punishment the city maintains. See Waller, 397 U. S., at 395. Florida law, we recognized in our pivotal case on the subject, treated a municipality as a “separate sovereign entit[y]” for all relevant real-world purposes: The city possessed broad home-rule authority, including the power to enact criminal ordinances and prosecute offenses. Id., at 391. But that functional control was not enough to escape the double jeopardy bar; indeed, it was wholly beside the point. The crucial legal inquiry was backward-looking: Did the city and State ultimately “derive their powers to prosecute from independent sources of authority”? Heath, 474 U. S., at 90 (describing Waller’s reasoning). Because the municipality, in the first instance, had received its power from the State, those two entities could not bring successive prosecutions for a like offense. And most pertinent here, this Court concluded in the early decades of the last century that U. S. territories—including an earlier incarnation of Puerto Rico itself—are not sovereigns distinct from the United States. In Grafton v. United States, 206 U. S. 333, 355 (1907) , we held that the Philippine Islands (then a U. S. territory, also acquired in the Spanish-American War) could not prosecute a defendant for murder after a federal tribunal had acquitted him of the same crime. We reasoned that whereas “a State does not derive its powers from the United States,” a territory does: The Philippine courts “exert[ed] all their powers by authority of” the Federal Government. Id., at 354. And then, in Shell Co., we stated that “[t]he situation [in Puerto Rico] was, in all essentials, the same.” 302 U. S., at 265. Commenting on a Puerto Rican statute that overlapped with a federal law, we explained that this “legislative duplication [gave] rise to no danger of a second prosecution” because “the territorial and federal laws [were] creations emanating from the same sovereignty.” Id., at 264; see also Heath, 474 U. S., at 90 (notingthat federal and territorial prosecutors “d[o] not derive their powers to prosecute from independent sources of authority”).[5] B With that background established, we turn to the question presented: Do the prosecutorial powers belonging to Puerto Rico and the Federal Government derive from wholly independent sources? See Brief for Petitioner 26–28 (agreeing with that framing of the issue). If so, the criminal charges at issue here can go forward; but if not, not. In addressing that inquiry, we do not view our decisions in Grafton and Shell Co. as, in and of themselves, controlling. Following 1952, Puerto Rico became a new kind of political entity, still closely associated with the United States but governed in accordance with, and exercising self-rule through, a popularly ratified constitution. The magnitude of that change requires us to consider the dual-sovereignty question anew. And yet the result we reach, given the legal test we apply, ends up the same. Puerto Rico today has a distinctive, indeed exceptional, status as a self-governing Commonwealth. But our approach is historical. And if we go back as far as our doctrine demands—to the “ultimate source” of Puerto Rico’s prosecutorial power, Wheeler, 435 U. S., at 320—we once again discover the U. S. Congress. Recall here the events of the mid-20th century—when Puerto Rico, just as petitioner contends, underwent a profound change in its political system. See Brief for Petitioner 1–2 (“[T]he people of Puerto Rico[ ] engaged in an exercise of popular sovereignty . . . by adopting their own Constitution establishing their own government to enact their own laws”); supra, at 3–4. At that time, Congress enacted Public Law 600 to authorize Puerto Rico’s adoption of a constitution, designed to replace the federal statute that then structured the island’s governance. The people of Puerto Rico capitalized on that opportunity, calling a constitutional convention and overwhelmingly approving the charter it drafted. Once Congress approved that proposal—subject to several important conditions accepted by the convention—the Commonwealth, a new political entity, came into being. Those constitutional developments were of great significance—and, indeed, made Puerto Rico “sovereign” in one commonly understood sense of that term. As this Court has recognized, Congress in 1952 “relinquished its control over [the Commonwealth’s] local affairs[,] grant[ing]Puerto Rico a measure of autonomy comparable to that possessed by the States.” Examining Bd. of Engineers, Architects and Surveyors v. Flores de Otero, 426 U. S. 572, 597 (1976) ; see id., at 594 (“[T]he purpose of Congress in the 1950 and 1952 legislation was to accord to Puerto Rico the degree of autonomy and independence normally associ-ated with States of the Union”); Rodriguez v. Popular Demo-cratic Party, 457 U. S. 1, 8 (1982) (“Puerto Rico, like a state, is an autonomous political entity, sovereign over matters not ruled by the [Federal] Constitution” (internal quotation marks omitted)). That newfound authority, including over local criminal laws, brought mutual benefit to the Puerto Rican people and the entire United States. See Brief for United States as Amicus Curiae 3. And if our double jeopardy decisions hinged on measuring an entity’s self-governance, the emergence of the Commonwealth would have resulted as well in the capacity to bring the kind of successive prosecutions attempted here. But as already explained, the dual-sovereignty test we have adopted focuses on a different question: not on the fact of self-rule, but on where it came from. See supra, at 7–8. We do not care, for example, that the States pres-ently exercise autonomous control over criminal law andother local affairs; instead, we treat them as separate sovereigns because they possessed such control as an original matter, rather than deriving it from the Federal Government. See supra, at 8–9. And in identifying a prosecuting entity’s wellspring of authority, we have insisted on going all the way back—beyond the immediate, or even an intermediate, locus of power to what we have termed the “ultimate source.” Wheeler, 435 U. S., at 320. That is why we have emphasized the “inherent,” “primeval,” and “pre-existing” capacities of the tribes and States—the power they enjoyed prior to the Union’s formation. Id., at 322–323, 328; Heath, 474 U. S., at 90; Santa Clara Pueblo, 436 U. S., at 56; see supra, at 8–10. And it is why cities fail our test even when they enact and enforce their own criminal laws under their own, popu-larly ratified charters: Because a State must initially authorize any such charter, the State is the furthest-back source of prosecutorial power. See Waller, 397 U. S., at 391–394; supra, at 10. On this settled approach, Puerto Rico cannot benefit from our dual-sovereignty doctrine. For starters, no one argues that when the United States gained possession of Puerto Rico, its people possessed independent prosecuto-rial power, in the way that the States or tribes did upon becoming part of this country. Puerto Rico was until then a colony “under Spanish sovereignty.” Treaty of Paris, Art. 2, 30Stat. 1755. And local prosecutors in the ensuing decades, as petitioner itself acknowledges, exercised only such power as was “delegated by Congress” through fed-eral statutes. Brief for Petitioner 28; see Shell Co., 302 U. S., at 264–265; supra, at 10–11. Their authority derived from, rather than pre-existed association with, the Federal Government. And contrary to petitioner’s claim, Puerto Rico’s transformative constitutional moment does not lead to a different conclusion. True enough, that the Commonwealth’s power to enact and enforce criminal law now proceeds, just as petitioner says, from the Puerto Rico Constitution as “ordain[ed] and establish[ed]” by “the people.” P. R. Const., Preamble; see Brief for Petitioner 28–30. But that makes the Puerto Rican populace only the most immediate source of such authority—and that is not what our dual-sovereignty decisions make relevant. Back of the Puerto Rican people and their Constitution, the “ultimate” source of prosecutorial power remains the U. S. Congress, just as back of a city’s charter lies a state government. Wheeler, 435 U. S., at 320. Congress, in Public Law 600, authorized Puerto Rico’s constitution-making process in the first instance; the people of a territory could not legally have initiated that process on their own. See, e.g., Simms v. Simms, 175 U. S. 162, 168 (1899) . And Congress, in later legislation, both amended the draft charter and gave it the indispensable stamp of approval; popular ratification, however meaningful, could not have turned the convention’s handiwork into law.[6] Put simply, Congress conferred the authority to create the Puerto Rico Constitution, which in turn confers the authority to bring criminal charges. That makes Congress the original source of power for Puerto Rico’s prosecutors—as it is for the Fed-eral Government’s. The island’s Constitution, significant though it is, does not break the chain. Petitioner urges, in support of its different view, that Congress itself recognized the new Constitution as “a democratic manifestation of the [people’s] will,” Brief for Petitioner 2—but far from disputing that point, we readily acknowledge it to be so. As petitioner notes, Public Law 600 affirmed the “principle of government by consent” and offered the Puerto Rican public a “compact,” under which they could “organize a government pursuant to a constitution of their own adoption.” §1, 64Stat. 319; see Brief for Petitioner 2, 29; supra, at 3. And the Constitution that Congress approved, as petitioner again underscores, declares that “[w]e, the people” of Puerto Rico, “create” the Commonwealth—a new political entity, “republican in form,” in which the people’s will is “sovereign[ ]” over the government. P. R. Const., Preamble and Art. I, §§1–2; see Brief for Petitioner 2, 29–30; supra, at 4. With thatconsented-to language, Congress “allow[ed] the people ofPuerto Rico,” in petitioner’s words, to begin a new chapter of democratic self-governance. Reply Brief 20. All that separates our view from petitioner’s is what that congressional recognition means for Puerto Rico’s ability to bring successive prosecutions. We agree that Congress has broad latitude to develop innovative approaches to territorial governance, see U. S. Const., Art. IV, §3, cl. 2; that Congress may thus enable a territory’s people to make large-scale choices about their own political institutions; and that Congress did exactly that in enacting Public Law 600 and approving the Puerto Rico Constitution—prime examples of what Felix Frankfurter once termed “inventive statesmanship” respecting the island. Memorandum for the Secretary of War, in Hearings on S. 4604 before the Senate Committee on Pacific Islands and Porto Rico, 63d Cong., 2d Sess., 22 (1914); see Reply Brief 18–20. But one power Congress does not have, just in the nature of things: It has no capacity, no magic wand or airbrush, to erase or otherwise rewrite its own foundational role in conferring political authority. Or otherwise said, the delegator cannot make itself any less so—no matter how much authority it opts to hand over. And our dual-sovereignty test makes this historical fact dispositive: If an entity’s authority to enact and enforce criminal law ultimately comes from Congress, then it cannot follow a federal prosecution with its own. That is true of Puerto Rico, because Congress authorized and approved its Constitution, from which prosecutorial power now flows. So the Double Jeopardy Clause bars both Puerto Rico and the United States from prosecuting a single person for the same conduct under equivalent criminal laws. III Puerto Rico boasts “a relationship to the United States that has no parallel in our history.” Examining Bd., 426 U. S., at 596. And since the events of the early 1950’s, an integral aspect of that association has been the Commonwealth’s wide-ranging self-rule, exercised under its own Constitution. As a result of that charter, Puerto Rico today can avail itself of a wide variety of futures. But for purposes of the Double Jeopardy Clause, the future is not what matters—and there is no getting away from the past. Because the ultimate source of Puerto Rico’s prosecutorial power is the Federal Government—because when we trace that authority all the way back, we arrive at the doorstep of the U. S. Capitol—the Commonwealth and the United States are not separate sovereigns. That means the two governments cannot “twice put” respondents Sánchez Valle and Gómez Vázquez “in jeopardy” for the “same offence.” U. S. Const., Amdt. 5. We accordingly affirm the judgment of the Supreme Court of Puerto Rico. It is so ordered.Notes 1 Because the parties in this case agree that the Double Jeopardy Clause applies to Puerto Rico, we have no occasion to consider that question here. See Brief for Petitioner 19–21; Brief for Respondents20, n. 4; see also Brief for United States as Amicus Curiae 10, n. 1 (concurring). 2 The dissent, ignoring our longstanding precedent to the contrary, see supra, at 6–7; infra, at 7–11, advances an approach of just this stripe: Its seven considerations all go to the question whether the Commonwealth, by virtue of Public Law 600, gained “the sovereign authority to enact and enforce” its own criminal laws. Post, at 5 (opinion of Breyer, J.). Our disagreement with the dissent arises entirely from its use of this test. If the question is whether, after the events of 1950–1952, Puerto Rico had authority to enact and enforce its own criminal laws (or, slightly differently phrased, whether Congress then decided that it should have such autonomy), the answer (all can and do agree) is yes. See infra, at 13–17. But as we now show, that is not the inquiry our double jeopardy law has made relevant: To the contrary, we have rejected that approach again and again—and so reached results inconsistent with its use. See, e.g., Heath v. Alabama, 474 U. S. 82 –91 (1985); Waller v. Florida, 397 U. S. 387 –395 (1970); see infra, at 7–11. 3 The Court has never explained its reasons for adopting this historical approach to the dual-sovereignty doctrine. It may appear counterintuitive, even legalistic, as compared to an inquiry focused on a governmental entity’s functional autonomy. But that alternative would raise serious problems of application. It would require deciding exactly how much autonomy is sufficient for separate sovereignty and whether a given entity’s exercise of self-rule exceeds that level. The results, we suspect, would often be uncertain, introducing error and inconsistency into our double jeopardy law. By contrast, as we go on to show, the Court has easily applied the “ultimate source” test to classify broad classes of governments as either sovereign or not for purposes of barring retrials. See infra, at 8–11. 4 Literalists might object that only the original 13 States can claim such an independent source of authority; for the other 37, Congress played some role in establishing them as territories, authorizing or approving their constitutions, or (at the least) admitting them to the Union. See U. S. Const., Art. IV, §3, cl. 1 (“New States may be admitted by the Congress into this Union”). And indeed, that is the tack the dissent takes. See post, at 3–4 (claiming that for this reason the Federal Government is “the ‘source’ of [later-admitted] States’ legislative powers”). But this Court long ago made clear that a new State, upon entry, necessarily becomes vested with all the legal characteristics and capabilities of the first 13. See Coyle v. Smith, 221 U. S. 559, 566 (1911) (noting that the very meaning of “ ‘a State’ is found in the powers possessed by the original States which adopted the Constitution”). That principle of “equal footing,” we have held, is essential to ensure that the nation remains “a union of States[ alike] in power, dignity and authority, each competent to exert that residuum of sovereignty not delegated to the United States.” Id., at 567; see Northwest Austin Municipal Util. Dist. No. One v. Holder, 557 U. S. 193, 203 (2009) (referring to the “fundamental principle of equal sovereignty” among the States). Thus, each later-admitted State exercises its authority to enact and enforce criminal laws by virtue not of congressional grace, but of the independent powers that its earliest counterparts both brought to the Union and chose to maintain. See Coyle, 221 U. S., at 573 (“[W]hen a new State is admitted into the Union, it is so admitted with all the powers of sovereignty and jurisdiction which pertain to the original States”). The dissent’s contrary view—that, say, Texas’s or California’s powers (including the power to make and enforce criminal law) derive from the Federal Government—contradicts the most fundamental conceptual premises of our constitutional order, indeed the very bedrock of our Union. 5 The dissent’s theory, see supra, at 7, n. 2, cannot explain any of these (many) decisions, whether involving States, Indian tribes, cities, or territories. We have already addressed the dissent’s misunderstanding with respect to the States, including the later-admitted ones. See supra, at 8, and n. 4. This Court’s reasoning could not have been plainer: The States (all of them) are separate sovereigns for double jeopardy purposes not (as the dissent claims) because they exercise authority over criminal law, but instead because that power derives from a source independent of the Federal Government. See Heath, 474 U. S., at 89. So too for the tribes, see supra, at 9–10; and, indeed, here the dissent’s contrary reasoning is deeply disturbing. According to the dissent, Congress is in fact “the ‘source’ of the Indian tribes’ criminal-enforcement power” because it has elected not to disturb the exercise of that authority. Post, at 5. But beginning with Chief Justice Marshall and continuing for nearly two centuries, this Court has held firm and fast to the view that Congress’s power over Indian affairs does nothing to gainsay the profound importance of the tribes’ pre-existing sovereignty. See Worcester v. Georgia, 6 Pet. 515, 559–561 (1832); Talton v. Mayes, 163 U. S. 376, 384 (1896) ; Michigan v. Bay Mills Indian Community, 572 U. S. ___, ___–___ (2014) (slip op., at 4–5). And once again, we have stated in no uncertain terms that the tribes are separate sovereigns precisely because of that inherent authority. See Wheeler, 435 U. S., at 328. Next, the dissent cannot (and does not even try to) explain our rule that a municipality is not a separate sovereign from a State. See supra, at 10. As this Court has explicitly recognized, many cities have (in the words of the dissent’s test) wide-ranging “authority to make and enforce [their] own criminal laws,” post, at 5; still, they cannot undertake successive prosecutions—because they received that power from state governments, see Waller, 397 U. S., at 395. And likewise (finally), the dissent fails to face up to our decisions that the territories are not distinct sovereigns from the United States because the powers they exercise are delegations from Congress. See Grafton v. United States, 206 U. S. 333, 355 (1907) ; supra, at 10–11. That, of course, is what makes them different from the current Philippines, see post, at 2–3, whose relevance here is hard to fathom. As an independent nation, the Philippines wields prosecutorial power that is not traceable to any congressional conferral of authority. And that, to repeat, is what matters: If an entity’s capacity to make and enforce criminal law ultimately comes from another government, then the two are not separate sovereigns for double jeopardy purposes. 6 Petitioner’s own statements are telling as to the role Congress necessarily played in this constitutional process. See, e.g., Reply Brief 1–2 (“Pursuant to Congress’ invitation, and with Congress’ consent, the people of Puerto Rico engaged in an exercise of popular sovereignty”); id., at 7 (“The Commonwealth’s legal cornerstone is Public Law 600”); Tr. of Oral Arg. 19 (describing the adoption of the Puerto Rico Constitution as “pursuant to the invitation of Congress and with the blessing of Congress”).

579.US.2015_15-138

The Racketeer Influenced and Corrupt Organizations Act (RICO) prohibits certain activities of organized crime groups in relation to an enterprise. RICO makes it a crime to invest income derived from a pattern of racketeering activity in an enterprise “which is engaged in, or the activities of which affect, interstate or foreign commerce,” 18 U. S. C. §1962(a); to acquire or maintain an interest in an enterprise through a pattern of racketeering activity, §1962(b); to conduct an enterprise’s affairs through a pattern of racketeering activity, §1962(c); and to conspire to violate any of the other three prohibitions, §1962(d). RICO also provides a civil cause of action for “[a]ny person injured in his business or property by reason of a violation” of those prohibitions. §1964(c). Respondents (the European Community and 26 of its member states) filed suit under RICO, alleging that petitioners (RJR Nabisco and related entities (collectively RJR)) participated in a global money-laundering scheme in association with various organized crime groups. Under the alleged scheme, drug traffickers smuggled narcotics into Europe and sold them for euros that—through transactions involving black-market money brokers, cigarette importers, and wholesalers—were used to pay for large shipments of RJR cigarettes into Europe. The complaint alleged that RJR violated §§1962(a)–(d) by engaging in a pattern of racketeering activity that included numerous predicate acts of money laundering, material support to foreign terrorist organizations, mail fraud, wire fraud, and violations of the Travel Act. The District Court granted RJR’s motion to dismiss on the ground that RICO does not apply to racketeering activity occurring outside U. S. territory or to foreign enterprises. The Second Circuit reinstated the claims, however, concluding that RICO applies extraterritorially to the same extent as the predicate acts of racketeering that underlie the alleged RICO violation, and that certain predicates alleged in this case expressly apply extraterritorially. In denying rehearing, the court held further that RICO’s civil action does not require a domestic injury, but permits recovery for a foreign injury caused by the violation of a predicate statute that applies extraterritorially. Held: 1. The law of extraterritoriality provides guidance in determining RICO’s reach to events outside the United States. The Court applies a canon of statutory construction known as the presumption against extraterritoriality: Absent clearly expressed congressional intent to the contrary, federal laws will be construed to have only domestic application. Morrison v. National Australia Bank Ltd., 561 U. S. 247 . Morrison and Kiobel v. Royal Dutch Petroleum Co., 569 U. S. ___, reflect a two-step framework for analyzing extraterritoriality issues. First, the Court asks whether the presumption against extraterritoriality has been rebutted—i.e., whether the statute gives a clear, affirmative indication that it applies extraterritorially. This question is asked regardless of whether the particular statute regulates conduct, affords relief, or merely confers jurisdiction. If, and only if, the statute is not found extraterritorial at step one, the Court moves to step two, where it examines the statute’s “focus” to determine whether the case involves a domestic application of the statute. If the conduct relevant to the statute’s focus occurred in the United States, then the case involves a permissible domestic application even if other conduct occurred abroad; but if the relevant conduct occurred in a foreign country, then the case involves an impermissible extraterritorial application regardless of whether other conduct occurred in U. S. territory. In the event the statute is found to have clear extraterritorial effect at step one, then the statute’s scope turns on the limits Congress has or has not imposed on the statute’s foreign application, and not on the statute’s “focus.” Pp. 7–10. 2. The presumption against extraterritoriality has been rebutted with respect to certain applications of RICO’s substantive prohibitions in §1962. Pp. 10–18. (a) RICO defines racketeering activity to include a number of predicates that plainly apply to at least some foreign conduct, such as the prohibition against engaging in monetary transactions in criminally derived property, §1957(d)(2), the prohibitions against the assassination of Government officials, §§351(i), 1751(k), and the prohibition against hostage taking, §1203(b). Congress has thus given a clear, affirmative indication that §1962 applies to foreign racketeering activity—but only to the extent that the predicates alleged in a particular case themselves apply extraterritorially. This fact is determinative as to §§1962(b) and (c), which both prohibit the employment of a pattern of racketeering. But §1962(a), which targets certain uses of income derived from a pattern of racketeering, arguably extends only to domestic uses of that income. Because the parties have not focused on this issue, and because its resolution does not affect this case, it is assumed that respondents have pleaded a domestic investment of racketeering income in violation of §1962(a). It is also assumed that the extraterritoriality of a violation of RICO’s conspiracy provision, §1962(d), tracks that of the RICO provision underlying the alleged conspiracy. Pp. 10–14. (b) RJR contends that RICO’s “focus” is its enterprise element, which gives no clear indication of extraterritorial effect. But focus is considered only when it is necessary to proceed to the inquiry’s second step. See Morrison, supra, at 267, n. 9. Here, however, there is a clear indication at step one that at least §§1962(b) and (c) apply to all transnational patterns of racketeering, subject to the stated limitation. A domestic enterprise requirement would lead to difficult line-drawing problems and counterintuitive results, such as excluding from RICO’s reach foreign enterprises that operate within the United States. Such troubling consequences reinforce the conclusion that Congress intended the §§1962(b) and (c) prohibitions to apply extraterritorially in tandem with the underlying predicates, without regard to the locus of the enterprise. Of course, foreign enterprises will qualify only if they engage in, or significantly affect, commerce directly involving the United States. Pp. 14–17. (c) Applying these principles here, respondents’ allegations that RJR violated §§1962(b) and (c) do not involve an impermissibly extraterritorial application of RICO. The Court assumes that the alleged pattern of racketeering activity consists entirely of predicate offenses that were either committed in the United States or committed in a foreign country in violation of a predicate statute that applies extraterritorially. The alleged enterprise also has a sufficient tie to U. S. commerce, as its members include U. S. companies and its activities depend on sales of RJR’s cigarettes conducted through “the U. S. mails and wires,” among other things. Pp. 17–18. 3. Irrespective of any extraterritoriality of §1962’s substantive provisions, §1964(c)’s private right of action does not overcome the presumption against extraterritoriality, and thus a private RICO plaintiff must allege and prove a domestic injury. Pp. 18–27. (a) The Second Circuit reasoned that the presumption against extraterritoriality did not apply to §1964(c) independently of its application to §1962’s substantive provisions because §1964(c) does not regulate conduct. But this view was rejected in Kiobel, 569 U. S., at ___, and the logic of that decision requires that the presumption be applied separately to RICO’s cause of action even though it has been overcome with respect to RICO’s substantive prohibitions. As in other contexts, allowing recovery for foreign injuries in a civil RICO action creates a danger of international friction that militates against recognizing foreign-injury claims without clear direction from Congress. Respondents, in arguing that such concerns are inapplicable here because the plaintiffs are not foreign citizens seeking to bypass their home countries’ less generous remedies but are foreign countries themselves, forget that this Court’s interpretation of §1964(c)’s injury requirement will necessarily govern suits by nongovernmental plaintiffs. The Court will not forgo the presumption against extraterritoriality to permit extraterritorial suits based on a case-by-case inquiry that turns on or looks to the affected sovereign’s consent. Nor will the Court adopt a double standard that would treat suits by foreign sovereigns more favorably than other suits. Pp. 18–22. (b) Section 1964(c) does not provide a clear indication that Congress intended to provide a private right of action for injuries suffered outside of the United States. It provides a cause of action to “[a]ny person injured in his business or property” by a violation of §1962, but neither the word “any” nor the reference to injury to “business or property” indicates extraterritorial application. Respondents’ arguments to the contrary are unpersuasive. In particular, while they are correct that RICO’s private right of action was modeled after §4 of the Clayton Act, which allows recovery for injuries suffered abroad as a result of antitrust violations, see Pfizer Inc. v. Government of India, 434 U. S. 308 –315, this Court has declined to transplant features of the Clayton Act’s cause of action into the RICO context where doing so would be inappropriate. Cf. Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479 . There is good reason not to do so here. Most importantly, RICO lacks the very language that the Court found critical to its decision in Pfizer, namely, the Clayton Act’s definition of a “person” who may sue, which “explicitly includes ‘corporations and associations existing under or authorized by . . . the laws of any foreign country,’ ” 434 U. S., at 313. Congress’s more recent decision to exclude from the antitrust laws’ reach most conduct that “causes only foreign injury,” F. Hoffmann-La Roche Ltd v. Empagran S. A., 542 U. S. 155 , also counsels against importing into RICO those Clayton Act principles that are at odds with the Court’s current extraterritoriality doctrine. Pp. 22–27. (c) Section 1964(c) requires a civil RICO plaintiff to allege and prove a domestic injury to business or property and does not allow recovery for foreign injuries. Respondents waived their domestic injury damages claims, so the District Court dismissed them with prejudice. Their remaining RICO damages claims therefore rest entirely on injury suffered abroad and must be dismissed. P. 27. 764 F. 3d 129, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy and Thomas, JJ., joined, and in which Ginsburg, Breyer, and Kagan, JJ., joined as to Parts I, II, and III. Ginsburg, J., filed an opinion concurring in part, dissenting in part, and dissenting from the judgment, in which Breyer and Kagan, JJ., joined. Breyer, J., filed an opinion concurring in part, dissenting in part, and dissenting from the judgment. Sotomayor, J., took no part in the consideration or decision of the case.

The Racketeer Influenced and Corrupt Organizations Act (RICO), 18 U. S. C. §§1961–1968, created four new criminal offenses involving the activities of organized criminal groups in relation to an enterprise. §§1962(a)–(d). RICO also created a new civil cause of action for “[a]ny person injured in his business or property by reason of a violation” of those prohibitions. §1964(c). We are asked to decide whether RICO applies extraterritorially—that is, to events occurring and injuries suffered outside the United States. I A RICO is founded on the concept of racketeering activity. The statute defines “racketeering activity” to encompass dozens of state and federal offenses, known in RICO parlance as predicates. These predicates include any act “indictable” under specified federal statutes, §§1961(1)(B)–(C), (E)–(G), as well as certain crimes “chargeable” under state law, §1961(1)(A), and any offense involving bankruptcy or securities fraud or drug-related activity that is “punishable” under federal law, §1961(1)(D). A predicate offense implicates RICO when it is part of a “pattern of racketeering activity”—a series of related predicates that together demonstrate the existence or threat of continued criminal activity. H. J. Inc. v. Northwestern Bell Telephone Co., 492 U. S. 229, 239 (1989) ; see §1961(5) (specifying that a “pattern of racketeering activity” requires at least two predicates committed within 10 years of each other). RICO’s §1962 sets forth four specific prohibitions aimed at different ways in which a pattern of racketeering activ-ity may be used to infiltrate, control, or operate “a[n] en-terprise which is engaged in, or the activities of which affect, interstate or foreign commerce.” These prohibitions can be summarized as follows. Section 1962(a) makes it unlawful to invest income derived from a pattern of racketeering activity in an enterprise. Section 1962(b) makes it unlawful to acquire or maintain an interest in an enterprise through a pattern of racketeering activity. Section 1962(c) makes it unlawful for a person employed by or associated with an enterprise to conduct the enterprise’s affairs through a pattern of racketeering activity. Finally, §1962(d) makes it unlawful to conspire to violate any of the other three prohibitions.[1] Violations of §1962 are subject to criminal penalties, §1963(a), and civil proceedings to enforce those prohibitions may be brought by the Attorney General, §§1964(a)–(b). Separately, RICO creates a private civil cause of action that allows “[a]ny person injured in his business or property by reason of a violation of section 1962” to sue in federal district court and recover treble damages, costs, and attorney’s fees. §1964(c).[2] B This case arises from allegations that petitioners—RJR Nabisco and numerous related entities (collectively RJR)—participated in a global money-laundering scheme in association with various organized crime groups. Respondents—the European Community and 26 of its member states—first sued RJR in the Eastern District of New York in 2000, alleging that RJR had violated RICO. Over the past 16 years, the resulting litigation (spread over at least three separate actions, with this case the lone survivor) has seen multiple complaints and multiple trips up and down the federal court system. See 2011 WL 843957, *1–*2 (EDNY, Mar. 8, 2011) (tracing the procedural his-tory through the District Court’s dismissal of the present complaint). In the interest of brevity, we confine our discussion to the operative complaint and its journey to this Court. Greatly simplified, the complaint alleges a scheme in which Colombian and Russian drug traffickers smuggled narcotics into Europe and sold the drugs for euros that—through a series of transactions involving black-market money brokers, cigarette importers, and wholesalers—were used to pay for large shipments of RJR cigarettes into Europe. In other variations of this scheme, RJR allegedly dealt directly with drug traffickers and money launderers in South America and sold cigarettes to Iraq in violation of international sanctions. RJR is also said to have acquired Brown & Williamson Tobacco Corporation for the purpose of expanding these illegal activities. The complaint alleges that RJR engaged in a pattern of racketeering activity consisting of numerous acts of money laundering, material support to foreign terrorist organizations, mail fraud, wire fraud, and violations of the Travel Act. RJR, in concert with the other participants in the scheme, allegedly formed an association in fact that was engaged in interstate and foreign commerce, and therefore constituted a RICO enterprise that the complaint dubs the “RJR Money-Laundering Enterprise.” App. to Pet. for Cert. 238a, Complaint ¶158; see §1961(4) (defining an enterprise to include “any union or group of individuals associated in fact although not a legal entity”). Putting these pieces together, the complaint alleges that RJR violated each of RICO’s prohibitions. RJR allegedly used income derived from the pattern of racketeering to invest in, acquire an interest in, and operate the RJR Money-Laundering Enterprise in violation of §1962(a); acquired and maintained control of the enterprise through the pattern of racketeering in violation of §1962(b); operated the enterprise through the pattern of racketeering in violation of §1962(c); and conspired with other participants in the scheme in violation of §1962(d).[3] These violations allegedly harmed respondents in various ways, including through competitive harm to their state-owned cigarette businesses, lost tax revenue from black-market cigarette sales, harm to European financial institutions, currency instability, and increased law enforcement costs.[4] RJR moved to dismiss the complaint, arguing that RICO does not apply to racketeering activity occurring outside U. S. territory or to foreign enterprises. The District Court agreed and dismissed the RICO claims as impermissibly extraterritorial. 2011 WL 843957, at *7. The Second Circuit reinstated the RICO claims. It concluded that, “with respect to a number of offenses that constitute predicates for RICO liability and are alleged in this case, Congress has clearly manifested an intent that they apply extraterritorially.” 764 F. 3d 129, 133 (2014). “By incorporating these statutes into RICO as predicate racketeering acts,” the court reasoned, “Congress has clearly communicated its intention that RICO apply to extraterritorial conduct to the extent that extraterritorial violations of these statutes serve as the basis for RICO liability.” Id., at 137. Turning to the predicates alleged in the complaint, the Second Circuit found that they passed muster. The court concluded that the money laundering and material support of terrorism statutes expressly apply extraterritorially in the circumstances alleged in the complaint. Id., at 139–140. The court held that the mail fraud, wire fraud, and Travel Act statutes do not apply extraterritorially. Id., at 141. But it concluded that the complaint states domestic violations of those predicates because it “allege[s] conduct in the United States that satisfies every essential element” of those offenses. Id., at 142. RJR sought rehearing, arguing (among other things) that RICO’s civil cause of action requires a plaintiff to allege a domestic injury, even if a domestic pattern of racketeering or a domestic enterprise is not necessary to make out a violation of RICO’s substantive prohibitions. The panel denied rehearing and issued a supplemental opinion holding that RICO does not require a domestic injury. 764 F. 3d 149 (CA2 2014) (per curiam). If a foreign injury was caused by the violation of a predicate statute that applies extraterritorially, the court concluded, then the plaintiff may seek recovery for that injury under RICO. Id., at 151. The Second Circuit later denied rehearing en banc, with five judges dissenting. 783 F. 3d 123 (2015). The lower courts have come to different conclusions regarding RICO’s extraterritorial application. Compare 764 F. 3d 129 (case below) (holding that RICO may apply extraterritorially) with United States v. Chao Fan Xu, 706 F. 3d 965, 974–975 (CA9 2013) (holding that RICO does not apply extraterritorially; collecting cases). Because of this conflict and the importance of the issue, we granted certiorari. 576 U. S. ___ (2015). II The question of RICO’s extraterritorial application really involves two questions. First, do RICO’s substantive prohibitions, contained in §1962, apply to conduct that occurs in foreign countries? Second, does RICO’s private right of action, contained in §1964(c), apply to injuries that are suffered in foreign countries? We consider each of these questions in turn. To guide our inquiry, we begin by reviewing the law of extraterritoriality. It is a basic premise of our legal system that, in general, “United States law governs domestically but does not rule the world.” Microsoft Corp. v. AT&T Corp., 550 U. S. 437, 454 (2007) . This principle finds expression in a canon of statutory construction known as the presumption against extraterritoriality: Absent clearly expressed congressional intent to the contrary, federal laws will be construed to have only domestic application. Morrison v. National Australia Bank Ltd., 561 U. S. 247, 255 (2010) . The question is not whether we think “Congress would have wanted” a statute to apply to foreign conduct “if it had thoughtof the situation before the court,” but whether Congress has affirmatively and unmistakably instructed that the statute will do so. Id., at 261. “When a statute gives no clear indication of an extraterritorial application, it has none.” Id., at 255. There are several reasons for this presumption. Most notably, it serves to avoid the international discord that can result when U. S. law is applied to conduct in foreign countries. See, e.g., Kiobel v. Royal Dutch Petroleum Co., 569 U. S. ___, ___–___ (2013) (slip op., at 4–5); EEOC v. Arabian American Oil Co., 499 U. S. 244, 248 (1991) (Aramco); Benz v. Compania Naviera Hidalgo, S. A., 353 U. S. 138, 147 (1957) . But it also reflects the more prosaic “commonsense notion that Congress generally legislates with domestic concerns in mind.” Smith v. United States, 507 U. S. 197 , n. 5 (1993). We therefore apply the presumption across the board, “regardless of whether there is a risk of conflict between the American statute and a foreign law.” Morrison, supra, at 255. Twice in the past six years we have considered whether a federal statute applies extraterritorially. In Morrison, we addressed the question whether §10(b) of the Securities Exchange Act of 1934 applies to misrepresentations made in connection with the purchase or sale of securities traded only on foreign exchanges. We first examined whether §10(b) gives any clear indication of extraterritorial effect, and found that it does not. 561 U. S., at 262–265. We then engaged in a separate inquiry to determine whether the complaint before us involved a permissible domestic application of §10(b) because it alleged that some of the relevant misrepresentations were made in the United States. At this second step, we considered the “ ‘focus’ of congressional concern,” asking whether §10(b)’s focus is “the place where the deception originated” or rather “purchases and sale of securities in the United States.” Id., at 266. We concluded that the statute’s focus is on domestic securities transactions, and we therefore held that the statute does not apply to frauds in connection with foreign securities transactions, even if those frauds involve domestic misrepresentations. In Kiobel, we considered whether the Alien Tort Statute (ATS) confers federal-court jurisdiction over causes of action alleging international-law violations committed overseas. We acknowledged that the presumption against extraterritoriality is “typically” applied to statutes “regulating conduct,” but we concluded that the principles supporting the presumption should “similarly constrain courts considering causes of action that may be brought under the ATS.” 569 U. S., at ___ (slip op., at 5). We applied the presumption and held that the ATS lacks any clear indication that it extended to the foreign violations alleged in that case. Id., at ___–___ (slip op., at 7–14). Because “all the relevant conduct” regarding those violations “took place outside the United States,” id., at ___ (slip op., at 14), we did not need to determine, as we did in Morrison, the statute’s “focus.” Morrison and Kiobel reflect a two-step framework for analyzing extraterritoriality issues. At the first step, we ask whether the presumption against extraterritoriality has been rebutted—that is, whether the statute gives a clear, affirmative indication that it applies extraterritorially. We must ask this question regardless of whether the statute in question regulates conduct, affords relief, or merely confers jurisdiction. If the statute is not extraterritorial, then at the second step we determine whether the case involves a domestic application of the statute, and we do this by looking to the statute’s “focus.” If the conduct relevant to the statute’s focus occurred in the United States, then the case involves a permissible domestic application even if other conduct occurred abroad; but if the conduct relevant to the focus occurred in a foreign country, then the case involves an impermissible extraterritorial application regardless of any other conduct that occurred in U. S. territory. What if we find at step one that a statute clearly does have extraterritorial effect? Neither Morrison nor Kiobel involved such a finding. But we addressed this issue in Morrison, explaining that it was necessary to consider §10(b)’s “focus” only because we found that the statute does not apply extraterritorially: “If §10(b) did apply abroad, we would not need to determine which transnational frauds it applied to; it would apply to all of them (barring some other limitation).” 561 U. S., at 267, n. 9. The scope of an extraterritorial statute thus turns on the limits Congress has (or has not) imposed on the statute’s foreign application, and not on the statute’s “focus.”[5] III With these guiding principles in mind, we first consider whether RICO’s substantive prohibitions in §1962 may apply to foreign conduct. Unlike in Morrison and Kiobel, we find that the presumption against extraterritoriality has been rebutted—but only with respect to certain applications of the statute. A The most obvious textual clue is that RICO defines racketeering activity to include a number of predicates that plainly apply to at least some foreign conduct. These predicates include the prohibition against engaging in monetary transactions in criminally derived property, which expressly applies, when “the defendant is a United States person,” to offenses that “tak[e] place outside the United States.” 18 U. S. C. §1957(d)(2). Other examples include the prohibitions against the assassination of Government officials, §351(i) (“There is extraterritorial jurisdiction over the conduct prohibited by this section”); §1751(k) (same), and the prohibition against hostage taking, which applies to conduct that “occurred outside the United States” if either the hostage or the offender is a U. S. national, if the offender is found in the United States, or if the hostage taking is done to compel action by the U. S. Government, §1203(b). At least one predicate—the prohibition against “kill[ing] a national of the United States, while such national is outside the United States”—applies only to conduct occurring outside the United States. §2332(a). We agree with the Second Circuit that Congress’s incorporation of these (and other) extraterritorial predicates into RICO gives a clear, affirmative indication that §1962 applies to foreign racketeering activity—but only to the extent that the predicates alleged in a particular case themselves apply extraterritorially. Put another way, a pattern of racketeering activity may include or consist of offenses committed abroad in violation of a predicate statute for which the presumption against extraterritoriality has been overcome. To give a simple (albeit grim) example, a violation of §1962 could be premised on a pattern of killings of Americans abroad in violation of §2332(a)—a predicate that all agree applies extraterritorially—whether or not any domestic predicates are also alleged.[6] We emphasize the important limitation that foreign conduct must violate “a predicate statute that manifests an unmistakable congressional intent to apply extraterritorially.” 764 F. 3d, at 136. Although a number of RICO predicates have extraterritorial effect, many do not. The inclusion of some extraterritorial predicates does not mean that all RICO predicates extend to foreign conduct. This is apparent for two reasons. First, “when a statute provides for some extraterritorial application, the presumption against extraterritoriality operates to limit thatprovision to its terms.” Morrison, 561 U. S., at 265. Second, RICO defines as racketeering activity only acts that are “indictable” (or, what amounts to the same thing, “chargeable” or “punishable”) under one of the statutes identified in §1961(1). If a particular statute does not apply extraterritorially, then conduct committed abroad is not “indictable” under that statute and so cannot qualify as a predicate under RICO’s plain terms. RJR resists the conclusion that RICO’s incorporation of extraterritorial predicates gives RICO commensurate extraterritorial effect. It points out that “RICO itself” does not refer to extraterritorial application; only the underlying predicate statutes do. Brief for Petitioners 42. RJR thus argues that Congress could have intended to capture only domestic applications of extraterritorial predicates, and that any predicates that apply only abroad could have been “incorporated . . . solely for when such offenses are part of a broader pattern whose overall locus is domestic.” Id., at 43. The presumption against extraterritoriality does not require us to adopt such a constricted interpretation. While the presumption can be overcome only by a clear indication of extraterritorial effect, an express statement of extraterritoriality is not essential. “Assuredly context can be consulted as well.” Morrison, supra, at 265. Context is dispositive here. Congress has not expressly said that §1962(c) applies to patterns of racketeering activity in foreign countries, but it has defined “racketeering activ-ity”—and by extension a “pattern of racketeering activ-ity”—to encompass violations of predicate statutes that do expressly apply extraterritorially. Short of an explicit declaration, it is hard to imagine how Congress could have more clearly indicated that it intended RICO to have (some) extraterritorial effect. This unique structure makes RICO the rare statute that clearly evidences extraterritorial effect despite lacking an express statement of extraterritoriality. We therefore conclude that RICO applies to some foreign racketeering activity. A violation of §1962 may be based on a pattern of racketeering that includes predicate offenses committed abroad, provided that each of those offenses violates a predicate statute that is itself extraterritorial. This fact is determinative as to §1962(b) and §1962(c), both of which prohibit the employment of a pattern of racketeering. Although they differ as to the end for which the pattern is employed—to acquire or maintain control of an enterprise under subsection (b), or to conduct an enterprise’s affairs under subsection (c)—this difference is immaterial for extraterritoriality purposes. Section 1962(a) presents a thornier question. Unlike subsections (b) and (c), subsection (a) targets certain uses of income derived from a pattern of racketeering, not the use of the pattern itself. Cf. Anza v. Ideal Steel Supply Corp., 547 U. S. 451 –462 (2006). While we have no difficulty concluding that this prohibition applies to income derived from foreign patterns of racketeering (within the limits we have discussed), arguably §1962(a) extends only to domestic uses of the income. The Second Circuit did not decide this question because it found that respondents have alleged “a domestic investment of racketeering proceeds in the form of RJR’s merger in the United States with Brown & Williamson and investments in other U. S. operations.” 764 F. 3d, at 138, n. 5. RJR does not dispute the basic soundness of the Second Circuit’s reasoning, but it does contest the court’s reading of the complaint. See Brief for Petitioners 57–58. Because the parties have not focused on this issue, and because it makes no difference to our resolution of this case, see infra, at 27, we assume without deciding that respondents have pleaded a domestic investment of racketeering income in violation of §1962(a). Finally, although respondents’ complaint alleges a violation of RICO’s conspiracy provision, §1962(d), the parties’ briefs do not address whether this provision should be treated differently from the provision (§1962(a), (b), or (c)) that a defendant allegedly conspired to violate. We therefore decline to reach this issue, and assume without deciding that §1962(d)’s extraterritoriality tracks that of the provision underlying the alleged conspiracy. B RJR contends that, even if RICO may apply to foreign patterns of racketeering, the statute does not apply to foreign enterprises. Invoking Morrison’s discussion of the Exchange Act’s “focus,” RJR says that the “focus” of RICO is the enterprise being corrupted—not the pattern of racketeering—and that RICO’s enterprise element gives no clear indication of extraterritorial effect. Accordingly, RJR reasons, RICO requires a domestic enterprise. This argument misunderstands Morrison. As explained above, supra, at 9–10, only at the second step of the inquiry do we consider a statute’s “focus.” Here, however, there is a clear indication at step one that RICO applies extraterritorially. We therefore do not proceed to the “focus” step. The Morrison Court’s discussion of the statutory “focus” made this clear, stating that “[i]f §10(b) did apply abroad, we would not need to determine which transnational frauds it applied to; it would apply to all of them (barring some other limitation).” 561 U. S., at 267, n. 9. The same is true here. RICO—or at least §§1962(b) and (c)—applies abroad, and so we do not need to determine which transnational (or wholly foreign) patterns of racketeering it applies to; it applies to all of them, regardless of whether they are connected to a “foreign” or “domestic” enterprise. This rule is, of course, subject to the important limitation that RICO covers foreign predicate offenses only to the extent that the underlying predicate statutes are extraterritorial. But within those bounds, the location of the affected enterprise does not impose an independent constraint. It is easy to see why Congress did not limit RICO to domestic enterprises. A domestic enterprise requirement would lead to difficult line-drawing problems and counterintuitive results. It would exclude from RICO’s reach foreign enterprises—whether corporations, crime rings, other associations, or individuals—that operate within the United States. Imagine, for example, that a foreign corporation has operations in the United States and that one of the corporation’s managers in the United States conducts its U. S. affairs through a pattern of extortion and mail fraud. Such domestic conduct would seem to fall well within what Congress meant to capture in enacting RICO. Congress, after all, does not usually exempt foreigners acting in the United States from U. S. legal requirements. See 764 F. 3d, at 138 (“Surely the presumption against extraterritorial application of United States laws does not command giving foreigners carte blanche to violate the laws of the United States in the United States”). Yet RJR’s theory would insulate this scheme from RICO liability—both civil and criminal—because the enterprise at issue is a foreign, not domestic, corporation. Seeking to avoid this result, RJR offers that any “ ‘emissaries’ ” a foreign enterprise sends to the United States—such as our hypothetical U. S.-based corporate manager—could be carved off and considered a “distinct domestic enterprise” under an association-in-fact theory. Brief for Petitioners 40. RJR’s willingness to gerrymander the enterprise to get around its proposed domestic enterprise requirement is telling. It suggests that RJR is not really concerned about whether an enterprise is foreign or domestic, but whether the relevant conduct occurred here or abroad. And if that is the concern, then it is the pattern of racketeering activity that matters, not the enterprise. Even spotting RJR its “domestic emissary” theory, this approach would lead to strange gaps in RICO’s coverage. If a foreign enterprise sent only a single “emissary” to engage in racketeering in the United States, there could be no RICO liability because a single person cannot be both the RICO enterprise and the RICO defendant. Cedric Kushner Promotions, Ltd. v. King, 533 U. S. 158, 162 (2001) . RJR also offers no satisfactory way of determining whether an enterprise is foreign or domestic. Like the District Court, RJR maintains that courts can apply the “nerve center” test that we use to determine a corporation’s principal place of business for purposes of federal diversity jurisdiction. See Hertz Corp. v. Friend, 559 U. S. 77 (2010) ; 28 U. S. C. §1332(c)(1); 2011 WL 843957, at *5–*6. But this test quickly becomes meaningless if, as RJR suggests, a corporation with a foreign nerve center can, if necessary, be pruned into an association-in-fact enterprise with a domestic nerve center. The nerve center test, developed with ordinary corporate command structures in mind, is also ill suited to govern RICO association-in-fact enterprises, which “need not have a hierarchical structure or a ‘chain of command.’ ” Boyle v. United States, 556 U. S. 938, 948 (2009) . These difficulties are largely avoided if, as we conclude today, RICO’s extraterritorial effect is pegged to the extraterritoriality judgments Congress has made in the predicate statutes, often by providing precise instructions as to when those statutes apply to foreign conduct. The practical problems we have identified with RJR’s proposed domestic enterprise requirement are not, by themselves, cause to reject it. Our point in reciting these troubling consequences of RJR’s theory is simply to reinforce our conclusion, based on RICO’s text and context, that Congress intended the prohibitions in 18 U. S. C. §§1962(b) and (c) to apply extraterritorially in tandem with the underlying predicates, without regard to the locus of the enterprise. Although we find that RICO imposes no domestic enterprise requirement, this does not mean that every foreign enterprise will qualify. Each of RICO’s substantive prohibitions requires proof of an enterprise that is “engaged in, or the activities of which affect, interstate or foreign commerce.” §§1962(a), (b), (c). We do not take this reference to “foreign commerce” to mean literally all commerce occurring abroad. Rather, a RICO enterprise must engage in, or affect in some significant way, commerce directly involving the United States—e.g., commerce between the United States and a foreign country. Enterprises whose activities lack that anchor to U. S. commerce cannot sustain a RICO violation. C Applying these principles, we agree with the Second Circuit that the complaint does not allege impermissibly extraterritorial violations of §§1962(b) and (c).[7] The alleged pattern of racketeering activity consists of five basic predicates: (1) money laundering, (2) material support of foreign terrorist organizations, (3) mail fraud, (4) wire fraud, and (5) violations of the Travel Act. The Second Circuit observed that the relevant provisions of the money laundering and material support of terrorism statutes expressly provide for extraterritorial application in certain circumstances, and it concluded that those circumstances are alleged to be present here. 764 F. 3d, at 139–140. The court found that the fraud statutes and the Travel Act do not contain the clear indication needed to overcome the presumption against extraterritoriality. But it held that the complaint alleges domestic violations of those statutes because it “allege[s] conduct in the United States that satisfies every essential element of the mail fraud, wire fraud, and Travel Act claims.” Id., at 142. RJR does not dispute these characterizations of the alleged predicates. We therefore assume without deciding that the alleged pattern of racketeering activity consists entirely of predicate offenses that were either committed in the United States or committed in a foreign country in violation of a predicate statute that applies extraterritorially. The alleged enterprise also has a sufficient tie to U. S. commerce, as its members include U. S. companies, and its activities depend on sales of RJR’s cigarettes conducted through “the U. S. mails and wires,” among other things. App. to Pet. for Cert. 186a, Complaint ¶96. On these premises, respondents’ allegations that RJR violated §§1962(b) and (c) do not involve an impermissibly extraterritorial application of RICO.[8] IV We now turn to RICO’s private right of action, on which respondents’ lawsuit rests. Section 1964(c) allows “[a]ny person injured in his business or property by reason of a violation of section 1962” to sue for treble damages, costs, and attorney’s fees. Irrespective of any extraterritorial application of §1962, we conclude that §1964(c) does not overcome the presumption against extraterritoriality. A private RICO plaintiff therefore must allege and prove a domestic injury to its business or property. A The Second Circuit thought that the presumption against extraterritoriality did not apply to §1964(c) independently of its application to §1962, reasoning that the presumption “is primarily concerned with the question of what conduct falls within a statute’s purview.” 764 F. 3d, at 151. We rejected that view in Kiobel, holding that the presumption “constrain[s] courts considering causes of action” under the ATS, a “ ‘strictly jurisdictional’ ” statute that “does not directly regulate conduct or afford relief.” 569 U. S., at ___ (slip op., at 5). We reached this conclusion even though the underlying substantive law consisted of well-established norms of international law, which by definition apply beyond this country’s borders. See id., at ___–___ (slip op., at 5–7). The same logic requires that we separately apply the presumption against extraterritoriality to RICO’s cause of action despite our conclusion that the presumption has been overcome with respect to RICO’s substantive prohibitions. “The creation of a private right of action raises issues beyond the mere consideration whether underlying primary conduct should be allowed or not, entailing, for example, a decision to permit enforcement without the check imposed by prosecutorial discretion.” Sosa v. Alvarez-Machain, 542 U. S. 692, 727 (2004) . Thus, as we have observed in other contexts, providing a private civil remedy for foreign conduct creates a potential for international friction beyond that presented by merely applying U. S. substantive law to that foreign conduct. See, e.g., Kiobel, supra, at ___ (slip op., at 6) (“Each of th[e] decisions” involved in defining a cause of action based on “conduct within the territory of another sovereign” “carries with it significant foreign policy implications”). Consider antitrust. In that context, we have observed that “[t]he application . . . of American private treble-damages remedies to anticompetitive conduct taking place abroad has generated considerable controversy” in other nations, even when those nations agree with U. S. substantive law on such things as banning price fixing. F. Hoffmann-La Roche Ltd v. Empagran S. A., 542 U. S. 155, 167 (2004). Numerous foreign countries—including some respondents in this case—advised us in Empagran that “to apply [U. S.] remedies would unjustifiably permit their citizens to bypass their own less generous remedial schemes, thereby upsetting a balance of competing considerations that their own domestic antitrust laws embody.” Ibid.[9] We received similar warnings in Morrison, where France, a respondent here, informed us that “most foreign countries proscribe securities fraud” but “have made very different choices with respect to the best way to implement that proscription,” such as “prefer[ring] ‘state actions, not private ones’ for the enforcement of law.” Brief for Republic of France as Amicus Curiae, O. T. 2009, No. 08–1191, p. 20; see id., at 23 (“Even when foreign countries permit private rights of action for securities fraud, they often have different schemes” for litigating them and “may approve of different measures of damages”). Allowing foreign investors to pursue private suits in the United States, we were told, “would upset that delicate balance and offend the sovereign interests of foreign nations.” Id., at 26. Allowing recovery for foreign injuries in a civil RICO action, including treble damages, presents the same danger of international friction. See Brief for United States as Amicus Curiae 31–34. This is not to say that friction would necessarily result in every case, or that Congress would violate international law by permitting such suits. It is to say only that there is a potential for international controversy that militates against recognizing foreign-injury claims without clear direction from Congress. Although “a risk of conflict between the American statute and a foreign law” is not a prerequisite for applying the presumption against extraterritoriality, Morrison, 561 U. S., at 255, where such a risk is evident, the need to enforce the presumption is at its apex. Respondents urge that concerns about international friction are inapplicable in this case because here the plaintiffs are not foreign citizens seeking to bypass their home countries’ less generous remedies but rather the foreign countries themselves. Brief for Respondents 52–53. Respondents assure us that they “are satisfied that the[ir] complaint . . . comports with limitations on prescriptive jurisdiction under international law and respects the dignity of foreign sovereigns.” Ibid. Even assuming that this is true, however, our interpretation of §1964(c)’s injury requirement will necessarily govern suits by nongovernmental plaintiffs that are not so sensitive to foreign sovereigns’ dignity. We reject the notion that we should forgo the presumption against extraterritoriality and instead permit extraterritorial suits based on a case-by-case inquiry that turns on or looks to the consent of the affected sovereign. See Morrison, supra, at 261 (“Rather than guess anew in each case, we apply the presumption in all cases”); cf. Empagran, 542 U. S., at 168. Respondents suggest that we should be reluctant to permit a foreign corporation to be sued in the courts of this country for events occurring abroad if the nation of incorporation objects, but that we should discard those reservations when a foreign state sues a U. S. entity in this country under U. S. law—instead of in its own courts and under its own laws—for conduct committed on its own soil. We refuse to adopt this double standard. “After all, in the law, what is sauce for the goose is normally sauce for the gander.” Heffernan v. City of Paterson, 578 U. S. ___, ___ (2016) (slip op., at 6). B Nothing in §1964(c) provides a clear indication that Congress intended to create a private right of action for injuries suffered outside of the United States. The statute provides a cause of action to “[a]ny person injured in his business or property” by a violation of §1962. §1964(c). The word “any” ordinarily connotes breadth, but it is insufficient to displace the presumption against extraterritoriality. See Kiobel, 569 U. S., at ___ (slip op., at 7). The statute’s reference to injury to “business or property” also does not indicate extraterritorial application. If anything, by cabining RICO’s private cause of action to particular kinds of injury—excluding, for example, personal injuries—Congress signaled that the civil remedy is not coextensive with §1962’s substantive prohibitions. The rest of §1964(c) places a limit on RICO plaintiffs’ ability to rely on securities fraud to make out a claim. This too suggests that §1964(c) is narrower in its application than §1962, and in any event does not support extraterritoriality. The Second Circuit did not identify anything in §1964(c) that shows that the statute reaches foreign injuries. Instead, the court reasoned that §1964(c)’s extraterritorial effect flows directly from that of §1962. Citing our holding in Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479 (1985) , that the “compensable injury” addressed by §1964(c) “necessarily is the harm caused by predicate acts sufficiently related to constitute a pattern,” id., at 497, the Court of Appeals held that a RICO plaintiff may sue for foreign injury that was caused by the violation of a predicate statute that applies extraterritorially, just as a substantive RICO violation may be based on extraterritorial predicates. 764 F. 3d, at 151. Justice Ginsburg advances the same theory. See post, at 4–5 (opinion concurring in part and dissenting in part). This reasoning has surface appeal, but it fails to appreciate that the presumption against extraterritoriality must be applied separately to both RICO’s substantive prohibitions and its private right of action. See supra, at 18–22. It is not enough to say that a private right of action must reach abroad because the underlying law governs conduct in foreign countries. Something more is needed, and here it is absent.[10] Respondents contend that background legal principles allow them to sue for foreign injuries, invoking what they call the “ ‘traditional rule’ that ‘a plaintiff injured in a foreign country’ could bring suit ‘in American courts.’ ” Brief for Respondents 41 (quoting Sosa, 542 U. S., at 706–707). But the rule respondents invoke actually provides that a court will ordinarily “apply foreign law to determine the tortfeasor’s liability” to “a plaintiff injured in a foreign country.” Id., at 706 (emphasis added). Respondents’ argument might have force if they sought to sue RJR for violations of their own laws and to invoke federal diversity jurisdiction as a basis for proceeding in U. S. courts. See U. S. Const., Art. III, §2, cl. 1 (“The judicial Power [of the United States] shall extend . . . to Controversies . . . between a State, or the Citizens thereof, and foreign States”); 28 U. S. C. §1332(a)(4) (“The district courts shall have original jurisdiction of all civil actions where the matter in controversy exceeds the sum or value of $75,000 . . . and is between . . . a foreign state . . . as plaintiff and citizens of a State or of different States”). The question here, however, is not “whether a federal court has jurisdiction to entertain a cause of action provided by foreign or even international law. The question is instead whether the court has authority to recognize a cause of action under U. S. law” for injury suffered overseas. Kiobel, supra, at ___ (slip op., at 8) (emphasis added). As to that question, the relevant background principle is the presumption against extraterritoriality, not the “traditional rule” respondents cite. Respondents and Justice Ginsburg point out that RICO’s private right of action was modeled after §4 of the Clayton Act, 15 U. S. C. §15; see Holmes v. Securities Investor Protection Corporation, 503 U. S. 258 –268 (1992), which we have held allows recovery for injuries suffered abroad as a result of antitrust violations, see Pfizer Inc. v. Government of India, 434 U. S. 308 –315 (1978). It follows, respondents and Justice Ginsburg contend, that §1964(c) likewise allows plaintiffs to sue for injuries suffered in foreign countries. We disagree. Al-though we have often looked to the Clayton Act for guidance in construing §1964(c), we have not treated the two statutes as interchangeable. We have declined to transplant features of the Clayton Act’s cause of action into the RICO context where doing so would be inappropriate. For example, in Sedima we held that a RICO plaintiff need not allege a special “racketeering injury,” rejecting a requirement that some lower courts had adopted by “[a]nalog[y]” to the “antitrust injury” required under the Clayton Act. 473 U. S., at 485, 495. There is good reason not to interpret §1964(c) to cover foreign injuries just because the Clayton Act does so. When we held in Pfizer that the Clayton Act allows recovery for foreign injuries, we relied first and foremost on the fact that the Clayton Act’s definition of “person”—which in turn defines who may sue under that Act—“explicitly includes ‘corporations and associations existing under or authorized by . . . the laws of any foreign country.’ ” 434 U. S., at 313; see 15 U. S. C. §12.[11] RICO lacks the language that the Pfizer Court found critical. See 18 U. S. C. §1961(3).[12] To the extent that the Pfizer Court cited other factors that might apply to §1964(c), they were not sufficient in themselves to show that the provision has extraterritorial effect. For example, the Pfizer Court, writing before we honed our extraterritoriality jurisprudence in Morrison and Kiobel, reasoned that Congress “[c]learly . . . did not intend to make the [Clayton Act’s] treble-damages remedy available only to consumers in our own country” because “the antitrust laws extend to trade ‘with foreign nations’ as well as among the several States of the Union.” 434 U. S., at 313–314. But we have emphatically rejected reliance on such language, holding that “ ‘even statutes . . . that expressly refer to “foreign commerce” do not apply abroad.’ ” Morrison, 561 U. S., at 262–263. This reasoning also fails to distinguish between extending substantive antitrust law to foreign conduct and extending a private right of action to foreign injuries, two separate issues that, as we have explained, raise distinct extraterritoriality problems. See supra, at 18–22. Finally, the Pfizer Court expressed concern that it would “defeat th[e] purposes” of the antitrust laws if a defendant could “escape full liability for his illegal actions.” 434 U. S., at 314. But this justification was merely an attempt to “divin[e] what Congress would have wanted” had it considered the question of extraterritoriality—an approach we eschewed in Morrison. 561 U. S., at 261. Given all this, and in particular the fact that RICO lacks the language that Pfizer found integral to its decision, we decline to extend this aspect of our Clayton Act jurisprudence to RICO’s cause of action. Underscoring our reluctance to read §1964(c) as broadly as we have read the Clayton Act is Congress’s more recent decision to define precisely the antitrust laws’ extraterritorial effect and to exclude from their reach most conduct that “causes only foreign injury.” Empagran, 542 U. S., at 158 (describing Foreign Trade Antitrust Improvements Act of 1982); see also id., at 169–171, 173–174 (discussing how the applicability of the antitrust laws to foreign injuries may depend on whether suit is brought by the Government or by private plaintiffs). Although this later enactment obviously does not limit §1964(c)’s scope by its own force, it does counsel against importing into RICO those Clayton Act principles that are at odds with our current extraterritoriality doctrine. C Section 1964(c) requires a civil RICO plaintiff to allege and prove a domestic injury to business or property and does not allow recovery for foreign injuries. The application of this rule in any given case will not always be self-evident, as disputes may arise as to whether a particular alleged injury is “foreign” or “domestic.” But we need not concern ourselves with that question in this case. As this case was being briefed before this Court, respondents filed a stipulation in the District Court waiving their damages claims for domestic injuries. The District Court accepted this waiver and dismissed those claims with prejudice. Respondents’ remaining RICO damages claims there-fore rest entirely on injury suffered abroad and must be dismissed.[13] * * * The judgment of the United States Court of Appeals for the Second Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. So ordered. Justice Sotomayor took no part in the consideration or decision of this case.Notes 1 In full, 18 U. S. C. §1962 provides: 2 In full, §1964(c) provides: 3 The complaint also alleges that RJR committed a variety of state-law torts. Those claims are not before us. 4 At an earlier stage of respondents’ litigation against RJR, the Second Circuit “held that the revenue rule barred the foreign sovereigns’ civil claims for recovery of lost tax revenue and law enforcement costs.” European Community v. RJR Nabisco, Inc., 424 F. 3d 175, 178 (2005) (Sotomayor, J.), cert. denied, 546 U. S. 1092 (2006) . It is unclear why respondents subsequently included these alleged injuries in their present complaint; they do not ask us to disturb or distinguish the Second Circuit’s holding that such injuries are not cognizable. We express no opinion on the matter. Cf. Pasquantino v. United States, 544 U. S. 349 , n. 1 (2005). 5 Because a finding of extraterritoriality at step one will obviate step two’s “focus” inquiry, it will usually be preferable for courts to proceed in the sequence that we have set forth. But we do not mean to preclude courts from starting at step two in appropriate cases. Cf. Pearson v. Callahan, 555 U. S. 223 –243 (2009). 6 The foreign killings would, of course, still have to satisfy the relatedness and continuity requirements of RICO’s pattern element. See H. J. Inc. v. Northwestern Bell Telephone Co., 492 U. S. 229 (1989) . 7 As to §§1962(a) and (d), see supra, at 13–14. 8 We stress that we are addressing only the extraterritoriality question. We have not been asked to decide, and therefore do not decide, whether the complaint satisfies any other requirements of RICO, or whether the complaint in fact makes out violations of the relevant predicate statutes. 9 See Brief for Governments of Federal Republic of Germany et al. as Amici Curiae, O. T. 2003, No. 03–724, p. 11 (identifying “controversial features of the U. S. legal system,” including treble damages, extensive discovery, jury trials, class actions, contingency fees, and punitive damages); id., at 15 (“Private plaintiffs rarely exercise the type of self-restraint or demonstrate the requisite sensitivity to the concerns of foreign governments that mark actions brought by the United States government”); Brief for United Kingdom et al. as Amici Curiae, O. T. 2003, No. 03–724, p. 13 (“No other country has adopted the United States’ unique ‘bounty hunter’ approach that permits a private plaintiff to ‘recover threefold the damages by him sustained, and the cost of suit, including a reasonable attorney’s fee.’ . . . Expanding the jurisdiction of this generous United States private claim system could skew enforcement and increase international business risks. It makes United States courts the forum of choice without regard to whose laws are applied, where the injuries occurred or even if there is any connection to the court except the ability to get in personam jurisdiction over the defendants”); see also Brief for Government of Canada as Amicus Curiae, O. T. 2003, No. 03–724, p. 14 (“[T]he attractiveness of the [U. S.] treble damages remedy would supersede the national policy decision by Canada that civil recovery by Canadian citizens for injuries resulting from anti-competitive behavior in Canada should be limited to actual damages”). Empagran concerned not the presumption against extraterritoriality per se, but the related rule that we construe statutes to avoid unreasonable interference with other nations’ sovereign authority where possible. See F. Hoffmann-La Roche Ltd v. Empagran S. A., 542 U. S. 155, 164 (2004) ; see also Hartford Fire Ins. Co. v. California, 509 U. S. 764 –815 (1993) (Scalia, J., dissenting) (discussing the two canons). As the foregoing discussion makes clear, considerations relevant to one rule are often relevant to the other. 10 Respondents note that Sedima itself involved an injury suffered by a Belgian corporation in Belgium. Brief for Respondents 45–46; see Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479 –484 (1985). Respondents correctly do not contend that this fact is controlling here, as the Sedima Court did not address the foreign-injury issue. 11 Pfizer most directly concerned whether a foreign government is a “person” that may be a Clayton Act plaintiff. But it is clear that the Court’s decision more broadly concerned recovery for foreign injuries, see 434 U. S., at 315 (expressing concern that “persons doing business both in this country and abroad might be tempted to enter into anticompetitive conspiracies affecting American consumers in the expectation that the illegal profits they could safely extort abroad would offset any liability to plaintiffs at home”), as respondents themselves contend, see Brief for Respondents 44 (“[T]his Court clearly recognized in Pfizer that Section 4 extends to foreign injuries”). The Court also permitted an antitrust plaintiff to sue for foreign injuries in Continental Ore Co. v. Union Carbide & Carbon Corp., 370 U. S. 690 (1962) , but the Court’s discussion in that case focused on the extraterritoriality of the underlying antitrust prohibitions, not the Clayton Act’s private right of action, see id., at 704–705, and so sheds little light on the interpretive question now before us. 12 This does not mean that foreign plaintiffs may not sue under RICO. The point is that RICO does not include the explicit foreign-oriented language that the Pfizer Court found to support foreign-injury suits under the Clayton Act. 13 In respondents’ letter notifying this Court of the waiver of their domestic-injury damages claims, respondents state that “[n]othing in the stipulation will affect respondents’ claims for equitable relief, including claims for equitable relief under state common law that are not at issue in this case before this Court.” Letter from David C. Frederick, Counsel for Respondents, to Scott S. Harris, Clerk of Court (Feb. 29, 2016). Although the letter mentions only state-law claims for equitable relief, Count 5 of respondents’ complaint seeks equitable relief under RICO. App. to Pet. for Cert. 260a–262a, Complaint ¶¶181–188. This Court has never decided whether equitable relief is available to private RICO plaintiffs, the parties have not litigated that question here, and we express no opinion on the issue today. We note, however, that any claim for equitable relief under RICO based on foreign injuries is necessarily foreclosed by our holding that §1964(c)’s cause of action requires a domestic injury to business or property. It is unclear whether respondents intend to seek equitable relief under RICO based on domestic injuries, and it may prove unnecessary to decide whether §1964(c) (or respondents’ stipulation) permits such relief in light of respondents’ state-law claims. We leave it to the lower courts to determine, if necessary, the status and availability of any such claims.

578.US.2015_15-339

Two guards—James Madigan and petitioner Michael Ross—undertook to move respondent Shaidon Blake, a Maryland inmate, to the prison’s segregation unit. During the transfer, Madigan assaulted Blake, punching him several times in the face. Blake reported the incident to a corrections officer, who referred the matter to the Maryland prison system’s Internal Investigative Unit (IIU). The IIU, which has authority under state law to investigate employee misconduct, issued a report condemning Madigan’s actions. Blake subsequently sued both guards under 42 U. S. C. §1983, alleging excessive force and failure to take protective action. A jury found Madigan liable. But Ross raised (as an affirmative defense) the exhaustion requirement of the Prison Litigation Reform Act of 1995 (PLRA), which demands that an inmate exhaust “such administrative remedies as are available” before bringing suit to challenge prison conditions. §1997e(a). Ross argued that Blake had filed suit without first following the prison’s prescribed procedures for obtaining an administrative remedy, while Blake argued that the IIU investigation was a substitute for those procedures. The District Court sided with Ross and dismissed the suit. The Fourth Circuit reversed, holding that “special circumstances” can excuse a failure to comply with administrative procedural requirements—particularly where the inmate reasonably, even though mistakenly, believed he had sufficiently exhausted his remedies. Held: 1. The Fourth Circuit’s unwritten “special circumstances” exception is inconsistent with the text and history of the PLRA. Pp. 3–8. (a) The PLRA speaks in unambiguous terms, providing that “[n]o action shall be brought” absent exhaustion of available administrative remedies. §1997e(a). Aside from one significant qualifier—that administrative remedies must indeed be “available”—the text suggests no limits on an inmate’s obligation to exhaust. That mandatory language means a court may not excuse a failure to exhaust, even to take “special circumstances” into account. When it comes to statutory exhaustion provisions, courts have a role in creating exceptions only if Congress wants them to. So mandatory exhaustion statutes like the PLRA establish mandatory exhaustion regimes, foreclosing judicial discretion. See, e.g., McNeil v. United States, 508 U. S. 106 . Time and again, this Court has rejected every attempt to deviate from the PLRA’s textual mandate. See Booth v. Churner, 532 U. S. 731 ; Porter v. Nussle, 534 U. S. 516 ; Woodford v. Ngo, 548 U. S. 81 . All those precedents rebut the Fourth Circuit’s “special circumstances” excuse for non-exhaustion. Pp. 3–6. (b) The PLRA’s history further underscores the mandatory nature of its exhaustion regime. The PLRA replaced a largely discretionary exhaustion scheme, see Nussle, 534 U. S., at 523, removing the conditions that administrative remedies be “plain, speedy, and effective,” that they satisfy federal minimum standards, and that exhaustion be “appropriate and in the interests of justice.” The Court of Appeals’ exception, if applied broadly, would resurrect that discretionary regime, in which a court could look to all the particulars of a case to decide whether to excuse a failure to exhaust. And if the exception were confined to cases in which a prisoner makes a reasonable mistake about the meaning of a prison’s grievance procedures, it would reintroduce the requirement that the remedial process be “plain.” When Congress amends legislation, courts must “presume it intends [the change] to have real and substantial effect.” Stone v. INS, 514 U. S. 386 . But the Court of Appeals acted as though no amendment had taken place. Pp. 6–8. 2. Blake’s contention that the prison’s grievance process was not in fact available to him warrants further consideration below. Pp. 8–14. (a) Blake’s suit may yet be viable. The PLRA contains its own, textual exception to mandatory exhaustion. Under §1997e(a), an inmate’s obligation to exhaust hinges on the “availab[ility]” of administrative remedies. A prisoner is thus required to exhaust only those grievance procedures that are “capable of use” to obtain “some relief for the action complained of.” Booth, 532 U. S., at 738. As relevant here, there are three kinds of circumstances in which an administrative remedy, although officially on the books, is not capable of use to obtain relief. First, an administrative procedure is unavailable when it operates as a simple dead end—with officers unable or consistently unwilling to provide any relief to aggrieved inmates. Next, an administrative scheme might be so opaque that it becomes, practically speaking, incapable of use—i.e., some mechanism exists to provide relief, but no ordinary prisoner can navigate it. And finally, a grievance process is rendered unavailable when prison administrators thwart inmates from taking advantage of it through machination, misrepresentation, or intimidation. Pp. 8–11. (b) The facts of this case raise questions about whether, given these principles, Blake had an “available” administrative remedy to exhaust. Ross’s exhaustion defense rests on Blake’s failure to seek relief through Maryland’s Administrative Remedy Procedure (ARP) process, which begins with a grievance to the warden. That process is the standard method for addressing inmate complaints in the State’s prisons. But Maryland separately maintains the IIU to look into charges of prison staff misconduct, and the IIU did just that here. Blake urged in the courts below that once the IIU commences such an inquiry, a prisoner cannot obtain relief through the ARP process. And in this Court, the parties have lodged additional materials relating to the interaction between the IIU and the ARP. Both sides’ submissions, although scattershot and in need of further review, lend some support to Blake’s account. Blake’s filings include many administrative dispositions indicating that Maryland wardens routinely dismiss ARP grievances as procedurally improper when parallel IIU investigations are pending. In addition, Blake has submitted briefs of the Maryland attorney general specifically recognizing that administrative practice. And Ross’s own submissions offer some confirmation of Blake’s view: Ross does not identify a single case in which a warden considered the merits of an ARP grievance while an IIU inquiry was underway. On remand, the Fourth Circuit should perform a thorough review of such materials, and then address whether the remedies Blake did not exhaust were “available” under the legal principles set out here. Pp. 11–14. 787 F. 3d 693, vacated and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Alito, and Sotomayor, JJ., joined. Thomas, J., filed an opinion concurring in part and concurring in the judgment. Breyer, J., filed an opinion concurring in part.

The Prison Litigation Reform Act of 1995 (PLRA) mandates that an inmate exhaust “such administrative remedies as are available” before bringing suit to challenge prison conditions. 42 U. S. C. §1997e(a). The court below adopted an unwritten “special circumstances” exception to that provision, permitting some prisoners to pursue litigation even when they have failed to exhaust available administrative remedies. Today, we reject that freewheeling approach to exhaustion as inconsistent with the PLRA. But we also underscore that statute’s built-in exception to the exhaustion requirement: A prisoner need not exhaust remedies if they are not “available.” The briefs and other submissions filed in this case suggest the possibility that the aggrieved inmate lacked an available administrative remedy. That issue remains open for consideration on remand, in light of the principles stated below. I Respondent Shaidon Blake is an inmate in a Maryland prison. On June 21, 2007, two guards—James Madigan and petitioner Michael Ross—undertook to move him from his regular cell to the facility’s segregation unit. According to Blake’s version of the facts, Ross handcuffed him and held him by the arm as they left the cell; Madigan followed close behind. Near the top of a flight of stairs, Madigan shoved Blake in the back. Ross told Madigan he had Blake under control, and the three continued walking. At the bottom of the stairs, Madigan pushed Blake again and then punched him four times in the face, driving his head into the wall. After a brief pause, Madigan hit Blake one last time. Ross kept hold of Blake throughout the assault. And when the blows subsided, Ross helped Madigan pin Blake to the ground until additional officersarrived. Later that day, Blake reported the assault to a senior corrections officer. That officer thought Madigan at fault, and so referred the incident to the Maryland prison system’s Internal Investigative Unit (IIU). Under state law, the IIU has authority to investigate allegations of employee misconduct, including the use of “excessive force.” Codeof Md. Regs., tit. 12, §11.01.05(A)(3) (2006). After conducting a year-long inquiry into the beating, the IIU issued a final report condemning Madigan’s actions, while making no findings with respect to Ross. See App. 191–195. Madigan resigned to avoid being fired. Blake subsequently sued both guards under 42 U. S. C. §1983, alleging that Madigan had used unjustifiable force and that Ross had failed to take protective action. The claim against Madigan went to a jury, which awarded Blake a judgment of $50,000. But unlike Madigan, Ross raised the PLRA’s exhaustion requirement as an affirmative defense, contending that Blake had brought suit without first following the prison’s prescribed procedures for obtaining an administrative remedy. As set out in Maryland’s Inmate Handbook, that process—called, not very fancifully, the Administrative Remedy Procedure (ARP)—begins with a formal grievance to the prison’s warden; it may also involve appeals to the Commissioner of Correction and then the Inmate Grievance Office (IGO). See Maryland Div. of Correction, Inmate Handbook 30–31 (2007). Blake acknowledged that he had not sought a remedy through the ARP—because, he thought, the IIU investigation served as a substitute for that otherwise standard process. The District Court rejected that explanation and dismissed the suit, holding that “the commencement of an internal investigation does not relieve prisoners from the [PLRA’s] exhaustion requirement.” Blake v. Maynard, No. 8:09–cv–2367 (D Md., Nov. 14, 2012), App. to Pet. for Cert. 38, 2012 WL 5568940, *5. The Court of Appeals for the Fourth Circuit reversed in a divided decision. Stating that the PLRA’s “exhaustion requirement is not absolute,” the court adopted an extra-textual exception originally formulated by the Second Circuit. 787 F. 3d 693, 698 (2015). Repeated the Court of Appeals: “[T]here are certain ‘special circumstances’ in which, though administrative remedies may have been available[,] the prisoner’s failure to comply with administrative procedural requirements may nevertheless have been justified.” Ibid. (quoting Giano v. Goord, 380 F. 3d 670, 676 (CA2 2004)). In particular, that was true when a prisoner “reasonably”—even though mistakenly—“believed that he had sufficiently exhausted his remedies.” 787 F. 3d, at 695. And Blake, the court concluded, fit within that exception because he reasonably thought that “the IIU’s investigation removed his complaint from the typical ARP process.” Id., at 700. Judge Agee dissented, stating that the PLRA’s mandatory exhaustion requirement is not “amenable” to “[j]udge-made exceptions.” Id., at 703. This Court granted certiorari. 577 U. S. ___ (2015). II The dispute here concerns whether the PLRA’s exhaustion requirement, §1997e(a), bars Blake’s suit. Statutory text and history alike foreclose the Fourth Circuit’s adoption of a “special circumstances” exception to that mandate. But Blake’s suit may yet be viable. Under the PLRA, a prisoner need exhaust only “available” administrative remedies. And Blake’s contention that the prison’s grievance process was not in fact available to him warrants further consideration below. A Statutory interpretation, as we always say, begins with the text, see, e.g., Hardt v. Reliance Standard Life Ins. Co., 560 U. S. 242, 251 (2010) —but here following that approach at once distances us from the Court of Appeals. As Blake acknowledges, that court made no attempt to ground its analysis in the PLRA’s language. See 787 F. 3d, at 697–698; Brief for Respondent 47–48, n. 20 (labeling the Court of Appeals’ rule an “extra-textual exception to the PLRA’s exhaustion requirement”). And that failure makes a difference, because the statute speaks in unambiguous terms opposite to what the Fourth Circuit said. Section 1997e(a) provides: “No action shall be brought with respect to prison conditions under section 1983 of this title, or any other Federal law, by a prisoner confined in any jail, prison, or other correctional facility until such administrative remedies as are available are exhausted.” As we have often observed, that language is “mandatory”: An inmate “shall” bring “no action” (or said more conversationally, may not bring any action) absent exhaustion of available administrative remedies. Woodford v. Ngo, 548 U. S. 81, 85 (2006) ; accord, Jones v. Bock, 549 U. S. 199, 211 (2007) (“There is no question that exhaustion is mandatory under the PLRA”). As later discussed, that edict contains one significant qualifier: the remedies must indeed be “available” to the prisoner. See infra, at 8–10. But aside from that exception, the PLRA’s text suggests no limits on an inmate’s obligation to exhaust—irrespective of any “special circumstances.” And that mandatory language means a court may not excuse a failure to exhaust, even to take such circumstances into account. See Miller v. French, 530 U. S. 327, 337 (2000) (explaining that “[t]he mandatory ‘shall’ . . . normally creates an obligation impervious to judicial discretion”). No doubt, judge-made exhaustion doctrines, even if flatly stated at first, remain amenable to judge-made exceptions. See McKart v. United States, 395 U. S. 185, 193 (1969) (“The doctrine of exhaustion of administrative remedies . . . is, like most judicial doctrines, subject to numerous exceptions”). But a statutory exhaustion provision stands on a different footing. There, Congress sets the rules—and courts have a role in creating exceptions only if Congress wants them to. For that reason, mandatory exhaustion statutes like the PLRA establish mandatory exhaustion regimes, foreclosing judicial discretion. See, e.g., McNeil v. United States, 508 U. S. 106, 111, 113 (1993) (“We are not free to rewrite the statutory text” when Congress has strictly “bar[red] claimants from bringing suit in federal court until they have exhausted their administrative remedies”). Time and again, this Court has taken such statutes at face value—refusing to add unwritten limits onto their rigorous textual requirements. See, e.g., id., at 111; Shalala v. Illinois Council on Long Term Care, Inc., 529 U. S. 1 –14 (2000); see also 2 R. Pierce, Administrative Law Treatise §15.3, p. 1241 (5th ed. 2010) (collecting cases). We have taken just that approach in construing the PLRA’s exhaustion provision—rejecting every attempt to deviate (as the Fourth Circuit did here) from its textual mandate. In Booth v. Churner, 532 U. S. 731 (2001) , for example, the prisoner argued that exhaustion was not necessary because he wanted a type of relief that the administrative process did not provide. But §1997e(a), we replied, made no distinctions based on the particular “forms of relief sought and offered,” and that legislative judgment must control: We would not read “exceptions into statutory exhaustion requirements where Congress has provided otherwise.” Id., at 741, n. 6. The next year, in Porter v. Nussle, 534 U. S. 516, 520 (2002) , the Court rejected a proposal to carve out excessive-force claims (like Blake’s) from the PLRA’s exhaustion regime, viewing that approach too as inconsistent with the uncompromising statutory text. And most recently, in Woodford, we turned aside a requested exception for constitutional claims. 548 U. S., at 91, n. 2. Our explanation was familiar: “We are interpreting and applying” not a judge-made doctrine but a “statutory requirement,” and therefore must honor Congress’s choice. Ibid.[1] All those precedents rebut the Court of Appeals’ adoption of a “special circumstances” excuse for non-exhaustion. So too, the history of the PLRA underscores the mandatory nature of its exhaustion regime. Section §1997e(a)’s precursor, enacted in the Civil Rights of Institutionalized Persons Act (CRIPA), §7, 94Stat. 352 (1980), was a “weak exhaustion provision.” Woodford, 548 U. S., at 84. Under CRIPA, a court would require exhaustion only if a State provided “plain, speedy, and effective” remedies meeting federal minimum standards—and even then, only if the court believed exhaustion “appropriate and in the interests of justice.” §7(a), 94Stat. 352. That statutory scheme made exhaustion “in large part discretionary.” Nussle, 534 U. S., at 523. And for that reason (among others), CRIPA proved inadequate to stem the then-rising tide of prisoner litigation. In enacting the PLRA, Congress thus substituted an “invigorated” exhaustion provision. Woodford, 548 U. S., at 84. “[D]iffer[ing] markedly from its predecessor,” the new §1997e(a) removed the conditions that administrative remedies be “plain, speedy, and effective” and that they satisfy minimum standards. Nussle, 534 U. S., at 524. Still more, the PLRA prevented a court from deciding that exhaustion would be unjust or inappropriate in a given case. As described earlier, see supra, at 4–5, all inmates must now exhaust all available remedies: “Exhaustion is no longer left to the discretion of the district court.” Woodford, 548 U. S., at 85. The PLRA’s history (just like its text) thus refutes a “special circumstances” exception to its rule of exhaustion. That approach, if applied broadly, would resurrect CRIPA’s scheme, in which a court could look to all the particulars of a case to decide whether to excuse a failure to exhaust available remedies. But as we have observed, such wide-ranging discretion “is now a thing of the past.” Booth, 532 U. S., at 739. And the conflict with the PLRA’s history (as again with its text) becomes scarcely less stark if the Fourth Circuit’s exception is confined, as the court may have intended, to cases in which a prisoner makes a reasonable mistake about the meaning of a prison’s grievance procedures. Understood that way, the exception reintroduces CRIPA’s requirement that the remedial process be “plain”—that is, not subject to any reasonable misunderstanding or disagreement. §7(a), 94Stat. 352. When Congress amends legislation, courts must “presume it intends [the change] to have real and substantial effect.” Stone v. INS, 514 U. S. 386, 397 (1995) . The Court of Appeals instead acted as though the amendment—from a largely permissive to a mandatory exhaustion regime—had not taken place.[2] B Yet our rejection of the Fourth Circuit’s “special circumstances” exception does not end this case—because the PLRA contains its own, textual exception to mandatory exhaustion. Under §1997e(a), the exhaustion requirement hinges on the “availab[ility]” of administrative remedies: An inmate, that is, must exhaust available remedies, but need not exhaust unavailable ones. And that limitation on an inmate’s duty to exhaust—although significantly different from the “special circumstances” test or the old CRIPA standard—has real content. As we explained in Booth, the ordinary meaning of the word “available” is “ ‘capable of use for the accomplishment of a purpose,’ and that which ‘is accessible or may be obtained.’ ” 532 U. S., at 737–738 (quoting Webster’s Third New International Dictionary 150 (1993)); see also Random House Dictionary of the English Language 142 (2d ed. 1987) (“suitable or ready for use”); 1 Oxford English Dictionary 812 (2d ed. 1989) (“capable of being made use of, at one’s disposal, within one’s reach”); Black’s Law Dictionary 135 (6th ed. 1990) (“useable”; “present or ready for immediate use”). Accordingly, an inmate is required to exhaust those, but only those, grievance procedures that are “capable of use” to obtain “some relief for the action complained of.” Booth, 532 U. S., at 738. To state that standard, of course, is just to begin; courts in this and other cases must apply it to the real-world workings of prison grievance systems. Building on our own and lower courts’ decisions, we note as relevant here three kinds of circumstances in which an administrative remedy, although officially on the books, is not capable of use to obtain relief. See Tr. of Oral Arg. 27–29 (Solicitor General as amicus curiae acknowledging these three kinds of unavailability). Given prisons’ own incentives to maintain functioning remedial processes, we expect that these circumstances will not often arise. See Woodford, 548 U. S., at 102. But when one (or more) does, an inmate’s duty to exhaust “available” remedies does not come into play. First, as Booth made clear, an administrative procedure is unavailable when (despite what regulations or guidance materials may promise) it operates as a simple dead end—with officers unable or consistently unwilling to provide any relief to aggrieved inmates. See 532 U. S., at 736, 738. Suppose, for example, that a prison handbook directs inmates to submit their grievances to a particular administrative office—but in practice that office disclaims the capacity to consider those petitions. The procedure is not then “capable of use” for the pertinent purpose. In Booth’s words: “[S]ome redress for a wrong is presupposed by the statute’s requirement” of an “available” remedy; “where the relevant administrative procedure lacks authority to provide any relief,” the inmate has “nothing to exhaust.” Id., at 736, and n. 4. So too if administrative officials have apparent authority, but decline ever to exercise it. Once again: “[T]he modifier ‘available’ requires the possibility of some relief.” Id., at 738. When the facts on the ground demonstrate that no such potential exists, the inmate has no obligation to exhaust the remedy. Next, an administrative scheme might be so opaque that it becomes, practically speaking, incapable of use. In this situation, some mechanism exists to provide relief, but no ordinary prisoner can discern or navigate it. As the Solicitor General put the point: When rules are “so confusing that . . . no reasonable prisoner can use them,” then “they’re no longer available.” Tr. of Oral Arg. 23. That is a significantly higher bar than CRIPA established or the Fourth Circuit suggested: The procedures need not be sufficiently “plain” as to preclude any reasonable mistake or debate with respect to their meaning. See §7(a), 94Stat. 352; 787 F. 3d, at 698–699; supra, at 3, 6–8. When an administrative process is susceptible of multiple reasonable interpretations, Congress has determined that the inmate should err on the side of exhaustion. But when a remedy is, in Judge Carnes’s phrasing, essentially “unknowable”—so that no ordinary prisoner can make sense of what it demands—then it is also unavailable. See Goebert v. Lee County, 510 F. 3d 1312, 1323 (CA11 2007); Turner v. Burnside, 541 F. 3d 1077, 1084 (CA11 2008) (“Remedies that rational inmates cannot be expected to use are not capable of accomplishing their purposes andso are not available”). Accordingly, exhaustion is not required. And finally, the same is true when prison administrators thwart inmates from taking advantage of a grievance process through machination, misrepresentation, or intimidation. In Woodford, we recognized that officials might devise procedural systems (including the blind alleys and quagmires just discussed) in order to “trip[] up all but the most skillful prisoners.” 548 U. S., at 102. And appellate courts have addressed a variety of instances in which officials misled or threatened individual inmates so as to prevent their use of otherwise proper procedures. As all those courts have recognized, such interference with an inmate’s pursuit of relief renders the administrative process unavailable.[3] And then, once again, §1997e(a) poses no bar. The facts of this case raise questions about whether, given these principles, Blake had an “available” administrative remedy to exhaust. As explained earlier, Ross’s exhaustion defense rests on Blake’s failure to seek relief through Maryland’s ARP process, which begins with a grievance to the warden and may continue with appeals to the Commissioner of Correction and the IGO. See supra, at 2–3; Inmate Handbook, at 30–31. That process is the standard method for addressing inmate complaints in the State’s prisons: The Inmate Handbook provides that prisoners may use the ARP for “all types” of grievances (subject to four exceptions not relevant here), including those relating to the use of force. Id., at 30; see App. 312. But recall that Maryland separately maintains the IIU to look into charges of staff misconduct in prisons, and the IIU did just that here. See supra, at 2. Blake urged in the courts below that once the IIU commences such an inquiry, a prisoner cannot obtain relief through the standard ARP process—whatever the Handbook may say to the contrary. See 787 F. 3d, at 697; App. to Pet. for Cert. 38, 2012 WL 5568940, at *5. And in this Court, that issue has taken on new life. Both Blake and Ross (as represented by the Maryland attorney general) have lodged additional materials relating to the interaction between the IIU and the ARP. And both sides’ submissions, although scattershot and in need of further review, lend some support to Blake’s account—while also revealing Maryland’s grievance process to have, at least at first blush, some bewildering features. Blake’s filings include many administrative dispositions (gleaned from the records of other prisoner suits) indicating that Maryland wardens routinely dismiss ARP grievances as procedurally improper when parallel IIU investigations are pending. One warden, for example, wrote in response to a prisoner’s complaint: “Your Request for Administrative Remedy has been received and is hereby dismissed. This issue has been assigned to the Division of Correction’s Internal Investigative Unit (Case #07–35–010621I/C), and will no longer be addressed through this process.” Lodging of Respondent 1; see also, e.g., id., at 18 (“Admin. Dismiss Final: This is being investigated outside of the ARP process by I.I.U.”). In addition, Blake has submitted briefs of the Maryland attorney general (again, drawn from former prisoner suits) specifically recognizing that administrative practice. As the attorney general stated in one case: “Wilkerson filed an ARP request,” but “his complaint already was being investigated by the [IIU], superceding an ARP investigation.” Id., at 23–24; see also, e.g., id., at 5 (Bacon’s grievance “was dismissed because the issue had been assigned to [the] IIU and would no longer be addressed through the ARP process”).[4] And Ross’s own submissions offer some confirmation of Blake’s view. Ross does not identify a single case in which a warden considered the merits of an ARP grievance while an IIU inquiry was underway. See Tr. of Oral Arg. 6 (Maryland attorney general’s office conceding that it had found none). To the contrary, his lodging contains still further evidence that wardens consistently dismiss such complaints as misdirected. See, e.g., Lodging of Petitioner 15 (District Court noting that “Gladhill was advised that no further action would be taken through the ARP process because the matter had been referred to the [IIU]”). Indeed, Ross’ materials suggest that some wardens use a rubber stamp specially devised for that purpose; the inmate, that is, receives a reply stamped with the legend: “Dismissed for procedural reasons . . . . This issue is being investigated by IIU case number: ____. No further action shall be taken within the ARP process.” Id., at 25, 32, 38; see Tr. of Oral Arg. 8–9 (Maryland attorney general’s office conceding the stamp’s existence and use). Complicating the picture, however, are several cases in which an inmate refused to take a warden’s jurisdictional “no” for an answer, resubmitted his grievance up the chain to the IGO, and there received a ruling on the merits, without any discussion of the ARP/IIU issue. We confess to finding these few cases perplexing in relation to normal appellate procedure. See id., at 3–10, 13–15, 18–20 (multiple Justices expressing confusion about Maryland’s procedures). If the IGO thinks the wardens wrong to dismiss complaints because of pending IIU investigations, why does it not say so and stop the practice? Conversely, if the IGO thinks the wardens right, how can it then issue merits decisions? And if that really is Maryland’s procedure—that when an IIU investigation is underway, the warden (and Commissioner of Correction) cannot consider a prisoner’s complaint, but the IGO can—why does the Inmate Handbook not spell this out? Are there, instead, other materials provided to prisoners that communicate how this seemingly unusual process works and how to navigate it so as to get a claim heard? In light of all these lodgings and the questions they raise about Maryland’s grievance process, we remand this case for further consideration of whether Blake had “available” remedies to exhaust. The materials we have seen are not conclusive; they may not represent the complete universe of relevant documents, and few have been analyzed in the courts below. On remand, in addition to considering any other arguments still alive in this case, the court must perform a thorough review of such materials, and then address the legal issues we have highlighted concerning the availability of administrative remedies. First, did Maryland’s standard grievance procedures potentially offer relief to Blake or, alternatively, did the IIU investigation into his assault foreclose that possibility? Second, even if the former, were those procedures knowable by an ordinary prisoner in Blake’s situation, or was the system so confusing that no such inmate could make use of it? And finally, is there persuasive evidence that Maryland officials thwarted the effective invocation of the administrative process through threats, game-playing, or misrepresentations, either on a system-wide basis or in the individual case? If the court accepts Blake’s probable arguments on one or more of these scores, then itshould find (consistent this time with the PLRA) that his suit may proceed even though he did not file an ARP complaint. III Courts may not engraft an unwritten “special circumstances” exception onto the PLRA’s exhaustion requirement. The only limit to §1997e(a)’s mandate is the one baked into its text: An inmate need exhaust only such administrative remedies as are “available.” On remand, the court below must consider how that modifying term affects Blake’s case—that is, whether the remedies he failed to exhaust were “available” under the principles set out here. We therefore vacate the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 We note that our adherence to the PLRA’s text runs both ways: The same principle applies regardless of whether it benefits the inmate or the prison. We have thus overturned judicial rulings that imposed extra-statutory limitations on a prisoner’s capacity to sue—reversing, for example, decisions that required an inmate to demonstrate exhaustion in his complaint, permitted suit against only defendants named in the administrative grievance, and dismissed an entire action because of a single unexhausted claim. See Jones v. Bock, 549 U. S. 199, 203 (2007) . “[T]hese rules,” we explained, “are not required by the PLRA,” and “crafting and imposing them exceeds the proper limits on the judicial role.” Ibid. 2 Of course, an exhaustion provision with a different text and history from §1997e(a) might be best read to give judges the leeway to create exceptions or to itself incorporate standard administrative-law exceptions. See 2 R. Pierce, Administrative Law Treatise §15.3, p. 1245 (5th ed. 2010). The question in all cases is one of statutory construction, which must be resolved using ordinary interpretive techniques. 3 See, e.g., Davis v. Hernandez, 798 F. 3d 290, 295 (CA5 2015) (“Grievance procedures are unavailable . . . if the correctional facility’s staff misled the inmate as to the existence or rules of the grievance process so as to cause the inmate to fail to exhaust such process” (emphasis deleted)); Schultz v. Pugh, 728 F. 3d 619, 620 (CA7 2013) (“A remedy is not available, therefore, to a prisoner prevented by threats or other intimidation by prison personnel from seeking an administrative remedy”); Pavey v. Conley, 663 F. 3d 899, 906 (CA7 2011) (“[I]f prison officials misled [a prisoner] into thinking that . . . he had done all he needed to initiate the grievance process,” then “[a]n administrative remedy is not ‘available’ ”); Tuckel v. Grover, 660 F. 3d 1249, 1252–1253 (CA10 2011) (“[W]hen a prison official inhibits an inmate from utilizing an administrative process through threats or intimidation, that process can no longer be said to be ‘available’ ”); Goebert v. Lee County, 510 F. 3d 1312, 1323 (CA11 2007) (If a prison “play[s] hide-and-seek with administrative remedies,” then they are not “available”). 4 Blake further notes that in 2008, a year after his beating, Maryland amended one of its prison directives to state expressly that when the IIU investigates an incident, an ARP grievance may not proceed. See App. 367, Md. Div. of Correction, Directive 185–003, §VI(N)(4) (Aug. 27, 2008) (The Warden “shall issue a final dismissal of [an ARP] request for procedural reasons when it has been determined that the basis of the complaint is the same basis of an investigation under the authority of the [IIU]”); Brief for Respondent 17–18. According to Blake, that amendment merely codified what his submissions show had long been the practice in Maryland prisons. See ibid.

577.US.2015_14-990

Since 1976, federal law has mandated that a “district court of three judges shall be convened . . . when an action is filed challenging the constitutionality of the apportionment of congressional districts . . . ,” 28 U. S. C. §2284(a), and has provided that “the judge [presented with a request for a three-judge court] shall, unless he determines that three judges are not required, immediately notify the chief judge of the circuit, who shall designate two other judges” to serve, §2284(b)(1). Petitioners requested that a three-judge court be convened to consider their claim that Maryland’s 2011 congressional redistricting plan burdens their First Amendment right of political association. Concluding that no relief could be granted for this claim, the District Judge dismissed the action instead of notifying the Chief Judge of the Circuit to convene a three-judge court. The Fourth Circuit affirmed. Held: Section 2284 entitles petitioners to make their case before a three-judge court. Pp. 3–8. (a) Section 2284(a)’s prescription could not be clearer. Because the present suit is indisputably “an action . . . challenging the constitutionality of the apportionment of congressional districts,” the District Judge was required to refer the case to a three-judge court. Section 2284(a) admits of no exception, and “the mandatory ‘shall’ . . . normally creates an obligation impervious to judicial discretion.” Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U. S. 26 . The subsequent provision of §2284(b)(1), that the district judge shall commence the process for appointment of a three-judge panel “unless he determines that three judges are not required,” should be read not as a grant of discretion to the district judge to ignore §2284(a), but as a compatible administrative detail requiring district judges to “determin[e]” only whether the “request for three judges” is made in a case covered by §2284(a). This conclusion is bolstered by §2284(b)(3)’s explicit command that “[a] single judge shall not . . . enter judgment on the merits.” Pp. 3–5. (b) Respondents’ alternative argument, that the District Judge should have dismissed petitioners’ claim as “constitutionally insubstantial” under Goosby v. Osser, 409 U. S. 512 , is unpersuasive. This Court has long distinguished between failing to raise a substantial federal question for jurisdictional purposes—what Goosby addressed—and failing to state a claim for relief on the merits—what the District Judge found here; only “wholly insubstantial and frivolous” claims implicate the former, Bell v. Hood, 327 U. S. 678 –683. Absent such obvious frivolity, “the failure to state a proper cause of action calls for a judgment on the merits and not for a dismissal for want of jurisdiction.” Id., at 682. Petitioners’ plea for relief, which was based on a legal theory put forward in Justice Kennedy’s concurrence in Vieth v. Jubelirer, 541 U. S. 267 , and uncontradicted in subsequent majority opinions, easily clears Goosby’s low bar. Pp. 5–7. 584 Fed. Appx. 140, reversed and remanded. Scalia, J., delivered the opinion for a unanimous Court.

We consider under what circumstances, if any, a district judge is free to “determin[e] that three judges are not required” for an action “challenging the constitutionality of the apportionment of congressional districts.” 28 U. S. C. §§2284(a), (b)(1). I A Rare today, three-judge district courts were more common in the decades before 1976, when they were required for various adjudications, including the grant of an “interlocutory or permanent injunction restraining the enforcement, operation or execution of any State statute . . . upon the ground of the unconstitutionality of such statute.” 28 U. S. C. §2281 (1970 ed.), repealed, Pub. L. 94–381, §1, 90Stat. 1119. See Currie, The Three-Judge District Courtin Constitutional Litigation, 32 U. Chi. L. Rev. 1, 3–12 (1964). Decisions of three-judge courts could, then as now, be appealed as of right directly to this Court. 28 U. S. C. §1253. In 1976, Congress substantially curtailed the circumstances under which a three-judge court is required. It was no longer required for the grant of an injunction against state statutes, see Pub. L. 94–381, §1, 90Stat. 1119 (repealing 28 U. S. C. §2281), but was mandated for “an action . . . challenging the constitutionality of the apportionment of congressional districts or the apportionment of any statewide legislative body.” Id., §3, now codified at 28 U. S. C. §2284(a). Simultaneously, Congress amended the procedures governing three-judge district courts. The prior statute had provided: “The district judge to whom the application for injunction or other relief is presented shall constitute one member of [the three-judge] court. On the filing of the application, he shall immediately notify the chief judge of the circuit, who shall designate two other judges” to serve. 28 U. S. C. §2284(1) (1970 ed.). The amended statute provides: “Upon the filing of a request for three judges, the judge to whom the request is presented shall, unless he determines that three judges are not required, immediately notify the chief judge of the circuit, who shall designate two other judges” to serve. 28 U. S. C. §2284(b)(1) (2012 ed.) (emphasis added). The dispute here concerns the scope of the italicized text. B In response to the 2010 Census, Maryland enacted a statute in October 2011 establishing—or, more pejora-tively, gerrymandering—the districts for the State’s eight congressional seats. Dissatisfied with the crazy-quilt results, see App. to Pet. for Cert. 23a, petitioners, a bipartisan group of citizens, filed suit pro se in Federal District Court. Their amended complaint alleges, inter alia, that Maryland’s redistricting plan burdens their First Amendment right of political association. Petitioners also requested that a three-judge court be convened to hear the case. The District Judge, however, thought the claim “not one for which relief can be granted.” Benisek v. Mack, 11 F. Supp. 3d 516, 526 (Md. 2014). “[N]othing about the congressional districts at issue in this case affects in any proscribed way [petitioners’] ability to participate in the political debate in any of the Maryland congressional districts in which they might find themselves. They are free to join preexisting political committees, form new ones, or use whatever other means are at their disposal to influence the opinions of their congressional representatives.” Ibid. (brackets, ellipsis, and internal quotation marks omitted). For that reason, instead of notifying the Chief Judge of the Circuit of the need for a three-judge court, the District Judge dismissed the action. The Fourth Circuit summar-ily affirmed in an unpublished disposition. Benisek v. Mack, 584 Fed. Appx. 140 (CA4 2014). Seeking review in this Court, petitioners pointed out that at least two other Circuits consider it reversible error for a district judge to dismiss a case under §2284 for failure to state a claim for relief rather than refer it for transfer to a three-judge court. See LaRouche v. Fowler, 152 F. 3d 974, 981–983 (CADC 1998); LULAC v. Texas, 113 F. 3d 53, 55–56 (CA5 1997) (per curiam). We granted certiorari. Shapiro v. Mack, 576 U. S. ___ (2015). II Petitioners’ sole contention is that the District Judge had no authority to dismiss the case rather than initiate the procedures to convene a three-judge court. Not so, argue respondents; the 1976 addition to §2284(b)(1) of the clause “unless he determines that three judges are not required” is precisely such a grant of authority. Moreover, say respondents, Congress declined to specify a standard to constrain the exercise of this authority. Choosing, as the District Judge did, the familiar standard for dismissal under Federal Rule of Civil Procedure 12(b)(6) best serves the purposes of a three-judge court, which (in respondents’ view) is to protect States from “hasty, imprudent invalidation” of their statutes by rogue district judges acting alone. Brief for Respondents 27. Whatever the purposes of a three-judge court may be, respondents’ argument needlessly produces a contradiction in the statutory text. That text’s initial prescription could not be clearer: “A district court of three judges shall be convened . . . when an action is filed challenging the constitutionality of the apportionment of congressional districts . . . .” 28 U. S. C. §2284(a) (emphasis added). Nobody disputes that the present suit is “an action . . . challenging the constitutionality of the apportionment of congressional districts.” It follows that the district judge was required to refer the case to a three-judge court, for §2284(a) admits of no exception, and “the mandatory ‘shall’ . . . normally creates an obligation impervious to judicial discretion.” Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U. S. 26, 35 (1998) ; see also National Assn. of Home Builders v. Defenders of Wildlife, 551 U. S. 644 –662 (2007) (same). The subsequent provision of §2284(b)(1), that the district judge shall commence the process for appointment of a three-judge panel “unless he determines that three judges are not required,” need not and therefore should not be read as a grant of discretion to the district judge to ignore §2284(a). It is not even framed as a proviso, or an exception from that provision, but rather as an administrative detail that is entirely compatible with §2284(a). The old §2284(1) triggered the district judge’s duty to refer the matter for the convening of a three-judge court “[o]n the filing of the application” to enjoin an unconstitutional state law. By contrast, the current §2284(b)(1) triggers the district judge’s duty “[u]pon the filing of a request for three judges” (emphasis added). But of course a party may—whether in good faith or bad, through ignorance or hope or malice—file a request for a three-judge court even if the case does not merit one under §2284(a). Section 2284(b)(1) merely clarifies that a district judge need not unthinkingly initiate the procedures to convene a three-judge court without first examining the allegations in the complaint. In short, all the district judge must “determin[e]” is whether the “request for three judges” is made in a case covered by §2284(a)—no more, no less. That conclusion is bolstered by §2284(b)(3)’s explicit command that “[a] single judge shall not . . . enter judgment on the merits.” It would be an odd interpretation that allowed a district judge to do under §2284(b)(1) what he is forbidden to do under §2284(b)(3). More likely that Congress intended a three-judge court, and not a single district judge, to enter all final judgments in cases satisfying the criteria of §2284(a). III Respondents argue in the alternative that a district judge is not required to refer a case for the convening of a three-judge court if the constitutional claim is (as they assert petitioners’ claim to be) “insubstantial.” In Goosby v. Osser, 409 U. S. 512 (1973) , we stated that the filing of a “constitutionally insubstantial” claim did not trigger the three-judge-court requirement under the pre-1976 statu-tory regime. Id., at 518. Goosby rested not on an interpretation of statutory text, but on the familiar proposition that “[i]n the absence of diversity of citizenship, it is essential to jurisdiction that a substantial federal question should be presented.” Ex parte Poresky, 290 U. S. 30, 31 (1933) (per curiam) (emphasis added). Absent a substantial federal question, even a single-judge district court lacks jurisdiction, and “[a] three-judge court is not required where the district court itself lacks jurisdiction of the complaint or the complaint is not justiciable in the federal courts.” Gonzalez v. Automatic Employees Credit Union, 419 U. S. 90, 100 (1974) . In the present case, however, the District Judge dismissed petitioners’ complaint not because he thought he lacked jurisdiction, but because he concluded that the allegations failed to state a claim for relief on the merits, citing Ashcroft v. Iqbal, 556 U. S. 662 (2009) , and Bell Atlantic Corp. v. Twombly, 550 U. S. 544 (2007) . See 11 F. Supp. 3d, at 520. That was in accord with Fourth Circuit precedent, which holds that where the “pleadings do not state a claim, then by definition they are insubstantial and so properly are subject to dismissal by the district court without convening a three-judge court.” Duckworth v. State Admin. Bd. of Election Laws, 332 F. 3d 769, 772–773 (CA4 2003) (emphasis added). We think this standard both too demanding and inconsistent with our precedents. “[C]onstitutional claims will not lightly be found insubstantial for purposes of” the three-judge-court statute. Washington v. Confederated Tribes of Colville Reservation, 447 U. S. 134 –148 (1980). We have long distinguished between failing to raise a substantial federal question for jurisdictional purposes—which is what Goosby addressed—and failing to state a claim for relief on the merits; only “wholly insubstantial and frivolous” claims implicate the former. Bell v. Hood, 327 U. S. 678 –683 (1946); see also Hannis Distilling Co. v. Mayor and City Council of Baltimore, 216 U. S. 285, 288 (1910) (“obviously frivolous or plainly insubstantial”); Bailey v. Patterson, 369 U. S. 31, 33 (1962) (per curiam) (“wholly insubstantial,” “legally speaking non-existent,” “essentially fictitious”); Steel Co. v. Citizens for Better Environment, 523 U. S. 83, 89 (1998) (“frivolous or immaterial”). Absent such frivolity, “the failure to state a proper cause of action calls for a judgment on the merits and not for a dismissal for want of jurisdiction.” Bell, supra, at 682. Consistent with this principle, Goosby clarified that “ ‘[c]onstitutional insubstantiality’ for this purpose has been equated with such concepts as ‘essen-tially fictitious,’ ‘wholly insubstantial,’ ‘obviously frivolous,’ and ‘obviously without merit.’ ” 409 U. S., at 518 (citations omitted). And the adverbs were no mere throwaways; “[t]he limiting words ‘wholly’ and ‘obviously’ have cogent legal significance.” Ibid. Without expressing any view on the merits of petitioners’ claim, we believe it easily clears Goosby’s low bar; after all, the amended complaint specifically challenges Maryland’s apportionment “along the lines suggested by Justice Kennedy in his concurrence in Vieth [v. Jubelirer, 541 U. S. 267 (2004) ].” App. to Brief in Opposition 44. Although the Vieth plurality thought all political gerrymandering claims nonjusticiable, Justice Kennedy, concurring in the judgment, surmised that if “a State did impose burdens and restrictions on groups or persons by reason of their views, there would likely be a First Amendment violation, unless the State shows some compelling interest. . . . Where it is alleged that a gerrymander had the purpose and effect of imposing burdens on a disfavored party and its voters, the First Amendment may offer a sounder and more prudential basis for intervention than does the Equal Protection Clause.” Vieth v. Jubelirer, 541 U. S. 267, 315 (2004) . Whatever “wholly insubstantial,” “obviously frivolous,” etc., mean, at a minimum they cannot include a plea for relief based on a legal theory put forward by a Justice of this Court and uncontradicted by the majority in any of our cases. Accordingly, the District Judge should not have dismissed the claim as “constitutionally insubstantial” under Goosby. Perhaps petitioners will ultimately fail on the merits of their suit, but §2284 entitles them to make their case before a three-judge district court. * * * The judgment of the Fourth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

578.US.2015_15-338

The Fair Debt Collection Practices Act (FDCPA or Act) aims to eliminate “abusive debt collection practices,” 15 U. S. C. §1692(a)–(d), by, as relevant here, barring “false, deceptive, or misleading representation[s] . . . in connection with the collection of any debt,” §1692e. Governing “debt collectors,” the Act excludes from the definition of that term “any officer . . . of . . . any State to the extent that collecting . . . any debt is in the performance of his official duties.” §1692a(6)(C). Under Ohio law, overdue debts owed to state-owned agencies and instrumentalities are certified to the State’s Attorney General for collection or disposition. Carrying out this responsibility, the Attorney General appoints, as independent contractors, private attorneys, naming them “special counsel” to act on the Attorney General’s behalf. The Attorney General requires special counsel to use the Attorney General’s letterhead in communicating with debtors. Among the special counsel appointed by the Attorney General in 2012 were petitioners Mark Sheriff and Eric Jones. Consistent with the Attorney General’s direction, Sheriff’s law firm and Jones sent debt collection letters on the Attorney General’s letterhead to respondents Hazel Meadows and Pamela Gillie, respectively. The signature block of each letter contained the name and address of the signatory as well as the designation “special” or “outside” counsel to the State Attorney General. Each letter also identified the sender as a debt collector seeking payment for debts to a state institution. Meadows and Gillie filed a putative class action in Federal District Court, alleging that defendants had, by using the Attorney General’s letterhead, employed deceptive and misleading means to attempt to collect consumer debts, in violation of the FDCPA. The Ohio Attorney General intervened, seeking a declaratory judgment that special counsel’s use of the Attorney General’s letterhead is neither false nor misleading, and urging that special counsel be deemed officers of the State exempted from the Act. The District Court granted summary judgment for defendants, holding that special counsel are “officers” of the State and, in any event, their use of the Attorney General’s letterhead is not false or misleading. The Sixth Circuit vacated that judgment, concluding that special counsel, as independent contractors, are not entitled to the FDCPA’s state-officer exemption. The appeals court remanded for trial the question whether use of the Attorney General’s letterhead would mislead a debtor into believing that it is the Attorney General who is collecting the debt. Held: Assuming, arguendo, that special counsel do not rank as “state officers” within the meaning of the Act, petitioners’ use of the Attorney General’s letterhead, nevertheless, does not offend §1692e. Special counsel’s use of the Attorney General’s letterhead at the Attorney General’s direction does not offend §1692e’s general prohibition against “false . . . or misleading representation[s].” The letterhead identifies the principal—Ohio’s Attorney General—and the signature block names the agent—a private lawyer hired as outside counsel to the Attorney General. The character of the relationship between special counsel and the Attorney General bolsters the Court’s determination. Special counsel work closely with attorneys in the Attorney General’s Office, providing legal services on the Attorney General’s behalf in furtherance of the Attorney General’s debt collection responsibilities for the State. A debtor’s impression that a letter from special counsel is a letter from the Attorney General’s Office is thus scarcely inaccurate. Special counsel’s use of the Attorney General’s letterhead is also consistent with §1692e(9)’s specific prohibition against “falsely represent[ing]” that a communication is “authorized, issued, or approved” by a State. Because the Attorney General authorized—indeed required—special counsel to use his letterhead, special counsel create no false impression in doing just that. Nor did special counsel use an untrue name in their letters, in violation of §1692e(14). Special counsel do not employ a false name when they use the Attorney General’s letterhead at his instruction, for special counsel act as the Attorney General’s agents in debt-related matters. The Court sees no reason, furthermore, to construe the FDCPA in a manner that would interfere with the Attorney General’s chosen method of fulfilling his statutory obligation to collect the State’s debts. The Sixth Circuit raises the specter of consumer confusion and the risk of intimidation from special counsel’s use of the Attorney General’s letterhead, but its exposition is unconvincing. Pp. 6–11. 785 F. 3d 1091, reversed and remanded. Ginsburg, J., delivered the opinion for a unanimous Court.

Ohio law authorizes the State’s Attorney General to retain, as independent contractors, “special counsel” to act on the Attorney General’s behalf in collecting certain debts owed to Ohio or an instrumentality of the State. Ohio Rev. Code Ann. §109.08 (Lexis 2014). As required by the Attorney General, special counsel use the Attorney General’s letterhead in communicating with debtors. App. 93. The Fair Debt Collection Practices Act, 91Stat. 874, 15 U. S. C. §1692 et seq. (FDCPA or Act), aims to eliminate “abusive debt collection practices.” §1692(a)–(d). To that end, the Act imposes various procedural and substantive obligations on debt collectors. See, e.g., §1692d (prohibiting harassing, oppressive, or abusive conduct); §1692e (barring “false, deceptive, or misleading representation[s] . . . in connection with the collection of any debt”); §1692g(a) (setting out requirements for the contents of initial notices to consumers). The FDCPA excludes from the definition of “debt collector” “any officer or employee of the United States or any State to the extent that collecting . . . any debt is in the performance of his official duties.” §1692a(6)(C). This case involves litigation between debtors to Ohio institutions and special counsel who sought to collect money owed to the institutions. The petition raises two questions: (1) Do special counsel appointed by Ohio’s Attorney General qualify as “state officers” exempt from the FDCPA’s governance? (2) Is special counsel’s use of the Attorney General’s letterhead a false or misleading representation proscribed by §1692e? Assuming, arguendo, that special counsel do not rank as “state officers,” we hold, nevertheless, that their use of the Attorney General’s letterhead does not offend §1692e. Not fairly described as “false” or “misleading,” use of the letterhead accurately conveys that special counsel, in seeking to collect debts owed to the State, do so on behalf of, and as instructed by, the Attorney General. I Responding to reports of abusive practices by third-party collectors of consumer debts, Congress enacted the FDCPA “to eliminate abusive debt collection practices by debt collectors, to insure that those debt collectors who refrain from using abusive debt collection practices are not competitively disadvantaged, and to promote consistent State action to protect consumers against debt collection abuses.” §1692(e). Primarily governing “debt collector[s],” the Act defines that term to include “any person . . . in any business the principal purpose of which is the collection of any debts, or who regularly collects or attempts to collect . . . debts owed or due or asserted to be owed or due another.” §1692a(6). Excluded from the definition is “any offi-cer or employee of the United States or any State to the extent that collecting or attempting to collect any debt is in the performance of his official duties.” §1692a(6)(C). Among other proscriptions, the FDCPA prohibits debt collectors from employing “false, deceptive, or misleading” practices. §1692e. “Without limiting” this general ban, §1692e enumerates 16 categories of conduct that qualify as false or misleading. Two of those categories are pertinent to our review: “[t]he use or distribution of any written communication which simulates or is falsely represented to be a document authorized, issued, or approved by any court, official, or agency of . . . any State, or which creates a false impression as to its source, authorization, or approval,” §1692e(9); and “[t]he use of any business, com-pany, or organization name other than the true name of the debt collector’s business, company, or organization,” §1692e(14). A debt collector who violates the Act is liable for both actual and statutory damages. §1692k(a). This case concerns the debt collection practices of those charged with collecting overdue debts owed to Ohio-owned agencies and instrumentalities. Among such debts are past-due tuition owed to public universities and unpaid medical bills from state-run hospitals. Under Ohio law, overdue debts are certified to the State’s Attorney General, who is responsible for collecting, settling, or otherwise disposing of them. Ohio Rev. Code Ann. §131.02(A), (C), (F). Carrying out this responsibility, the Attorney General may appoint private attorneys as “special counsel to represent the state” in collecting certified claims. §109.08. Special counsel enter into year-long retention agreements “on an independent contractor basis” to “provide legal services on behalf of the Attorney General to one or more State Clients.” App. 143–144. The Attorney General’s Office assigns individual claims to special counsel, who are paid a set percentage of the funds they collect for the State. §109.08; id., at 144–145, 149–152. With “the prior approval of the Attorney General,” special counsel may litigate and settle claims on behalf of the State. Id., at 149. Special counsel may continue to represent private clients so long as doing so does not create a conflict of interest with their work for the Attorney General. Among the special counsel appointed by the Attorney General in 2012 were Mark Sheriff, a partner at the law firm of Wiles, Boyle, Burkholder, and Bringardner Co. LPA (Wiles firm), and Eric Jones, of the Law Offices of Eric A. Jones, LLC. When special counsel contact debtors on behalf of the State, the Attorney General requires them to use his letterhead. Id., at 93. Consistent with this requirement, Sarah Sheriff, an employee of the Wiles firm, sent respondent Hazel Meadows a debt collection letter on the Ohio Attorney General’s letterhead. The letter reads: “Sir/Madam: Per your request, this is a letter with the current balance owed for your University of Akron loan that has been placed with the Ohio Attorney General. Feel free to contact me at [telephone number] should you have any further questions.” Gillie v. Law Office of Eric A. Jones, LLC, 785 F. 3d 1091, 1119 (CA6 2015). The amount Meadows owed is listed in the letter’s subject line. Ibid. After the body of the letter, Sheriff’s signature appears, followed by the firm’s name, its address, and the designation “Special Counsel to the Attorney General for the State of Ohio.” Ibid.[1] The letter concludes with a notice that it is “an attempt to collect a debt” and that the senders “are debt collectors.” Ibid. Respondent Pamela Gillie received a letter, also on the Ohio Attorney General’s letterhead, in relation to a debt she owed to a state-run hospital: “Dear Sir/Madam, You have chosen to ignore repeated attempts to resolv[e] the referenced . . . medical claim. If you cannot make immediate full payment call DENISE HALL at Eric A. Jones, L.L.C., [phone number] at my office to make arrangements to pay this debt.” Id., at 1118. That text is followed by a bolded, all-caps notice that the letter is “a communication from a debt collector.” Ibid. Signed by Eric A. Jones, “Outside Counsel for the Attorney General’s Office,” the letter includes Jones’s telephone and fax numbers. Ibid. A tear-away portion at the bottom of the page for return of payment is addressed to Jones’s law office. Ibid. After receiving these letters, Meadows and Gillie filed a putative class action in the United States District Court for the Southern District of Ohio, asserting that Mark Sheriff, Sarah Sheriff, Jones, and their law firms had violated the FDCPA. By sending debt collection notices on the Attorney General’s letterhead rather than the letterhead of their private firms, Meadows and Gillie alleged, defendants had employed deceptive and misleading means to attempt to collect consumer debts. The Ohio Attorney General intervened as a defendant and counterclaimant, seeking a declaratory judgment that special counsel’s use of his letterhead, as authorized by Ohio law,[2] is neither false nor misleading. Further, the Attorney General urged, special counsel should be deemed officers of the State and therefore outside the FDCPA’s compass. The District Court granted summary judgment for defendants, concluding that special counsel are “officers” of the State of Ohio and, in any event, their use of the Attorney General’s letterhead is not false or misleading. Gillie v. Law Office of Eric A. Jones, LLC, 37 F. Supp. 3d 928 (2014). The Court of Appeals for the Sixth Circuit vacated the District Court’s judgment. Because special counsel are independent contractors, the court determined, they are not entitled to the FDCPA’s state-officer exemption. 785 F. 3d, at 1097–1098. Turning to the deceptive and misleading practices charge, the Court of Appeals concluded that there is a genuine issue of material fact as to whether an unsophisticated consumer would be misled “into believing it is the Attorney General who is collecting on the account.” Id., at 1106. The court therefore remanded the case for trial on this issue. Id., at 1110. Judge Sutton dissented from both holdings. In his view, “deputizing . . . private lawyers to act as assistant attorneys general makes them ‘officers’ of the State for . . . collection purposes.” Ibid. He further concluded that special counsel’s use of the Attorney General’s letterhead “accurately describes the relevant legal realities—that the law firm acts as an agent of the Attorney General and stands in [his] shoes . . . in collecting money owed to the State.” Id., at 1110–1111. The Sixth Circuit denied en banc rehearing. We granted certiorari, 577 U. S. ___ (2015), and now reverse.[3] II As they did below, petitioners maintain that, as special counsel appointed by the Attorney General, they are “officers” exempt from the FDCPA’s governance, and that, in any case, the debt collection letters they sent to respondents comply with the Act. We pretermit the question whether, as petitioners contend and Judge Sutton would have held, special counsel qualify as state officers. For purposes of this decision, we assume, arguendo, that special counsel are not “officers” within the meaning of the Act and, therefore, rank simply as “debt collectors” within the FDCPA’s compass. We conclude, nevertheless, that petitioners complied with the Act, as their use of the Attorney General’s letterhead accurately conveys that special counsel act on behalf of the Attorney General. Special counsel’s use of the Attorney General’s letterhead at the Attorney General’s direction does not offend §1692e’s general prohibition against “false . . . or misleading representation[s].” The letterhead identifies the principal—Ohio’s Attorney General—and the signature block names the agent—a private lawyer hired as outside counsel to the Attorney General. It would not transgress §1692e, respondents acknowledge, if, in lieu of using the Attorney General’s letterhead, special counsel’s communications opened with a bold-face statement: “We write to you as special counsel to the [A]ttorney [G]eneral who has authorized us to collect a debt you owe to [the State or an instrumentality thereof].” Tr. of Oral Arg. 31 (internal quotation marks omitted). If that representation is accurate, i.e., not “false . . . or misleading,” it would make scant sense to rank as unlawful use of a letterhead conveying the very same message, particularly in view of the inclusion of special counsel’s separate contact information and the conspicuous notation that the letter is sent by a debt collector.[4] Our conclusion is bolstered by the character of the relationship between special counsel and the Attorney General. As earlier recounted, special counsel “provide legal services on behalf of the Attorney General to one or more State Clients” in furtherance of the Attorney General’s responsibilities as debt collector for state-owned entities and instrumentalities. App. 143–144. In performing this function, special counsel work closely with attorneys in the Attorney General’s Office. For example, Assistant Attorneys General “frequently assist Special Counsel in drafting pleadings, and sometimes join cases as co-counsel to assist Special Counsel with particularly sensitive or complex cases.” Id., at 102. Special counsel and Assistant Attorneys General even stand in one another’s stead, as needed, to cover proceedings in ongoing litigation. Ibid. Given special counsel’s alliance with attorneys within the Attorney General’s Office, a debtor’s impression that a letter from special counsel is a letter from the Attorney General’s Office is scarcely inaccurate.[5] On safe ground with respect to §1692e’s general proscription against false and misleading representations, special counsel’s use of the Attorney General’s letterhead is consistent too with §1692e(9)’s specific prohibition against “falsely represent[ing]” that a communication is “authorized, issued, or approved” by a State. In enacting this provision, Congress sought to prevent debt collectors from “misrepresenting” that they are “government official[s].” S. Rep. No. 95–382, p. 8 (1977). Here, the Attorney General authorized—indeed required—special counsel to use his letterhead in sending debt collection communications. Special counsel create no false impression in doing just what they have been instructed to do. Instead, their use of the Attorney General’s letterhead conveys on whose authority special counsel writes to the debtor. As a whole, the communication alerts the debtor to both the basis for the payment obligation and the official responsible for enforcement of debts owed to the State, while the signature block conveys who the Attorney General has engaged to collect the debt. Nor did special counsel, in sending letters on the Attorney General’s letterhead, use a name other than their “true name,” in violation of §1692e(14). Although the FDCPA does not say “what a ‘true name’ is, its import is straightforward: A debt collector may not lie about his institutional affiliation.” 785 F. 3d, at 1115 (Sutton, J., dissenting). Special counsel do not employ a false name when using the Attorney General’s letterhead at his instruction, for special counsel, as the Attorney General’s agents, act for him in debt-related matters. Far from misrepresenting special counsel’s identity, letters sent by special counsel accurately identify the office primarily responsible for collection of the debt (the Attorney General), special counsel’s affiliation with that office, and the address (special counsel’s law firm) to which payment should be sent.[6] We further note a federalism concern. “Ohio’s enforcement of its civil code—by collecting money owed to it—[is] a core sovereign function.” Gillie v. Law Office of Eric A. Jones, LLC, No. 14–3836 (CA6, July 14, 2015), App. to Pet. for Cert. 10a (Sutton, J., dissenting from denial of rehearing en banc). Ohio’s Attorney General has chosen toappoint special counsel to assist him in fulfilling his obliga-tion to collect the State’s debts, and he has instructed his appointees to use his letterhead when acting on his behalf. There is no cause, in this case, to construe federal law in a manner that interferes with “States’ arrangements for conducting their own governments.” Nixon v. Missouri Municipal League, 541 U. S. 125, 140 (2004) (citing Greg-ory v. Ashcroft, 501 U. S. 452, 460 (1991) ). The Sixth Circuit’s contrary exposition is unconvincing. Use of the Attorney General’s letterhead, the Court of Appeals emphasized, has led to confusion among debtors, as the Attorney General has received phone calls inquiring whether letters sent by special counsel are authentic. 785 F. 3d, at 1107. But the Sixth Circuit overlooked that the Attorney General’s prompt and invariable answer to those inquiries was “yes.” To the extent that consumers may be concerned that the letters are a “scam,” the solution is for special counsel to say more, not less, about their role as agents of the Attorney General. Special counsel’s use of the Attorney General’s letterhead, furthermore, encourages consumers to use official channels to ensure the legitimacy of the letters, assuaging the very concern the Sixth Circuit identified. In addition to the specter of consumer confusion, the Sixth Circuit stressed the risk of intimidation—that the Attorney General’s letterhead would “place pressure on those individuals receiving the letters” to pay their state debts. Id., at 1105. There are two bases for this concern, neither of which is persuasive. First, invocation of the Attorney General’s imprimatur could lead debtors to prioritize their debt to the State over other, private debts out of a belief that the consequences of failing to pay a state debt would be more severe. This impression is not false; the State does have enforcement powers beyond those afforded private creditors. A debtor’s tax refund, for example, “may be applied in satisfaction” of her debt, regardless of whether the State has obtained a judgment, Ohio Rev. Code Ann. §5747.12 (Lexis 2013), and a debt owed to the State takes priority over most private debts in state probate proceedings, §2117.25(A) (Lexis Supp. 2015). “The special consequences of state debts explain why the Act bars debt collectors unaffiliated with a State from using the State’s name to scare debtors into paying. When the State itself is doing the demanding, however, nothing about the resulting fear misleads.” 785 F. 3d, at 1116 (Sutton, J., dissenting). In other words, §1692e bars debt collectors from deceiving or misleading consumers; it does not protect consumers from fearing the actual consequences of their debts. Second, debtors might worry that the letters imply that the Attorney General, as the State’s top law enforcement official, intends to take punitive action against them. “But neither of the milquetoast letters [received by respondents] . . . threatens criminal prosecution, civil penalties, or any action whatsoever.” Id., at 1116–1117. Use of the Attorney General’s letterhead merely clarifies that the debt is owed to the State, and the Attorney General is the State’s debt collector. The FDCPA is not sensibly read to require special counsel to obscure that reality.[7] * * * For the reasons stated, the judgment of the Court of Appeals for the Sixth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 As noted above, Mark Sheriff, not Sarah Sheriff, was appointed special counsel. 2 Ohio Rev. Code Ann. §109.08 (Lexis 2014) requires the Attorney General to provide special counsel with his “official letterhead stationery” for the collection of tax debts. The Attorney General has interpreted this provision as mandating the use of his letterhead for tax claims, but permitting its use for the collection of other debts. Whether this is a correct interpretation of Ohio law is not before us. 3 We granted the petition for certiorari filed by Mark Sheriff, Sarah Sheriff, the Wiles firm, and the Ohio Attorney General. Jones and the Law Offices of Eric A. Jones, LLC, filed a separate petition for certio-rari as well as a separate brief in this case in support of petitioners. We refer to defendants collectively as “petitioners.” 4 Although respondents argued below that Sarah Sheriff’s inaccurate use of the “special counsel” designation also violates the FDCPA, they have not pursued that argument before this Court. In any case, the letter merely conveyed the debtor’s remaining balance, without any suggestion of followup action. Sarah Sheriff’s misstatement of her title thus qualifies as an immaterial, harmless mistake. 5 We address here only “special counsel.” The considerationsrelevant to that category may not carry over to other debt-collector relationships. 6 Because we conclude that the letters sent by petitioners were truthful, we need not consider the parties’ arguments as to whether a false or misleading statement must be material to violate the FDCPA, or whether a potentially false or misleading statement should be viewed from the perspective of “the least sophisticated consumer,” Brief for Respondent Gillie et al. 57, or “[t]he average consumer who has defaulted on a debt,” Brief for Petitioners 41. 7 Having determined that use of the Attorney General’s letterhead inaccurately suggested that the letters were from the Attorney General’s Office, the Sixth Circuit remanded to the District Court for trial on whether this practice was “materially false, deceptive and misleading.” Gillie v. Law Office of Eric A. Jones, LLC, 785 F. 3d 1091, 1109–1110 (2015). But all of the relevant facts are undisputed, and the application of the FDCPA to those facts is a question of law. The District Court therefore properly granted summary judgment for defendants.

578.US.2015_15-109

This case began with two suits filed by respondent Walter Himmelreich, a federal prisoner. He first filed suit against the United States, alleging that a severe beating he received from a fellow inmate was the result of negligence by prison officials. The Government treated the suit as a claim under the Federal Tort Claims Act (FTCA), which allows plaintiffs to seek damages from the United States for certain torts committed by federal employees, 28 U. S. C. §1346(b), “[s]ubject to the provisions of chapter 171” of Title 28. But an “Exceptions” section of the FTCA dictates that “the provisions of [Chapter 171] and section 1346(b) of this title . . . shall not apply” to certain categories of claims. The Government moved to dismiss the action on the ground that the claim fell into the exception for “[a]ny claim based upon . . . the exercise or performance . . . [of] a discretionary function,” namely, deciding where to house inmates, §2680(a). While the motion was pending, Himmelreich filed a second suit: a constitutional tort suit against individual Bureau of Prison employees, again alleging that his beating was the result of prison officials’ negligence. Ordinarily, the FTCA would have no bearing on that claim. But after the dismissal of Himmelreich’s first suit, the individual employee defendants argued that Himmelreich’s second suit was foreclosed by the FTCA’s judgment bar provision, according to which a judgment in an FTCA suit forecloses any future suit against individual employees. Agreeing, the District Court granted summary judgment in favor of the individual prison employees. The Sixth Circuit reversed, however, holding that the judgment bar provision did not apply to Himmelreich’s suit. Held: The judgment bar provision does not apply to the claims dismissed for falling within the “Exceptions” section of the FTCA. Pp. 3–10. (a) The FTCA explicitly excepts from its coverage certain categories of claims, including the one into which Himmelreich’s first suit fell. If, as the Government maintains, Chapter 171’s judgment bar provision applies to claims in that “Exceptions” category, it applied to Himmelreich’s first suit and would preclude any future actions, including his second suit. On Himmelreich’s reading, however, the provision does not apply and he may proceed with his second suit. Pp. 3–5. (b) Himmelreich is correct. The FTCA’s “Exceptions” section reads: “[T]he provisions of this chapter”—Chapter 171—“shall not apply to . . . [a]ny claim based upon . . . the exercise or performance . . . [of] a discretionary function or duty.” §2680(a). The judgment bar is a provision of Chapter 171. The “Exceptions” section’s plain text thus dictates that the judgment bar does “not apply” to cases that, like Himmelreich’s first suit, are based on the performance of a discretionary function. Because the judgment bar provision does not apply to Himmelreich’s first suit, his second suit—against individual prison employees—should be permitted to go forward. Nothing about the “Exceptions” section or the judgment bar provision gives this Court any reason to disregard the plain text of the statute. P. 5. (c) United States v. Smith, 499 U. S. 160 , does not require a different result. There, the Court found that the exclusive remedies provision of Chapter 171—which prevents a plaintiff from suing an employee where the FTCA would allow him to sue the United States instead, see §2679(b)(1)—applied to a claim for injuries sustained at a hospital in Italy, even though that claim fell within the category of “[a]ny claim arising in a foreign country,” one of the “Exceptions” to which “the provisions of [Chapter 171] . . . shall not apply,” §2680(k). Smith’s outcome, the Government argues, forecloses a literal reading of the “Exceptions” provision, but Smith does not control here. First, Smith does not even mention the “Exceptions” section’s “shall not apply” language. Second, the exclusive remedies provision at issue there was enacted as part of the Federal Employee Liability Reform and Tort Compensation Act of 1988, which also contained a mechanism to convert tort suits against Government employees into FTCA suits “subject to the limitations and exceptions applicable to those actions.” 499 U. S., at 166 (quoting §2679(d)(4); emphasis in Smith). By taking note of those “limitations and exceptions,” the Smith Court reasoned, the Liability Reform Act was intended to apply to the “Exceptions” categories of claims. Nothing in the text of the judgment bar provision compels the same result here. Pp. 5–7. (d) The Government’s remaining counterargument is a parade of horribles that it believes will come to pass if every provision of Chapter 171 “shall not apply” to the “Exceptions” categories of claims, but it raises few concerns about the judgment bar provision itself. If the Government is right about Chapter 171’s other provisions, the Court may hold so in the appropriate case, see Smith, 499 U. S., at 175, but the reading adopted here yields utterly sensible results. Had the District Court in this case issued a judgment dismissing Himmelreich’s first suit because, e.g., the prison employees were not negligent, it would make sense that the judgment bar provision would prevent a second suit against the employees. But where an FTCA claim is dismissed because it falls within one of the “Exceptions,” the dismissal signals merely that the United States cannot be held liable for a particular claim; it has no logical bearing on whether an employee can be liable instead. Pp. 7–9. 766 F. 3d 576, affirmed and remanded. Sotomayor, J., delivered the opinion for a unanimous Court.

The Federal Tort Claims Act (FTCA) allows plaintiffs to seek damages from the United States for certain torts committed by federal employees. 28 U. S. C. §§1346(b), 2674. Many of the FTCA’s procedural provisions are contained in a single chapter of the United States Code, Chapter 171. See §§2671–2680. But an “Exceptions” section of the FTCA dictates that “the provisions of [Chapter 171] . . . shall not apply” to certain categories of claims. At issue in this case is whether one of the “provisions of [Chapter 171]”—the so-called judgment bar provision, §2676—might nonetheless apply to one of the excepted claims. We conclude it does not. I A This case began with two suits filed by Walter Himmelreich. In each, Himmelreich alleged that he had been severely beaten by a fellow inmate in federal prison and that the beating was the result of prison officials’ negligence. At the time of the beating, Himmelreich was incarcerated for producing child pornography. His assailant had warned prison officials that he would “ ‘smash’ ” a pedophile if given the opportunity but was nonetheless released into the general prison population, where he assaulted Himmelreich. App. 46. Himmelreich filed a first suit against the United States. The Government treated this first suit as a claim under the FTCA and moved to dismiss the action, arguing that the claim fell into one of the “Exceptions” to the FTCA for “[a]ny claim based upon . . . the exercise or performance . . . [of] a discretionary function,” namely, deciding where to house inmates. §2680(a). The District Court granted the Government’s motion to dismiss. (Neither party here challenges the outcome of that first suit.) But before the District Court dismissed that first suit, Himmelreich filed a second suit, this one a constitutional tort suit against individual Bureau of Prison employees rather than against the United States. Ordinarily, the FTCA would have nothing to say about such claims. But after the dismissal of Himmelreich’s first suit, the individual employee defendants argued that Himmelreich’s second suit was foreclosed by the FTCA’s judgment bar provision, according to which a judgment in an FTCA suit forecloses any future suit against individual employees. See §2676. As relevant here, the District Court agreed and granted summary judgment in favor of the individual prison employees. Himmelreich appealed that ruling. The Sixth Circuit reversed, holding that the judgment bar provision did not apply to Himmelreich’s suit. Himmelreich v. Federal Bureau of Prisons et al., 766 F. 3d 576 (2014) (per curiam). We granted certiorari to resolve a Circuit split on whether the judgment bar provision applies to suits that, like Himmelreich’s, are dismissed as falling within an “Exceptio[n]” to the FTCA.[1] 577 U. S. ___ (2015). B The FTCA’s provisions are contained in two areas of the United States Code. One, 28 U. S. C. §1346(b), gives federal district courts exclusive jurisdiction over tort claims against the United States for the acts of its employees “[s]ubject to the provisions of chapter 171” of Title 28.[2] Chapter 171, in turn, is labeled “Tort Claims Procedure” and comprises the remaining provisions of the FTCA. §§2671–2680. Chapter 171 contains an array of provisions. Some provisions govern how FTCA claims are to be adjudicated. See, e.g., §2674 (specifying scope of United States’ liability); §2675(a) (exhaustion requirement); §2678 (restricting attorney’s fees). Other provisions limit plaintiffs’ remedies outside the FTCA. See, e.g., §2679(a) (cannot sue agency for claims within scope of FTCA); §2679(d)(1) (suit against federal employee acting within scope of employment automatically converted to FTCA action). The District Court in this case relied on one suchremedies-limiting provision of Chapter 171, the judgment bar provision.[3] See §2676. Under the judgment bar provision, once a plaintiff receives a judgment (favorable or not) in an FTCA suit, he generally cannot proceed with a suit against an individual employee based on the same underlying facts. The District Court below held that Himmelreich had received a judgment in the first suit (the FTCA suit against the United States) and so could not proceed with the second suit (the individual employee suit based on the same underlying facts). The FTCA explicitly excepts from its coverage certain categories of claims, including the one into which Himmelreich’s first suit fell: “Exceptions “The provisions of this chapter and section 1346(b) of this title shall not apply to— “(a) Any claim based upon . . . the exercise or performance or the failure to exercise or perform a discretionary function or duty . . . whether or not the discretion involved be abused. . . .” §2680. “The provisions of this chapter” referenced in the first line are the provisions of Chapter 171. “[S]ection 1346(b) of this title” is the provision giving district courts FTCA jurisdiction. And the “Exceptions” to which those portions of the FTCA “shall not apply” are 13 categories of claims, such as any claim that—like Himmelreich’s first suit—arises from the performance of a “discretionary function,” §2680(a); “[a]ny claim arising in a foreign country,” §2680(k); and “[a]ny claim arising from the activities of the Tennessee Valley Authority,” §2680(l). Both parties agree that district courts do not have jurisdiction over claims that fall into one of the 13 categories of “Exceptions” because “section 1346(b) of this title”—the provision conferring jurisdiction on district courts—does “not apply” to such claims. Both parties also agree that at least one of “[t]he provisions of [Chapter 171]”—the provision delimiting the United States’ liability, §2674—need “not apply” to claims in the “Exceptions” categories because no court will have jurisdiction to hold the United States liable on such claims in any event. The parties disagree, however, about whether the judgment bar provision of Chapter 171 “shall not apply” to claims in one of the “Exceptions” categories. The Government maintains that the judgment bar provision does apply to such claims. In that case, it applied to Himmelreich’s first suit and would preclude any future actions, including his second suit. Himmelreich urges that it does not apply. On that reading, there is no reason he cannot proceed with his second suit. II Himmelreich is correct. The “Exceptions” section of the FTCA reads: “[T]he provisions of this chapter”—Chapter 171—“shall not apply to . . . [a]ny claim based upon . . . the exercise or performance . . . [of] a discretionary function or duty.” §2680(a). The judgment bar is a provision of Chapter 171; the plain text of the “Exceptions” section therefore dictates that it does “not apply” to cases that, like Himmelreich’s first suit, are based on the performance of a discretionary function. Because the judgment bar provision does not apply to Himmelreich’s first suit, Himmelreich’s second suit—the one against individual prison employees—should be permitted to go forward. Absent persuasive indications to the contrary, we presume Congress says what it means and means what it says. Nothing about the “Exceptions” section or the judgment bar provision gives us any reason to doubt the plain-text result in this case. III A Given the clarity of the “Exceptions” section’s command, a reader might be forgiven for wondering how there could be any confusion about the statute’s operation. The main source of uncertainty on this score, the Government submits, is United States v. Smith, 499 U. S. 160 (1991) . In Smith, we considered another provision of Chapter 171, the exclusive remedies provision. Id., at 162. Under the exclusive remedies provision, a plaintiff generally cannot sue an employee where the FTCA would allow him to sue the United States instead. See §2679(b)(1).[4] The Smith Court held that this exclusive remedies provision applied to a claim for injuries sustained at an Army hospital in Italy, even though that claim fell within the category of “[a]ny claim arising in a foreign country,” one of the “Exceptions” to which “the provisions of [Chapter 171] . . . shall not apply.” §2680(k). The Government argues that our literal reading of the “Exceptions” provision would foreclose Smith’s outcome because the Smith Court applied a provision of Chapter 171 (the exclusive remedies provision) to a claim falling within one of the “Exceptions” categories (a claim arising in a foreign country). Smith, the Government argues, thus establishes that we cannot read the command of the “Exceptions” section literally and that the judgment bar provision therefore should apply to Himmelreich’s discretionary function claim. The Government’s position has some force. Nonetheless, Smith does not control this case. First, Smith does not even cite, let alone discuss, the “shall not apply” language “Exceptions” provision. Second, the exclusive remedies provision at issue in Smith was enacted as part of the Federal Employee Liability Reform and Tort Compensation Act of 1988, which contained a mechanism to reduce the number of tort suits against Government employees. As the Smith Court explained, if “the Attorney General . . . certif[ies] that a Government employee named as defendant was acting within the scope of his employment when he committed the alleged tort,” the Liability Reform Act dictates that the United States be substituted as the sole defendant, and that the action “ ‘shall proceed in the same manner’ ” as an FTCA action “ ‘and shall be subject to the limitations and exceptions applicable to those actions.’ ” 499 U. S., at 166 (quoting §2679(d)(4)); (emphasis in Smith). The Smith Court held that the Liability Reform Act’s reference to “limitations and exceptions” was most naturally read to refer to the “Exceptions” section of the FTCA. And by taking note of the “Exceptions” section, the Smith court reasoned, the Liability Reform Act was intended to apply to those “Exceptions.” In light of the unique language of the Liability Reform Act, Smith is distinguishable from this case. Nothing in the text of the judgment bar provision compels the same result. B The Government’s remaining counterargument amounts to a parade of horribles that it believes will come to pass if every provision of Chapter 171 “shall not apply” to the “Exceptions” categories of claims. See Brief for Petitioners 52. If the Government is right about the other provisions of Chapter 171, the Court may hold so in the appropriate case. See Smith, 499 U. S., at 175. But this case deals only with the judgment bar provision, and, aside from a passing concern about duplicative litigation, the Government does not argue that any such cavalcade would follow if that provision does not apply to the excepted claims. It is enough for our purposes that the statute’s clear directive would not lead to hard-to-explain results when applied to the judgment bar provision in particular. To the contrary, our holding that the judgment bar provision “shall not apply” to the categories of claims in the “Exceptions” section in fact allows the statute to operate in an utterly sensible manner. Ordinarily, the judgment bar provision prevents unnecessarily duplicative litigation. If the District Court in this case had issued a judgment dismissing Himmelreich’s first suit because the prison employees were not negligent, because Himmelreich was not harmed, or because Himmelreich simply failed to prove his claim, it would make little sense to give Himmelreich a second bite at the money-damages apple by allowing suit against the employees: Himmelreich’s first suit would have given him a fair chance to recover dam-ages for his beating. Where an FTCA claim is dismissed because it falls within one of the “Exceptions,” by contrast, the judgment bar provision makes much less sense. The dismissal of a claim in the “Exceptions” section signals merely that the United States cannot be held liable for a particular claim; it has no logical bearing on whether an employee can be held liable instead.[5] To apply the judgment bar so as to foreclose a future suit against an employee thus would be passing strange. The Government’s reading would yield another strange result. According to the Government, the viability of a plaintiff’s meritorious suit against an individual employee should turn on the order in which the suits are filed (or the order in which the district court chooses to address motions). For example, had the District Court in this case addressed the individual employee suit first, there would be no FTCA judgment in the picture, and so the judgment bar provision would not affect the outcome of the suit. The Government’s reading would thus encourage litigants to file suit against individual employees before suing the United States to avoid being foreclosed from recovery altogether. Yet this result is at odds with one of the FTCA’s purposes, channeling liability away from individual employees and toward the United States. See Dalehite v. United States, 346 U. S. 15, 25 (1953) . We decline to ignore the text of the statute to achieve these imprudently restrictive results. Accordingly, we read “[t]he provisions of this chapter . . . shall not apply” as it was written. The judgment bar provision—one of the “provisions of this chapter”—does not apply to the categories of claims in the “Exceptions” sections of the FTCA. We therefore affirm the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 See Hallock v. Bonner, 387 F. 3d 147 (CA2 2004), vacated on other grounds sub nom. Will v. Hallock, 546 U. S. 345 (2006) ; Pesnell v. Arsenault, 543 F. 3d 1038 (CA9 2008); Williams v. Fleming, 597 F. 3d 820, 823–824 (CA7 2010). 2 The precise claims at issue are “claims against the United States, for money damages, accruing on and after January 1, 1945, for injury or loss of property, or personal injury or death caused by the negligent or wrongful act or omission of any employee of the Government while acting within the scope of his office or employment, under circum-stances where the United States, if a private person, would be liable to the claimant in accordance with the law of the place where the act or omission occurred.” 28 U. S. C. §1346(b). 3 It reads in full: “The judgment in an action under section 1346(b) of this title shall constitute a complete bar to any action by the claimant, by reason of the same subject matter, against the employee of the Government whose act or omission gave rise to the claim.” §2676. 4 There is an exception to this provision for suits alleging constitutional violations. See §2679(b)(2)(A). Himmelreich’s second suit—the one against individual prison employees—alleged a violation of the Constitution and so was not foreclosed by the exclusive remedies provision. 5 This conclusion is buttressed by analogy to the common-law doctrine of claim preclusion, which prevents duplicative litigation by barring one party from again suing the other over the same underlying facts. This Court has said that the judgment bar provision “functions in much the same way” as that doctrine. Will, 546 U. S., at 354. (The judgment bar provision supplements common-law claim preclusion by closing a narrow gap: At the time that the FTCA was passed, common-law claim preclusion would have barred a plaintiff from suing the United States after having sued an employee but not vice versa. See Restatement of Judgments §§99, 96(1)(a), Comments b and d (1942). The judgment bar provision applies where a plaintiff first sues the United States and then sues an employee.)

578.US.2015_13-1339

The Fair Credit Reporting Act of 1970 (FCRA) requires consumer reporting agencies to “follow reasonable procedures to assure maximum possible accuracy of” consumer reports, 15 U. S. C. §1681e(b), and imposes liability on “[a]ny person who willfully fails to comply with any requirement [of the Act] with respect to any” individual, §1681n(a). Petitioner Spokeo, Inc., an alleged consumer reporting agency, operates a “people search engine,” which searches a wide spectrum of databases to gather and provide personal information about individuals to a variety of users, including employers wanting to evaluate prospective employees. After respondent Thomas Robins discovered that his Spokeo-generated profile contained inaccurate information, he filed a federal class-action complaint against Spokeo, alleging that the company willfully failed to comply with the FCRA’s requirements. The District Court dismissed Robins’ complaint, holding that he had not properly pleaded injury in fact as required by Article III. The Ninth Circuit reversed. Based on Robins’ allegation that “Spokeo violated his statutory rights” and the fact that Robins’ “personal interests in the handling of his credit information are individualized,” the court held that Robins had adequately alleged an injury in fact. Held: Because the Ninth Circuit failed to consider both aspects of the injury-in-fact requirement, its Article III standing analysis was incomplete. Pp. 5–11. (a) A plaintiff invoking federal jurisdiction bears the burden of establishing the “irreducible constitutional minimum” of standing by demonstrating (1) an injury in fact, (2) fairly traceable to the challenged conduct of the defendant, and (3) likely to be redressed by a favorable judicial decision. Lujan v. Defenders of Wildlife, 504 U. S. 555 –561. Pp. 5–6. (b) As relevant here, the injury-in-fact requirement requires a plaintiff to show that he or she suffered “an invasion of a legally protected interest” that is “concrete and particularized” and “actual or imminent, not conjectural or hypothetical.” Lujan, supra, at 560. Pp. 7–11. (1) The Ninth Circuit’s injury-in-fact analysis elided the independent “concreteness” requirement. Both observations it made concerned only “particularization,” i.e., the requirement that an injury “affect the plaintiff in a personal and individual way,” Lujan, supra, at 560, n. 1, but an injury in fact must be both concrete and particularized, see, e.g., Susan B. Anthony List v. Driehaus, 573 U. S. ___, ___. Concreteness is quite different from particularization and requires an injury to be “de facto,” that is, to actually exist. Pp. 7–8. (2) The Ninth Circuit also failed to address whether the alleged procedural violations entail a degree of risk sufficient to meet the concreteness requirement. A “concrete” injury need not be a “tangible” injury. See, e.g., Pleasant Grove City v. Summum, 555 U. S. 460 . To determine whether an intangible harm constitutes injury in fact, both history and the judgment of Congress are instructive. Congress is well positioned to identify intangible harms that meet minimum Article III requirements, but a plaintiff does not automatically satisfy the injury-in-fact requirement whenever a statute grants a right and purports to authorize a suit to vindicate it. Article III standing requires a concrete injury even in the context of a statutory violation. This does not mean, however, that the risk of real harm cannot satisfy that requirement. See, e.g., Clapper v. Amnesty Int’l USA, 568 U. S. ____. The violation of a procedural right granted by statute can be sufficient in some circumstances to constitute injury in fact; in such a case, a plaintiff need not allege any additional harm beyond the one identified by Congress, see Federal Election Comm’n v. Akins, 524 U. S. 11 –25. This Court takes no position on the correctness of the Ninth Circuit’s ultimate conclusion, but these general principles demonstrate two things: that Congress plainly sought to curb the dissemination of false information by adopting procedures designed to decrease that risk and that Robins cannot satisfy the demands of Article III by alleging a bare procedural violation. Pp. 8–11. 742 F. 3d 409, vacated and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Breyer, and Kagan, JJ., joined. Thomas, J., filed a concurring opinion. Ginsburg, J., filed a dissenting opinion, in which Sotomayor, J., joined.

This case presents the question whether respondent Robins has standing to maintain an action in federal court against petitioner Spokeo under the Fair Credit Reporting Act of 1970 (FCRA or Act), 84Stat. 1127, as amended, 15 U. S. C. §1681 et seq. Spokeo operates a “people search engine.” If an individual visits Spokeo’s Web site and inputs a person’s name, a phone number, or an e-mail address, Spokeo conducts a computerized search in a wide variety of databases and provides information about the subject of the search. Spokeo performed such a search for information about Robins, and some of the information it gathered and then disseminated was incorrect. When Robins learned of these inaccuracies, he filed a complaint on his own behalf and on behalf of a class of similarly situated individuals. The District Court dismissed Robins’ complaint for lack of standing, but a panel of the Ninth Circuit reversed. The Ninth Circuit noted, first, that Robins had alleged that “Spokeo violated his statutory rights, not just the statu-tory rights of other people,” and, second, that “Robins’s personal interests in the handling of his credit information are individualized rather than collective.” 742 F. 3d 409, 413 (2014). Based on these two observations, the Ninth Circuit held that Robins had adequately alleged injury in fact, a requirement for standing under Article III of the Constitution. Id., at 413–414. This analysis was incomplete. As we have explained in our prior opinions, the injury-in-fact requirement requires a plaintiff to allege an injury that is both “concrete and particularized.” Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U. S. 167 –181 (2000) (emphasis added). The Ninth Circuit’s analysis focused on the second characteristic (particularity), but it overlooked the first (concreteness). We therefore vacate the decision below and remand for the Ninth Circuit to consider both aspects of the injury-in-fact requirement. I The FCRA seeks to ensure “fair and accurate credit reporting.” §1681(a)(1). To achieve this end, the Act regulates the creation and the use of “consumer report[s]”[1] by “consumer reporting agenc[ies]”[2] for certain specified purposes, including credit transactions, insurance, licensing, consumer-initiated business transactions, and employment. See §§1681a(d)(1)(A)–(C); §1681b. Enacted long before the advent of the Internet, the FCRA applies to companies that regularly disseminate information bearing on an individual’s “credit worthiness, credit standing, credit capacity, character, general reputation, personal characteristics, or mode of living.” §1681a(d)(1). The FCRA imposes a host of requirements concerning the creation and use of consumer reports. As relevant here, the Act requires consumer reporting agencies to “follow reasonable procedures to assure maximum possible accuracy of” consumer reports, §1681e(b); to notify providers and users of consumer information of their responsibilities under the Act, §1681e(d); to limit the circumstances in which such agencies provide consumer reports “for employment purposes,” §1681b(b)(1); and to post toll-free numbers for consumers to request reports, §1681j(a). The Act also provides that “[a]ny person who willfully fails to comply with any requirement [of the Act] with respect to any [individual[3]] is liable to that [individual]” for, among other things, either “actual damages” or statutory damages of $100 to $1,000 per violation, costs of the action and attorney’s fees, and possibly punitive damages. §1681n(a). Spokeo is alleged to qualify as a “consumer reporting agency” under the FCRA.[4] It operates a Web site that allows users to search for information about other individuals by name, e-mail address, or phone number. In response to an inquiry submitted online, Spokeo searches a wide spectrum of databases and gathers and provides information such as the individual’s address, phone number, marital status, approximate age, occupation, hobbies, finances, shopping habits, and musical preferences. App. 7, 10–11. According to Robins, Spokeo markets its services to a variety of users, including not only “employers who want to evaluate prospective employees,” but also “those who want to investigate prospective romantic partners or seek other personal information.” Brief for Respondent 7. Persons wishing to perform a Spokeo search need not disclose their identities, and much information is available for free. At some point in time, someone (Robins’ complaint does not specify who) made a Spokeo search request for information about Robins, and Spokeo trawled its sources and generated a profile. By some means not detailed in Robins’ complaint, he became aware of the contents of that profile and discovered that it contained inaccurate information. His profile, he asserts, states that he is married, has children, is in his 50’s, has a job, is relatively affluent, and holds a graduate degree. App. 14. According to Robins’ complaint, all of this information is incorrect. Robins filed a class-action complaint in the United States District Court for the Central District of California, claiming, among other things, that Spokeo willfully failed to comply with the FCRA requirements enumerated above. The District Court initially denied Spokeo’s motion to dismiss the complaint for lack of jurisdiction, but later reconsidered and dismissed the complaint with prejudice. App. to Pet. for Cert. 23a. The court found that Robins had not “properly pled” an injury in fact, as required by Article III. Ibid. The Court of Appeals for the Ninth Circuit reversed. Relying on Circuit precedent,[5] the court began by stating that “the violation of a statutory right is usually a sufficient injury in fact to confer standing.” 742 F. 3d, at 412. The court recognized that “the Constitution limits the power of Congress to confer standing.” Id., at 413. But the court held that those limits were honored in this case because Robins alleged that “Spokeo violated his statutory rights, not just the statutory rights of other people,” and because his “personal interests in the handling of his credit information are individualized rather than collective.” Ibid. (emphasis in original). The court thus concluded that Robins’ “alleged violations of [his] statutory rights [were] sufficient to satisfy the injury-in-fact requirement of Article III.” Id., at 413–414. We granted certiorari. 575 U. S. ___ (2015). II A The Constitution confers limited authority on each branch of the Federal Government. It vests Congress with enumerated “legislative Powers,” Art. I, §1; it confers upon the President “[t]he executive Power,” Art. II, §1, cl. 1; and it endows the federal courts with “[t]he judicial Power of the United States,” Art. III, §1. In order to remain faithful to this tripartite structure, the power of the Federal Judiciary may not be permitted to intrude upon the powers given to the other branches. See DaimlerChrysler Corp. v. Cuno, 547 U. S. 332, 341 (2006) ; Lujan v. Defenders of Wildlife, 504 U. S. 555 –560 (1992). Although the Constitution does not fully explain what is meant by “[t]he judicial Power of the United States,” Art. III, § 1, it does specify that this power extends only to “Cases” and “Controversies,” Art. III, §2. And “ ‘[n]o principle is more fundamental to the judiciary’s proper role in our system of government than the constitutional limitation of federal-court jurisdiction to actual cases or controversies.’ ” Raines v. Byrd, 521 U. S. 811, 818 (1997) . Standing to sue is a doctrine rooted in the traditional understanding of a case or controversy. The doctrine developed in our case law to ensure that federal courts do not exceed their authority as it has been traditionally understood. See id., at 820. The doctrine limits the category of litigants empowered to maintain a lawsuit in federal court to seek redress for a legal wrong. See Valley Forge Christian College v. Americans United for Separation of Church and State, Inc., 454 U. S. 464, 473 (1982) ; Warth v. Seldin, 422 U. S. 490 –499 (1975). In this way, “[t]he law of Article III standing . . . serves to prevent the judicial process from being used to usurp the powers of the political branches,” Clapper v. Amnesty Int’l USA, 568 U. S. ___, ___ (2013) (slip op., at 9); Lujan, supra, at 576–577, and confines the federal courts to a properly judicial role, see Warth, supra, at 498. Our cases have established that the “irreducible constitutional minimum” of standing consists of three elements. Lujan, 504 U. S., at 560. The plaintiff must have (1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision. Id., at 560–561; Friends of the Earth, Inc., 528 U. S., at 180–181. The plaintiff, as the party invoking federal jurisdiction, bears the burden of establishing these elements. FW/PBS, Inc. v. Dallas, 493 U. S. 215, 231 (1990) . Where, as here, a case is at the pleading stage, the plaintiff must “clearly . . . allege facts demonstrating” each element. Warth, supra, at 518.[6] B This case primarily concerns injury in fact, the “[f ]irst and foremost” of standing’s three elements. Steel Co. v. Citizens for Better Environment, 523 U. S. 83, 103 (1998) . Injury in fact is a constitutional requirement, and “[i]t is settled that Congress cannot erase Article III’s standing requirements by statutorily granting the right to sue to a plaintiff who would not otherwise have standing.” Raines, supra, at 820, n. 3; see Summers v. Earth Island Institute, 555 U. S. 488, 497 (2009) ; Gladstone, Realtors v. Village of Bellwood, 441 U. S. 91, 100 (1979) (“In no event . . . may Congress abrogate the Art. III minima”). To establish injury in fact, a plaintiff must show that he or she suffered “an invasion of a legally protected interest” that is “concrete and particularized” and “actual or imminent, not conjectural or hypothetical.” Lujan, 504 U. S., at 560 (internal quotation marks omitted). We discuss the particularization and concreteness requirements below. 1 For an injury to be “particularized,” it “must affect the plaintiff in a personal and individual way.” Ibid., n. 1; see also, e.g., Cuno, supra, at 342 (“ ‘plaintiff must allege personal injury’ ”); Whitmore v. Arkansas, 495 U. S. 149, 155 (1990) (“ ‘distinct’ ”); Allen v. Wright, 468 U. S. 737, 751 (1984) (“personal”); Valley Forge, supra, at 472 (standing requires that the plaintiff “ ‘personally has suffered some actual or threatened injury’ ”); United States v. Richardson, 418 U. S. 166, 177 (1974) (not “undifferenti-ated”); Public Citizen, Inc. v. National Hwy. Traffic Safety Admin., 489 F. 3d 1279, 1292–1293 (CADC 2007) (collecting cases).[7] Particularization is necessary to establish injury in fact, but it is not sufficient. An injury in fact must also be “concrete.” Under the Ninth Circuit’s analysis, however, that independent requirement was elided. As previously noted, the Ninth Circuit concluded that Robins’ complaint alleges “concrete, de facto” injuries for essentially two reasons. 742 F. 3d, at 413. First, the court noted that Robins “alleges that Spokeo violated his statutory rights, not just the statutory rights of other people.” Ibid. Second, the court wrote that “Robins’s personal interests in the handling of his credit information are individualized rather than collective.” Ibid. (emphasis added). Both of these observations concern particularization, not concreteness. We have made it clear time and time again that an injury in fact must be both concrete and particularized. See, e.g., Susan B. Anthony List v. Driehaus, 573 U. S. ___, ___ (2014) (slip op., at 8); Summers, supra, at 493; Sprint Communications Co. v. APCC Services, Inc., 554 U. S. 269, 274 (2008) ; Massachusetts v. EPA, 549 U. S. 497, 517 (2007) . A “concrete” injury must be “de facto”; that is, it must actually exist. See Black’s Law Dictionary 479 (9th ed. 2009). When we have used the adjective “concrete,” we have meant to convey the usual meaning of the term—“real,” and not “abstract.” Webster’s Third New International Dictionary 472 (1971); Random House Dictionary of the English Language 305 (1967). Concreteness, therefore, is quite different from particularization. 2 “Concrete” is not, however, necessarily synonymous with “tangible.” Although tangible injuries are perhaps easier to recognize, we have confirmed in many of our previous cases that intangible injuries can nevertheless be concrete. See, e.g., Pleasant Grove City v. Summum, 555 U. S. 460 (2009) (free speech); Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U. S. 520 (1993) (free exercise). In determining whether an intangible harm constitutes injury in fact, both history and the judgment of Congress play important roles. Because the doctrine of standing derives from the case-or-controversy requirement, and because that requirement in turn is grounded in historical practice, it is instructive to consider whether an alleged intangible harm has a close relationship to a harm that has traditionally been regarded as providing a basis for a lawsuit in English or American courts. See Vermont Agency of Natural Resources v. United States ex rel. Stevens, 529 U. S. 765 –777 (2000). In addition, because Congress is well positioned to identify intangible harms that meet minimum Article III requirements, its judgment is also instructive and important. Thus, we said in Lujan that Congress may “elevat[e] to the status of legally cognizable injuries concrete, de facto injuries that were previously inadequate in law.” 504 U. S., at 578. Similarly, Justice Kennedy’s concurrence in that case explained that “Congress has the power to define injuries and articulate chains of causation that will give rise to a case or controversy where none existed before.” Id., at 580 (opinion concurring in part and concurring in judgment). Congress’ role in identifying and elevating intangible harms does not mean that a plaintiff automatically satisfies the injury-in-fact requirement whenever a statute grants a person a statutory right and purports to authorize that person to sue to vindicate that right. Article III standing requires a concrete injury even in the context of a statutory violation. For that reason, Robins could not, for example, allege a bare procedural violation, divorced from any concrete harm, and satisfy the injury-in-fact requirement of Article III. See Summers, 555 U. S., at 496 (“[D]eprivation of a procedural right without some concrete interest that is affected by the deprivation . . . is insufficient to create Article III standing”); see also Lujan, supra, at 572. This does not mean, however, that the risk of real harm cannot satisfy the requirement of concreteness. See, e.g., Clapper v. Amnesty Int’l USA, 568 U. S. ____. For example, the law has long permitted recovery by certain tort victims even if their harms may be difficult to prove or measure. See, e.g., Restatement (First) of Torts §§569 (libel), 570 (slander per se) (1938). Just as the common law permitted suit in such instances, the violation of a procedural right granted by statute can be sufficient in some circumstances to constitute injury in fact. In other words, a plaintiff in such a case need not allege any additional harm beyond the one Congress has identified. See Federal Election Comm’n v. Akins, 524 U. S. 11 –25 (1998) (confirming that a group of voters’ “inability to obtain information” that Congress had decided to make public is a sufficient injury in fact to satisfy Article III); Public Citizen v. Department of Justice, 491 U. S. 440, 449 (1989) (holding that two advocacy organizations’ failure to obtain information subject to disclosure under the Federal Advisory Committee Act “constitutes a sufficiently distinct injury to provide standing to sue”). In the context of this particular case, these general principles tell us two things: On the one hand, Congress plainly sought to curb the dissemination of false information by adopting procedures designed to decrease that risk. On the other hand, Robins cannot satisfy the demands of Article III by alleging a bare procedural violation. A violation of one of the FCRA’s procedural requirements may result in no harm. For example, even if a consumer reporting agency fails to provide the required notice to a user of the agency’s consumer information, that information regardless may be entirely accurate. In addition, not all inaccuracies cause harm or present any material risk of harm. An example that comes readily to mind is an incorrect zip code. It is difficult to imagine how the dissemination of an incorrect zip code, without more, could work any concrete harm.[8] Because the Ninth Circuit failed to fully appreciate the distinction between concreteness and particularization, its standing analysis was incomplete. It did not address the question framed by our discussion, namely, whether the particular procedural violations alleged in this case entail a degree of risk sufficient to meet the concreteness requirement. We take no position as to whether the Ninth Circuit’s ultimate conclusion—that Robins adequately alleged an injury in fact—was correct. * * * The judgment of the Court of Appeals is vacated, and the case is remanded for proceedings consistent with this opinion. It is so ordered.Notes 1 The Act defines the term “consumer report” as: 2 “The term ‘consumer reporting agency’ means any person which, for monetary fees, dues, or on a cooperative nonprofit basis, regularly engages in whole or in part in the practice of assembling or evaluating consumer credit information or other information on consumers for the purpose of furnishing consumer reports to third parties, and which uses any means or facility of interstate commerce for the purpose of preparing or furnishing consumer reports.” §1681a(f ). 3 This statutory provision uses the term “consumer,” but that term is defined to mean “an individual.” §1681a(c). 4 For purposes of this opinion, we assume that Spokeo is a consumer reporting agency. 5 See Edwards v. First American Corp., 610 F. 3d 514 (CA9 2010), cert. granted sub nom. First American Financial Corp. v. Edwards, 564 U. S. 1018 (2011) , cert. dism’d as improvidently granted, 567 U. S. ___ (2012) ( per curiam). 6 “That a suit may be a class action . . . adds nothing to the question of standing, for even named plaintiffs who represent a class ‘must allege and show that they personally have been injured, not that injury has been suffered by other, unidentified members of the class to which they belong.’ ” Simon v. Eastern Ky. Welfare Rights Organization, 426 U. S. 26, 40, n. 20 (1976) (quoting Warth, 422 U. S., at 502). 7 The fact that an injury may be suffered by a large number of people does not of itself make that injury a nonjusticiable generalized grievance. The victims’ injuries from a mass tort, for example, are widely shared, to be sure, but each individual suffers a particularized harm. 8 We express no view about any other types of false information that may merit similar treatment. We leave that issue for the Ninth Circuit to consider on remand.

577.US.2015_14-1209

The Alaska National Interest Lands Conservation Act (ANILCA) set aside 104 million acres of land in Alaska for preservation purposes. Under ANILCA, those lands were placed into “conservation system units,” which were defined to include “any unit in Alaska of the National Park System, National Wildlife Refuge System, National Wild and Scenic Rivers Systems, National Trails System, National Wilderness Preservation System, or a National Forest Monument.” 16 U. S. C. §3102(4). In addition to federal land, over 18 million acres of state, Native Corporation, and private land were also included within the boundaries of those conservation system units. In 2007, John Sturgeon was piloting his hovercraft over a stretch of the Nation River that flows through the Yukon-Charley Rivers National Preserve, a conservation system unit in Alaska that is managed by the National Park Service. Alaska law permits the use of hovercraft. National Park Service regulations do not. See 36 CFR §2.17(e). Park Service rangers approached Sturgeon, informing him that hovercraft were prohibited within the preserve under Park Service regulations. Sturgeon protested that Park Service regulations did not apply because the river was owned by the State of Alaska. The rangers ordered Sturgeon to remove his hovercraft from the preserve, and he complied. Sturgeon later filed suit against the Park Service in the United States District Court for the District of Alaska, seeking declaratory and injunctive relief permitting him to operate his hovercraft within the boundaries of the Yukon-Charley. Alaska intervened in support of Sturgeon. The Secretary of the Interior has authority to “prescribe regulations” concerning “boating and other activities on or relating to water located within System units.” 54 U. S. C. §100751(b). The Park Service’s hovercraft regulation was adopted pursuant to Section 100751(b). The hovercraft ban is not limited to Alaska, but instead has effect in federally managed preservation areas across the country. Section 103(c) of ANILCA, in contrast, addresses the scope of the Park Service’s authority over lands within the boundaries of conservation system units in Alaska. The first sentence of Section 103(c) specifies the property included as a portion of those units. It states: “Only those lands within the boundaries of any conservation system unit which are public lands (as such term is defined in this Act) shall be deemed to be included as a portion of such unit.” 16 U. S. C. §3103(c). ANILCA defines the word “land” to include “lands, waters, and interests therein,” and the term “public lands” to include lands to which the United States has “title,” with certain exceptions. §3102. The second sentence of Section 103(c) concerns the Park Service’s authority to regulate “non-public” lands in Alaska, which include state, Native Corporation, and private property. It provides: “No lands which, before, on, or after December 2, 1980, are conveyed to the State, to any Native Corporation, or to any private party shall be subject to the regulations applicable solely to public lands within such units.” §3103(c). The third sentence of Section 103(c) explains how new lands become part of conservation system units: “If the State, a Native Corporation, or other owner desires to convey any such lands, the Secretary may acquire such lands in accordance with applicable law (including this Act), and any such lands shall become part of the unit, and be administered accordingly.” Ibid. Interpreting Section 103(c) of ANILCA, the District Court granted summary judgment to the Park Service, and the Ninth Circuit affirmed in pertinent part. According to the Ninth Circuit, because the hovercraft regulation “applies to all federal-owned lands and waters administered by [the Park Service] nationwide, as well as all navigable waters lying within national parks,” the hovercraft ban does not apply “solely” within conservation system units in Alaska. 768 F. 3d 1066, 1077. The Ninth Circuit concluded that the Park Service therefore has authority to enforce its hovercraft regulation on the Nation River. The Ninth Circuit did not address whether the Nation River counts as “public land” for purposes of ANILCA. Held: The Ninth Circuit’s interpretation of Section 103(c) is inconsistent with both the text and context of ANILCA. Pp. 12–16. (a) The Ninth Circuit’s interpretation of Section 103(c) violates “a fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme,” Roberts v. Sea-Land Services, Inc., 566 U. S. ___, ___. ANILCA repeatedly recognizes that Alaska is different, and ANILCA itself accordingly carves out numerous Alaska-specific exceptions to the Park Service’s general authority over federally managed preservation areas. Those Alaska-specific provisions reflect the simple truth that Alaska is often the exception, not the rule. Yet the reading below would prevent the Park Service from recognizing Alaska’s unique conditions. Under that reading, the Park Service could regulate “non-public” lands in Alaska only through rules applicable outside Alaska as well. The Court concludes that, whatever the reach of the Park Service’s authority under ANILCA, Section 103(c) did not adopt such a “topsy-turvy” approach. Pp. 12–14. (b) Moreover, it is clear that Section 103(c) draws a distinction between “public” and “non-public” lands within the boundaries of conservation system units in Alaska. And yet, according to the court below, if the Park Service wanted to differentiate between that “public” and “non-public” land in an Alaska-specific way, it would have to regulate the “non-public” land pursuant to rules applicable outside Alaska, and the “public” land pursuant to Alaska-specific provisions. Assuming the Park Service has authority over “non-public” land in Alaska (an issue the Court does not decide), the Court concludes that this is an implausible reading of the statute. The Court therefore rejects the interpretation of Section 103(c) adopted by the court below. Pp. 14–15. (c) The Court does not reach the remainder of the parties’ arguments. In particular, it does not decide whether the Nation River qualifies as “public land” for purposes of ANILCA. It also does not decide whether the Park Service has authority under Section 100751(b) to regulate Sturgeon’s activities on the Nation River, even if the river is not “public” land, or whether—as Sturgeon argues—any such authority is limited by ANILCA. Finally, the Court does not consider whether the Park Service has authority under ANILCA over both “public” and “non-public” lands within the boundaries of conservation system units in Alaska, to the extent a regulation is written to apply specifically to both types of land. The Court leaves those arguments to the lower courts for consideration as necessary. Pp. 15–16. 768 F. 3d 1066, vacated and remanded. Roberts, C. J., delivered the opinion for a unanimous Court.

For almost 40 years, John Sturgeon has hunted moose along the Nation River in Alaska. Because parts of the river are shallow and difficult to navigate, Sturgeon travels by hovercraft, an amphibious vehicle capable of gliding over land and water. To reach his preferred hunting grounds, Sturgeon must pilot his hovercraft over a stretch of the Nation River that flows through the Yukon-Charley Rivers National Preserve, a 1.7 million acre federal preservation area managed by the National Park Service. 16 U. S. C. §410hh(10). Alaska law permits the use of hovercraft. National Park Service regulations do not. See 36 CFR §2.17(e) (2015). After Park Service rangers informed Sturgeon that he was prohibited from using his hovercraft within the boundaries of the preserve, Sturgeon filed suit, seeking declaratory and injunctive relief. He argues that the Nation River is owned by the State, and that the Alaska National Interest Lands Conservation Act (ANILCA) prohibits the Park Service from enforcing its regulations on state-owned land in Alaska. The Park Service disagrees, contending that it has authority to regulate waters flowing through federally managed preservation areas. The District Court and the Court of Appeals ruled in favor of the Park Service. We granted certiorari. I In 1867, Secretary of State William Seward, serving under President Andrew Johnson, negotiated a treaty to purchase Alaska from Russia for $7.2 million. Treaty Concerning the Cession of the Russian Possessions in North America, Mar. 30, 1867, 15Stat. 539. In a single stroke, the United States gained 365 million acres of land—an area more than twice the size of Texas. Despite the bargain price of two cents an acre, however, the purchase was mocked by contemporaries as “Seward’s Folly” and President Johnson’s “Polar Bear Garden.” See C. Naske & H. Slotnick, Alaska: A History 92–94 (2011) (Naske & Slotnick); S. Rep. No. 1163, 85th Cong., 1st Sess., 2 (1957). The monikers didn’t stick. In 1898, the “Three Lucky Swedes”—Jafet Lindeberg, Eric Lindblom, and Jon Brynteson—struck gold in Nome, Alaska. As word of their discovery spread, thousands traveled to Alaska to try their hand at mining. Once the gold rush subsided, settlers turned to other types of mining, fishing, and trapping, fueling an emerging export economy. See Naske & Slotnick 128–129, 155, 249–251; D. Wharton, The Alaska Gold Rush 186–187 (1972). Despite newfound recognition of Alaska’s economic potential, however, it was not until the 1950’s that Congress seriously considered admitting Alaska as a State. By that time, it was clear that Alaska was strategically important both in the Pacific and Arctic, and that the Territory was rich in natural resources, including oil. Moreover, the people of Alaska favored statehood. See Naske & Slotnick 201, 224–235. But there was a problem: Out of the 365 million acres of land in Alaska, 98 percent were owned by the Federal Government. As a result, absent a land grant from the Federal Government to the State, there would be little land available to drive private economic activity and contribute to the state tax base. See S. Rep. No. 1163, at 2, 12 (“The expenses of the State of Alaska will be comparatively high, partially due to the vast land areas within the State; but the State would be able to realize revenues from only 2 percent of this vast area unless some provision were made to modify the present land-ownership conditions”). A solution was struck. The 1958 Alaska Statehood Act permitted Alaska to select 103 million acres of “vacant, unappropriated, and unreserved” federal land—just over a quarter of all land in Alaska—for state ownership. §§6(a)–(b), 72Stat. 340. That land grant included “mineral deposits,” which were “subject to lease by the State as the State legislature may direct.” §6(i), id., at 342. Upon statehood, Alaska also gained “title to and ownership of the lands beneath navigable waters” within the State, in addition to “the natural resources within such lands and waters,” including “the right and power to manage, administer, lease, develop, and use the said lands and natural resources.” §3(a), 67Stat. 30, 43 U. S. C. §1311(a); §6(m), 72Stat. 343. With over 100 million acres of land now available to the new State, Alaska could begin to fulfill its state policy “to encourage the settlement of its land and the development of its resources by making them available for maximum use consistent with the public interest.” Alaska Const., Art. VIII, §1 (2014). The Statehood Act did not, however, determine the rights of the Alaska Natives, who asserted aboriginal title to much of the same land now claimed by the State. Naske & Slotnick 287–289. To resolve the dispute, Congress in 1971 passed the Alaska Native Claims Settlement Act (ANCSA), which extinguished aboriginal land claims in Alaska. 85Stat. 688, as amended, 43 U. S. C. §1601 et seq. In exchange, Congress provided for a $960 million settlement and permitted corporations organized by groups of Alaska Natives to select 40 million acres of federal land to manage within the State. §§1605, 1610–1615; Naske & Slotnick 296–297. Congress sought to implement the settlement “rapidly, with certainty, in conformity with the real economic and social needs” of Alaska Natives. §1601(b). In addition to settling the claims of the Alaska Natives, ANCSA directed the Secretary of the Interior to select up to 80 million acres of unreserved federal land in Alaska for addition to the National Park, Forest, Wildlife Refuge, and Wild and Scenic Rivers Systems, subject to congressional approval. §1616(d)(2). When Congress failed to approve the Secretary’s selections, however, President Carter unilaterally designated 56 million acres of federal land in Alaska as national monuments. See Presidential Proclamation Nos. 4611–4627, 3 CFR 69–104 (1978 Comp.). President Carter’s actions were unpopular among many Alaskans, who were concerned that the new monuments would be subject to restrictive federal regulations. Protesters demonstrated in Fairbanks, and more than 2,500 Alaskans participated in the “Great Denali-McKinley Trespass.” The goal of the trespass was to break over 25 Park Service rules in a two-day period—including by camping, hunting, snowmobiling, setting campfires, shooting guns, and unleashing dogs. During the event, a “rider on horseback, acting the part of Paul Revere, galloped through the crowd yelling, ‘The Feds are coming! The Feds are coming!’ ” N. Y. Times, Jan. 15, 1979, p. A8; Anchorage Daily News, Jan. 15, 1979, pp. 1–2. Congress once again stepped in to settle the controversy, passing the Alaska National Interest Lands Conservation Act. 94Stat. 2371, 16 U. S. C. §3101 et seq. ANILCA had two stated goals: First, to provide “sufficient protection for the national interest in the scenic, natural, cultural and environmental values on the public lands in Alaska.” §3101(d). And second, to provide “adequate opportunity for satisfaction of the economic and social needs of the State of Alaska and its people.” Ibid. ANILCA set aside 104 million acres of land in Alaska for preservation purposes, in the process creating ten new national parks, preserves, and monuments—including the Yukon-Charley Rivers National Preserve—and tripling the number of acres set aside in the United States for federal wilderness preservation. See §410hh; Naske & Slotnick 315–316. At the same time, ANILCA specified that the Park Service could not prohibit on those lands certain activities of particular importance to Alaskans. See, e.g., §3170(a) (Secretary must permit reasonable use of vehicles “for travel to and from villages and homesites”); §3201 (Secretary must permit “the taking of fish and wildlife for sport purposes and subsistence uses” within National Preserves in Alaska, subject to regulation and certain exceptions). President Carter’s earlier land designations were rescinded. See §3209(a). Under ANILCA, federal preservation lands in Alaska were placed into “conservation system units,” which were defined to include “any unit in Alaska of the National Park System, National Wildlife Refuge System, National Wild and Scenic Rivers Systems, National Trails System, National Wilderness Preservation System, or a National Forest Monument.” §3102(4). Congress drew the bound-aries of those units to “follow hydrographic divides or em-brace other topographic or natural features,” however, rather than to map the Federal Government’s landholdings. §3103(b). As a consequence, in addition to federal land, over 18 million acres of state, Native Corporation, and private land ended up inside the boundaries of conservation system units. See Brief for Petitioner 6. This brings us back to Sturgeon and his hovercraft. II A One fall day in 2007, Sturgeon was piloting his hovercraft on the Nation River, which rises in the Ogilvie Mountains in Canada and joins the Yukon River within the boundaries of the Yukon-Charley Rivers National Preserve conservation system unit (Yukon-Charley). Sturgeon was headed to a hunting ground upstream from the preserve, just shy of the Canadian border. To reach that hunting ground, dubbed “moose meadows,” Sturgeon had to travel on a portion of the river that flows through the preserve. About two miles into his trip on the Nation River, Sturgeon stopped on a gravel bar to repair the steering cable of his hovercraft. As he was performing the repairs, Sturgeon was approached by three Park Service rangers. The rangers informed him that hovercraft were prohibited under Park Service regulations, and that he was committing a crime by operating his hovercraft within the boundaries of the Yukon-Charley. Despite Sturgeon’s protests that Park Service regulations did not apply because the river was owned by the State of Alaska, the rangers ordered Sturgeon to remove his hovercraft from the preserve. Sturgeon complied, heading home without a moose. Sturgeon now fears that he will be criminally prosecuted if he returns to hunt along the Nation River in his hovercraft. To avoid prosecution, Sturgeon sued the Park Service and several federal officials in the United States District Court for the District of Alaska. He seeks declaratory and injunctive relief permitting him to operate his hovercraft within the boundaries of the Yukon-Charley. Alaska intervened in support of Sturgeon, and the Park Service opposed the suit. The District Court granted summary judgment to the Park Service. Sturgeon v. Masica, 2013 WL 5888230 (Oct. 30, 2013). The Court of Appeals for the Ninth Circuit affirmed in pertinent part. Sturgeon v. Masica, 768 F. 3d 1066 (2014). We granted certiorari. 576 U. S. ___ (2015). B The Secretary of the Interior has authority to “prescribe regulations” concerning “boating and other activities on or relating to water located within System units, including water subject to the jurisdiction of the United States.” 54 U. S. C. §100751(b) (2012 ed., Supp. II). “System units” are in turn defined as “any area of land and water administered by the Secretary, acting through the Director [of the Park Service], for park, monument, historic, parkway, recreational, or other purposes.” §§100102, 100501. The Park Service’s hovercraft regulation was adopted pursuant to Section 100751(b). The hovercraft ban applies not only within “[t]he boundaries of federally owned lands and waters administered by the National Park Service,” but also to “[w]aters subject to the jurisdiction of theUnited States located within the boundaries of the National Park System, including navigable waters . . . withoutregard to the ownership of submerged lands.” 36 CFR §1.2(a). The hovercraft ban is not limited to Alaska, but instead has effect in federally managed preservation areas across the country. Section 103(c) of ANILCA, in contrast, addresses the scope of the Park Service’s authority over lands within the boundaries of conservation system units in Alaska. The first sentence of Section 103(c) specifies the property included as a portion of those units. It states: “Only those lands within the boundaries of any conservation system unit which are public lands (as such term is defined in this Act) shall be deemed to be included as a portion of such unit.” 16 U. S. C. §3103(c). ANILCA defines the word “land” to include “lands, waters, and interests therein,” and the term “public lands” to include “lands the title to which is in the United States after December 2, 1980,” with certain exceptions. §3102. In sum, only “lands, waters, and interests therein” to which the United States has “title” are considered “public” land “included as a portion” of the conservation system units in Alaska. The second sentence of Section 103(c) concerns the Park Service’s authority to regulate “non-public” lands in Alaska, which include state, Native Corporation, and private property. It provides: “No lands which, before, on, or after December 2, 1980, are conveyed to the State, to any Native Corporation, or to any private party shall be subject to the regulations applicable solely to public lands within such units.” §3103(c). The third sentence of Section 103(c) explains how new lands become part of conservation system units: “If the State, a Native Corporation, or other owner desires to convey any such lands, the Secretary may acquire such lands in accordance with applicable law (including this Act), and any such lands shall become part of the unit, and be administered accordingly.” Ibid. C The parties dispute whether Section 103(c) of ANILCA created an Alaska-specific exception to the Park Service’s general authority over boating and related activities in federally managed preservation areas. Sturgeon, the Park Service, and the Ninth Circuit each adopt a different reading of Section 103(c), reaching different conclusions about the scope of the Park Service’s powers. Sturgeon, joined by the State, understands Section 103(c) to stand for a simple proposition: The Park Service is prohibited from regulating “non-public” land in Alaska as if that land were owned by the Federal Government. He contends that his reading is consistent with the history of federal land management in Alaska, beginning with the Alaska Statehood Act and culminating in ANILCA. Sturgeon’s argument proceeds in two steps. First, he asserts that the Nation River is not “public land” for purposes of ANILCA and is therefore not part of the Yukon-Charley. As discussed, ANILCA defines “public lands” as lands to which the United States has “title.” 16 U. S. C. §3102. And Section 103(c) provides that “[o]nly those lands within the boundaries of any conservation system unit which are public lands (as such term is defined in this Act) shall be deemed to be included as a portion of such unit.” §3103(c). Sturgeon argues that the Nation River is not “public land” because it is owned by the State and not by the Federal Government. To support his argument, Sturgeon relies on the Alaska Statehood Act, which granted ownership of the submerged lands beneath the navigable waters in Alaska, and the resources within those waters, to the State. See §6(m), 72Stat. 343; 43 U. S. C. §1311(a). He also cites this Court’s decision in United States v. California, 436 U. S. 32 (1978) , which stated that “the Submerged Lands Act transferred title to and ownership of the submerged lands and waters” to the States. Id., at 40 (internal quotation marks omitted). Because the State and not the Federal Government owns the Nation River, Sturgeon urges, it is not “public” land under ANILCA and is therefore not part of the Yukon-Charley. Second, Sturgeon asserts that because the Nation River is not part of the Yukon-Charley, the Park Service lacks authority to regulate it. His argument rests on the second sentence of Section 103(c), which states that “[n]o lands which, before, on, or after December 2, 1980, are conveyed to the State, to any Native Corporation, or to any private party shall be subject to the regulations applicable solely to public lands within such units.” 16 U. S. C. §3103(c). Sturgeon argues that the phrase “regulations applicable solely to public lands within such units” refers to those regulations that apply “solely” by virtue of the Park Service’s “authority to manage national parks.” Brief for Petitioner 18, 26–27. The word “solely,” Sturgeon contends, simply ensures that “non-public” lands within the boundaries of those units remain subject to laws generally “applicable to both public and private lands (such as the Clean Air Act and Clean Water Act).” Id., at 19. Because the hovercraft regulation was adopted pursuant to the Park Service’s authority over federally managed preservation areas, and is not a law of general applicability like the Clean Air Act or the Clean Water Act, Sturgeon concludes that Section 103(c) bars enforcement of the regulation. The Park Service, in contrast, reads Section 103(c) more narrowly. In its brief in this Court, the Park Service, while defending the reasoning of the Ninth Circuit, relies primarily on very different arguments. The agency stresses that it has longstanding authority to regulate waters within federally managed preservation areas, and that Section 103(c) does not take any of that authority away. In reaching its conclusion, the Park Service disagrees with Sturgeon at each step. First, the Park Service contends that the Nation River is part of the Yukon-Charley. To support that contention, the agency cites ANILCA’s definition of “public lands,” which—as noted—includes “lands, waters, and interests therein” to which the United States has “title.” 16 U. S. C. §3102. The Park Service argues that the United States has “title” to an “interest” in the water within the boundaries of the Yukon-Charley under the reserved water rights doctrine. The reserved water rights doctrine specifies that “when the Federal Government withdraws its land from the public domain and reserves it for a federal purpose, the Government, by implication, reserves appurtenant water then unappropriated to the extent needed to accomplish the purpose of the reservation.” Cappaert v. United States, 426 U. S. 128, 138 (1976) . By creating the Yukon-Charley, the Park Service urges, the Federal Government reserved the water within the boundaries of the conservation system unit to achieve the Government’s conservation goals. As a result, the Federal Government has “title” to an “interest” in the Nation River, making it “public” land subject to Park Service regulations. Second, the Park Service contends that even if the Nation River is not “public” land, the agency still has authority to regulate it. According to the Park Service, the second sentence of Section 103(c) imposes only a limited restriction on the agency’s power, prohibiting it from enforcing on “non-public” lands only those regulations that explicitly apply “solely to public lands.” The hovercraft regulation applies both within “[t]he boundaries of feder-ally owned lands and waters administered by the National Park Service” and to “[w]aters subject to the jurisdiction of the United States located within the boundaries of the National Park System, including navigable waters . . . without regard to the ownership of submerged lands.” 36 CFR §1.2(a). Accordingly, the Park Service asserts, the hovercraft regulation does not apply “solely to public lands,” and Section 103(c) therefore does not prevent enforcement of the regulation. See Brief for Respondents 56–58. The Ninth Circuit, for its part, adopted a reading of Section 103(c) different from the primary argument advanced by the Park Service in this Court. The Court of Appeals did not reach the question whether the Nation River counts as “public” land for purposes of ANILCA. Instead, it held that the phrase “regulations applicable solely to public lands within such units” distinguishes between Park Service regulations that apply solely to “public” lands in Alaska, and Park Service regulations that apply to federally managed preservation areas across the country. In the Ninth Circuit’s view, the Park Service may enforce nationally applicable regulations on both “public” and “non-public” property within the boundaries of conservation system units in Alaska, because such regulations do not apply “solely to public lands within such units.” The Park Service may not, however, apply Alaska-specific regulations to “non-public” lands within the boundaries of those units. According to the Ninth Circuit, because the hovercraft regulation “applies to all federal-owned lands and waters administered by [the Park Service] nationwide, as well as all navigable waters lying within national parks,” the hovercraft ban does not apply “solely” within conservation system units in Alaska. 768 F. 3d, at 1077. The Ninth Circuit concluded that the Park Service therefore has authority to enforce its hovercraft regulation on the Nation River. Id., at 1078. The Ninth Circuit’s holding is subject to some interpretation, but Sturgeon, the State, the Alaska Native Corporations, and the Park Service (at least at times) concur in our understanding of the decision below. See Brief for Petitioner 25; Brief for State of Alaska as Amicus Curiae 23; Brief for Arctic Slope Regional Corporation et al. as Amici Curiae 12–13; Brief for Doyon, Ltd., et al. as Amici Curiae 31–32; Brief for Respondents 20; Tr. of Oral Arg. 61; 80 Fed. Reg. 65573 (2015). III We reject the interpretation of Section 103(c) adopted by the Ninth Circuit. The court’s reading of the phrase “regulations applicable solely to public lands within such units” may be plausible in the abstract, but it is ultimately inconsistent with both the text and context of the statute as a whole. Statutory language “cannot be construed in a vacuum. It is a fundamental canon of statutory construction that the words of a statute must be read in their context and with a view to their place in the overall statutory scheme.” Roberts v. Sea-Land Services, Inc., 566 U. S. ___, ___ (2012) (slip op., at 6) (internal quotation marks omitted). Under the reading of the statute adopted below, the Park Service may apply nationally applicable regulations to “non-public” lands within the boundaries of conservation system units in Alaska, but it may not apply Alaska-specific regulations to those lands. That is a surprising conclusion. ANILCA repeatedly recognizes that Alaska is different—from its “unrivaled scenic and geological values,” to the “unique” situation of its “rural residents dependent on subsistence uses,” to “the need for development and use of Arctic resources with appropriate recognition and consideration given to the unique nature of the Arctic environment.” 16 U. S. C. §§3101(b), 3111(2), 3147(b)(5). ANILCA itself accordingly carves out numerous Alaska-specific exceptions to the Park Service’s general authority over federally managed preservation areas. For example, ANILCA requires the Secretary of the Interior to permit “the exercise of valid commercial fishing rights or privi-leges” within the National Wildlife Refuge System in Alaska, including the use of “campsites, cabins, motorized vehicles, and aircraft landings directly incident to the exercise of such rights or privileges,” with certain exceptions. 94Stat. 2393. ANILCA also requires the Secretary to “permit on the public lands appropriate use for subsistence purposes of snowmobiles, motorboats, and other means of surface transportation traditionally employed for such purposes by local residents, subject to reasonable regulation.” 16 U. S. C. §3121(b). And it provides that National Preserves “in Alaska shall be administered and managed as a unit of the National Park System in the same manner as a national park except as otherwise provided in this Act and except that the taking of fish and wildlife for sport purposes and subsistence uses, and trapping shall be allowed” pursuant to applicable law. §3201 (emphasis added). Many similar examples are woven throughout ANILCA. See, e.g., 94Stat. 2393 (Secretary must administer wildlife refuge “so as to not impede the passage of navigation and access by boat on the Yukon and Kuskokwim Rivers,” subject to reasonable regulation); id., at 2388 (Secretary must allow reindeer grazing uses in certain areas, including construction of necessary facilities); 16 U. S. C. §3203(a) (Alaska-specific rules for wilderness management apply “in recognition of the unique conditions in Alaska”); §3170(a) (Secretary must permit reasonable use of snowmachines, motorboats, and airplanes within conserva-tion system units “for travel to and from villages and homesites”). All those Alaska-specific provisions reflect the simple truth that Alaska is often the exception, not the rule. Yet the reading below would prevent the Park Service from recognizing Alaska’s unique conditions. Under that reading, the Park Service could regulate “non-public” lands in Alaska only through rules applicable outside Alaska as well. Thus, for example, if the Park Service elected to allow hovercraft during hunting season in Alaska—in a departure from its nationwide rule—the more relaxed regulation would apply only to the “public” land within the boundaries of the unit. Hovercraft would still be banned from the “non-public” land, even during hunting season. Whatever the reach of the Park Service’s authority under ANILCA, we cannot conclude that Section 103(c) adopted such a topsy-turvy approach. Moreover, it is clear that Section 103(c) draws a distinction between “public” and “non-public” lands within the boundaries of conservation system units in Alaska. See §3103(c) (“Only those lands within the boundaries of any conservation system unit which are public lands . . . shall be deemed to be included as a portion of such unit”); ibid. (No lands “conveyed to the State, to any Native Corporation, or to any private party shall be subject to the regulations applicable solely to public lands within such units”). And yet, according to the court below, if the Park Service wanted to differentiate between that “public” and “non-public” land in an Alaska-specific way, it would have to regulate the “non-public” land pursuant to rules applicable outside Alaska, and the “public” land pursuant to Alaska-specific provisions. Assuming the Park Service has authority over “non-public” land in Alaska (an issue we do not decide), that strikes us as an implausible reading of the statute. Looking at ANILCA both as a whole and with respect to Section 103(c), the Act contemplates the possibility that all the land within the boundaries of conservation system units in Alaska may be treated differently from federally managed preservation areas across the country, and that “non-public” lands within the boundaries of those units may be treated differently from “public” lands within the unit. Under the Ninth Circuit’s reading of Section 103(c), however, the former is not an option, and the latter would require contorted and counterintuitive measures. We therefore reject the interpretation of Section 103(c) adopted by the court below. That reading of the statute was the sole basis for the disposition of this case by the Court of Appeals. We accordingly vacate the judgment of that court and remand for further proceedings. We do not reach the remainder of the parties’ arguments. In particular, we do not decide whether the Nation River qualifies as “public land” for purposes of ANILCA. Sturgeon claims that it does not; the Park Service that it does. The parties’ arguments in this respect touch on vital issues of state sovereignty, on the one hand, and federal authority, on the other. We find that in this case those issues should be addressed by the lower courts in the first instance. Given this determination, we also do not decide whether the Park Service has authority under Section 100751(b) to regulate Sturgeon’s activities on the Nation River, even if the river is not “public” land, or whether—as Sturgeon argues—any such authority is limited by ANILCA. Fin-ally, we do not consider the Park Service’s alternative ar-gument that it has authority under ANILCA over both “public” and “non-public” lands within the boundaries of conservation system units in Alaska, to the extent a regulation is written to apply specifically to both types of land. We leave those arguments to the lower courts for consideration as necessary. The judgment of the Court of Appeals for the Ninth Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

579.US.2015_14-6166

Petitioner Taylor was indicted under the Hobbs Act on two counts of affecting commerce or attempting to do so through robbery for his participation in two home invasions targeting marijuana dealers. In both cases, Taylor and other gang members broke into the homes, confronted the residents, demanded the location of drugs and money, found neither, and left relatively empty handed. Taylor’s trial resulted in a hung jury. At his retrial, the Government urged the trial court to preclude Taylor from offering evidence that the drug dealers he targeted dealt only in locally-grown marijuana. The trial court excluded that evidence and Taylor was convicted on both counts. The Fourth Circuit affirmed, holding that, given the aggregate effect of drug dealing on interstate commerce, the Government needed only to prove that Taylor robbed or attempted to rob a drug dealer of drugs or drug proceeds to satisfy the commerce element. Held: 1. The prosecution in a Hobbs Act robbery case satisfies the Act’s commerce element if it shows that the defendant robbed or attempted to rob a drug dealer of drugs or drug proceeds. Pp. 4–9. (a) The language of the Hobbs Act is unmistakably broad and reaches any obstruction, delay, or other effect on commerce, 18 U. S. C. §1951(a), “over which the United States has jurisdiction,” §1951(b)(3). See United States v. Culbert, 435 U. S. 371 . Pp. 4–5. (b) Under its commerce power, this Court has held, Congress may regulate, among other things, activities that have a substantial aggregate effect on interstate commerce, see Wickard v. Filburn, 317 U. S. 111 . This includes “purely local activities that are part of an economic ‘class of activities’ that have a substantial effect on interstate commerce,” Gonzales v. Raich, 545 U. S. 1 , so long as those activities are economic in nature. See United States v. Morrison, 529 U. S. 598 . One such “class of activities” is the production, possession, and distribution of controlled substances. 545 U. S., at 22. Grafting the holding in Raich onto the Hobbs Act’s commerce element, it follows that a robber who affects even the intrastate sale of marijuana affects commerce over which the United States has jurisdiction. Pp. 5–6. (c) In arguing that Raich should be distinguished because the Controlled Substances Act lacks the Hobbs Act’s additional commerce element, Taylor confuses the standard of proof with the meaning of the element that must be proved. The meaning of the Hobbs Act’s commerce element is a question of law, which, Raich establishes, includes purely intrastate drug production and sale. Applying, without expanding, Raich’s interpretation of the scope of Congress’s Commerce Clause power, if the Government proves beyond a reasonable doubt that a robber targeted a marijuana dealer’s drugs or illegal proceeds, the Government has proved beyond a reasonable doubt that commerce over which the United States has jurisdiction was affected. Pp. 6–9. 2. Here, the Government met its burden by introducing evidence that Taylor’s gang intentionally targeted drug dealers to obtain drugs and drug proceeds. That evidence included information that the gang members targeted the victims because of their drug dealing activities, as well as explicit statements made during the course of the robberies that revealed their belief that drugs and money were present. Such proof is sufficient to meet the Hobbs Act’s commerce element. P. 9. 754 F. 3d 217, affirmed. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion.

The Hobbs Act makes it a crime for a person to affect commerce, or to attempt to do so, by robbery. 18 U. S. C. §1951(a). The Act defines “commerce” broadly as interstate commerce “and all other commerce over which the United States has jurisdiction.” §1951(b)(3). This case requires us to decide what the Government must prove to satisfy the Hobbs Act’s commerce element when a defendant commits a robbery that targets a marijuana dealer’s drugs or drug proceeds. The answer to this question is straightforward and dictated by our precedent. We held in Gonzales v. Raich, 545 U. S. 1 (2005) , that the Commerce Clause gives Congress authority to regulate the national market for marijuana, including the authority to proscribe the purely intrastate production, possession, and sale of this controlled substance. Because Congress may regulate these intrastate activities based on their aggregate effect on interstate commerce, it follows that Congress may also regulate intrastate drug theft. And since the Hobbs Act criminalizes robberies and attempted robberies that affect any commerce “over which the United States has jurisdiction,” §1951(b)(3), the prosecution in a Hobbs Act robbery case satisfies the Act’s commerce element if it shows that the defendant robbed or attempted to rob a drug dealer of drugs or drug proceeds. By targeting a drug dealer in this way, a robber necessarily affects or attempts to affect commerce over which the United States has jurisdiction. In this case, petitioner Anthony Taylor was convicted on two Hobbs Act counts based on proof that he attempted to rob marijuana dealers of their drugs and drug money. We hold that this evidence was sufficient to satisfy the Act’s commerce element. I Beginning as early as 2009, an outlaw gang called the “Southwest Goonz” committed a series of home invasion robberies targeting drug dealers in the area of Roanoke, Virginia. 754 F. 3d 217, 220 (CA4 2014). For obvious reasons, drug dealers are more likely than ordinary citizens to keep large quantities of cash and illegal drugs in their homes and are less likely to report robberies to the police. For participating in two such home invasions, Taylor was convicted of two counts of Hobbs Act robbery, in violation of §1951(a), and one count of using a firearm in furtherance of a crime of violence, in violation of §924(c). The first attempted drug robbery for which Taylor was convicted occurred in August 2009. Id., at 220. Taylor and others targeted the home of Josh Whorley, having obtained information that Whorley dealt “exotic and high grade” marijuana. Ibid. “The robbers expected to find both drugs and money” in Whorley’s home. Ibid. Taylor and the others broke into the home, searched it, and assaulted Whorley and his girlfriend. They demanded to be told the location of money and drugs but, not locating any, left with only jewelry, $40, two cell phones, and a marijuana cigarette. Ibid. The second attempted drug robbery occurred two months later in October 2009 at the home of William Lynch. Ibid. A source informed the leader of the gang that, on a prior occasion, the source had robbed Lynch of 20 pounds of marijuana in front of Lynch’s home. The gang also received information that Lynch continued to deal drugs. Taylor and others broke into Lynch’s home, held his wife and young children at gunpoint, assaulted his wife, and demanded to know the location of his drugs and money. Again largely unsuccessful, the robbers made off with only a cell phone. Id., at 221. For his participation in these two home invasions, Taylor was indicted under the Hobbs Act on two counts of affecting commerce or attempting to do so through robbery. App. 11a–13a. His first trial resulted in a hung jury. On retrial, at the urging of the Government, the District Court precluded Taylor from introducing evidence that the drug dealers he targeted might be dealing in only locally grown marijuana. Id., at 60a; see 754 F. 3d, at 221. During the second trial, Taylor twice moved for a judgment of acquittal on the ground that the prosecution had failed to meet its burden on the commerce element, Tr. 445–447, 532–533; see 754 F. 3d, at 221, but the District Court denied those motions, holding that the proof that Taylor attempted to rob drug dealers was sufficient as a matter of law to satisfy that element. Tr. 446, 532–533. The jury found Taylor guilty on both of the Hobbs Act counts and one of the firearms counts. App. 67a–69a. On appeal, Taylor challenged the sufficiency of the evidence to prove the commerce element of the Hobbs Act, but the Fourth Circuit affirmed. “Because drug dealing in the aggregate necessarily affects interstate commerce,” the court reasoned, “the government was simply required to prove that Taylor depleted or attempted to deplete the assets of such an operation.” 754 F. 3d, at 224. We granted certiorari to resolve a conflict in the Circuits regarding the demands of the Hobbs Act’s commerce element in cases involving the theft of drugs and drug proceeds from drug dealers. 576 U. S. ___ (2015). II A The Hobbs Act provides in relevant part as follows: “Whoever in any way or degree obstructs, delays, or affects commerce or the movement of any article or commodity in commerce, by robbery . . . or attemptsor conspires so to do . . . shall be fined under this title or imprisoned not more than twenty years, or both.” 18U. S. C. §1951(a). The Act then defines the term “commerce” to mean “commerce within the District of Columbia, or any Territory or Possession of the United States; all commerce between any point in a State, Territory, Possession, or the District of Columbia and any point outside thereof; all commerce between points within the same State through any place outside such State; and all other commerce over which the United States has jurisdiction.” §1951(b)(3). The language of the Hobbs Act is unmistakably broad. It reaches any obstruction, delay, or other effect on commerce, even if small, and the Act’s definition of commerce encompasses “all . . . commerce over which the United States has jurisdiction.” Ibid. We have noted the sweep of the Act in past cases. United States v. Culbert, 435 U. S. 371, 373 (1978) (“These words do not lend themselves to restrictive interpretation”); Stirone v. United States, 361 U. S. 212, 215 (1960) (The Hobbs Act “speaks in broad language, manifesting a purpose to use all the constitutional power Congress has to punish interference with interstate commerce by extortion, robbery or physical violence”). B To determine how far this commerce element extends—and what the Government must prove to meet it—we look to our Commerce Clause cases. We have said that there are three categories of activity that Congress may regulate under its commerce power: (1) “the use of the channels of interstate commerce”; (2) “the instrumentalities of interstate commerce, or persons or things in interstate commerce, even though the threat may come only from intrastate activities”; and (3) “those activities having a substantial relation to interstate commerce, . . . i. e., those activities that substantially affect interstate commerce.” United States v. Lopez, 514 U. S. 549 –559 (1995). We have held that activities in this third category—those that “substantially affect” commerce—may be regulated so long as they substantially affect interstate commerce in the aggregate, even if their individual impact on interstate commerce is minimal. See Wickard v. Filburn, 317 U. S. 111, 125 (1942) (“[E]ven if appellee’s activity be local and though it may not be regarded as commerce, it may still, whatever its nature, be reached by Congress if it exerts a substantial economic effect on interstate commerce”). While this final category is broad, “thus far in our Nation’s history our cases have upheld Commerce Clause regulation of intrastate activity only where that activity is economic in nature.” United States v. Morrison, 529 U. S. 598, 613 (2000) . In this case, the activity at issue, the sale of marijuana, is unquestionably an economic activity. It is, to be sure, a form of business that is illegal under federal law and the laws of most States. But there can be no question that marijuana trafficking is a moneymaking endeavor—and a potentially lucrative one at that. In Raich, the Court addressed Congress’s authority to regulate the marijuana market. The Court reaffirmed “Congress’ power to regulate purely local activities that are part of an economic ‘class of activities’ that have a substantial effect on interstate commerce.” 545 U. S., at 17. The production, possession, and distribution of controlled substances constitute a “class of activities” that in the aggregate substantially affect interstate commerce, and therefore, the Court held, Congress possesses the authority to regulate (and to criminalize) the production, possession, and distribution of controlled substances even when those activities occur entirely within the boundaries of a single State. Any other outcome, we warned, would leave a gaping enforcement hole in Congress’s regulatory scheme. Id., at 22. The case now before us requires no more than that we graft our holding in Raich onto the commerce element of the Hobbs Act. The Hobbs Act criminalizes robberies affecting “commerce over which the United States has jurisdiction.” §1951(b)(3). Under Raich, the market for marijuana, including its intrastate aspects, is “commerce over which the United States has jurisdiction.” It therefore follows as a simple matter of logic that a robberwho affects or attempts to affect even the intrastate saleof marijuana grown within the State affects or attemptsto affect commerce over which the United States has jurisdiction. C Rejecting this logic, Taylor takes the position that the robbery or attempted robbery of a drug dealer’s inventory violates the Hobbs Act only if the Government proves something more. This argument rests in part on the fact that Raich concerned the Controlled Substances Act (CSA), the criminal provisions of which lack a jurisdictional element. See 21 U. S. C. §§841(a), 844. The Hobbs Act, by contrast, contains such an element—namely, the conduct criminalized must affect or attempt to affect commerce in some way or degree. See 18 U. S. C. §1951(a). Therefore, Taylor reasons, the prosecution must prove beyond a reasonable doubt either (1) that the particular drugs in question originated or were destined for sale out of State or (2) that the particular drug dealer targeted in the robbery operated an interstate business. See Brief for Petitioner 25–27; Reply Brief 8. The Second and Seventh Circuits have adopted this same argument. See United States v. Needham, 604 F. 3d 673, 681 (CA2 2010); United States v. Peterson, 236 F. 3d 848, 855 (CA7 2001). This argument is flawed. It confuses the standard of proof with the meaning of the element that must be proved. There is no question that the Government in a Hobbs Act prosecution must prove beyond a reasonable doubt that the defendant engaged in conduct that satisfies the Act’s commerce element, but the meaning of that element is a question of law. And, as noted, Raich established that the purely intrastate production and sale of marijuana is commerce over which the Federal Government has jurisdiction. Therefore, if the Government proves beyond a reasonable doubt that a robber targeted a marijuana dealer’s drugs or illegal proceeds, the Government has proved beyond a reasonable doubt that commerce over which the United States has jurisdiction was affected. The only way to escape that conclusion would be to hold that the Hobbs Act does not exercise the full measure of Congress’s commerce power. But we reached the opposite conclusion more than 50 years ago, see Stirone, 361 U. S., at 215, and it is not easy to see how the expansive language of the Act could be interpreted in any other way. This conclusion does not make the commerce provision of the Hobbs Act superfluous. That statute, unlike the criminal provisions of the CSA, applies to forms of conduct that, even in the aggregate, may not substantially affect commerce. The Act’s commerce element ensures that applications of the Act do not exceed Congress’s authority. But in a case like this one, where the target of a robbery is a drug dealer, proof that the defendant’s conduct in and of itself affected or threatened commerce is not needed. All that is needed is proof that the defendant’s conduct fell within a category of conduct that, in the aggregate, had the requisite effect. D Contrary to the dissent, see post, at 10–12 (opinion of Thomas, J.), today’s holding merely applies—it in no way expands—Raich’s interpretation of the scope of Congress’s power under the Commerce Clause. The dissent resists the substantial-effects approach and the aggregation principle on which Raich is based, see post, at 11–12. But we have not been asked to reconsider Raich. So our decision in Raich controls the outcome here. As long as Congress may regulate the purely intrastate possession and sale of illegal drugs, Congress may criminalize the theft or attempted theft of those same drugs. We reiterate what this means. In order to obtain a conviction under the Hobbs Act for the robbery or attempted robbery of a drug dealer, the Government need not show that the drugs that a defendant stole or attempted to steal either traveled or were destined for transport across state lines. Rather, to satisfy the Act’s commerce element, it is enough that a defendant knowingly stole or attempted to steal drugs or drug proceeds, for, as a matter of law, the market for illegal drugs is “commerce over which the United States has jurisdiction.” And it makes no difference under our cases that any actual or threatened effect on commerce in a particular case is minimal. See Perez v. United States, 402 U. S. 146, 154 (1971) (“Where the class of activities is regulated and that class is within the reach of federal power, the courts have no power ‘to excise, as trivial, individual instances’ of the class” (emphasis deleted)). E In the present case, the Government met its burden by introducing evidence that Taylor’s gang intentionally targeted drug dealers to obtain drugs and drug proceeds. One of the victims had been robbed of substantial quantities of drugs at his residence in the past, and the other was thought to possess high-grade marijuana. The robbers also made explicit statements in the course of the robberies revealing that they believed that the victims possessed drugs and drug proceeds. Tr. 359 (asking Lynch “where the weed at”); id., at 93 (asking Whorley “where the money was at, where the weed was at”); id., at 212–213 (asking Whorley, “Where is your money and where is your weed at?”). Both robberies were committed with the express intent to obtain illegal drugs and the proceeds from the sale of illegal drugs. Such proof is sufficient to meet the commerce element of the Hobbs Act. Our holding today is limited to cases in which the defendant targets drug dealers for the purpose of stealing drugs or drug proceeds. We do not resolve what the Government must prove to establish Hobbs Act robbery where some other type of business or victim is targeted. See, e.g., Stirone, supra, at 215 (Government offered evidence that the defendant attempted to extort a concrete business that actually obtained supplies and materials from out of State). * * * The judgment of the Fourth Circuit is affirmed. It is so ordered.

577.US.2015_14-1146

Respondents, employees of petitioner Tyson Foods, work in the kill, cut, and retrim departments of a pork processing plant in Iowa. Respondents’ work requires them to wear protective gear, but the exact composition of the gear depends on the tasks a worker performs on a given day. Petitioner compensated some, but not all, employees for this donning and doffing, and did not record the time each employee spent on those activities. Respondents filed suit, alleging that the donning and doffing were integral and indispensable to their hazardous work and that petitioner’s policy not to pay for those activities denied them overtime compensation required by the Fair Labor Standards Act of 1938 (FLSA). Respondents also raised a claim under an Iowa wage law. They sought certification of their state claims as a class action under Federal Rule of Civil Procedure 23 and certification of their FLSA claims as a “collective action.” See 29 U. S. C. §216. Petitioner objected to certification of both classes, arguing that, because of the variance in protective gear each employee wore, the employees’ claims were not sufficiently similar to be resolved on a classwide basis. The District Court concluded that common questions, such as whether donning and doffing protective gear was compensable under the FLSA, were susceptible to classwide resolution even if not all of the workers wore the same gear. To recover for a violation of the FLSA’s overtime provision, the employees had to show that they each worked more than 40 hours a week, inclusive of the time spent donning and doffing. Because petitioner failed to keep records of this time, the employees primarily relied on a study performed by an industrial relations expert, Dr. Kenneth Mericle. Mericle conducted videotaped observations analyzing how long various donning and doffing activities took, and then averaged the time taken to produce an estimate of 18 minutes a day for the cut and retrim departments and 21.25 minutes for the kill department. These estimates were then added to the timesheets of each employee to ascertain which class members worked more than 40 hours a week and the value of classwide recovery. Petitioner argued that the varying amounts of time it took employees to don and doff different protective gear made reliance on Mericle’s sample improper, and that its use would lead to recovery for individuals who, in fact, had not worked the requisite 40 hours. The jury awarded the class about $2.9 million in unpaid wages. The award has not yet been disbursed to individual employees. The Eighth Circuit affirmed the judgment and the award. Held: The District Court did not err in certifying and maintaining the class. Pp. 8–17. (a) Before certifying a class under Rule 23(b)(3), a district court must find that “questions of law or fact common to class members predominate over any questions affecting only individual members.” The parties agree that the most significant question common to the class is whether donning and doffing protective gear is compensable under the FLSA. Petitioner claims, however, that individual inquiries into the time each worker spent donning and doffing predominate over this common question. Respondents argue that individual inquiries are unnecessary because it can be assumed each employee donned and doffed for the same average time observed in Mericle’s sample. Whether and when statistical evidence such as Mericle’s sample can be used to establish classwide liability depends on the purpose for which the evidence is being introduced and on “the elements of the underlying cause of action,” Erica P. John Fund, Inc. v. Halliburton Co., 563 U.S. 804, 809. Because a representative sample may be the only feasible way to establish liability, it cannot be deemed improper merely because the claim is brought on behalf of a class. Respondents can show that Mericle’s sample is a permissible means of establishing hours worked in a class action by showing that each class member could have relied on that sample to establish liability had each brought an individual action. Anderson v. Mt. Clemens Pottery Co., 328 U.S. 680, shows why Mericle’s sample was permissible in the circumstances of this case. There, where an employer violated its statutory duty to keep proper records, the Court concluded the employees could meet their burden by proving that they in fact “performed work for which [they were] improperly compensated and . . . produc[ing] sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference.” Id., at 687. Here, similarly, respondents sought to introduce a representative sample to fill an evidentiary gap created by the employer’s failure to keep adequate records. Had the employees proceeded with individual lawsuits, each employee likely would have had to introduce Mericle’s study to prove the hours he or she worked. The representative evidence was a permissible means of showing individual hours worked. This holding is in accord with Wal-Mart Stores, Inc. v. Dukes, 564 U.S. 338, where the underlying question was, as here, whether the sample at issue could have been used to establish liability in an individual action. There, the employees were not similarly situated, so none of them could have prevailed in an individual suit by relying on depositions detailing the ways in which other employees were discriminated against by their particular store managers. In contrast, the employees here, who worked in the same facility, did similar work, and were paid under the same policy, could have introduced Mericle’s study in a series of individual suits. This case presents no occasion for adoption of broad and categorical rules governing the use of representative and statistical evidence in class actions. Rather, the ability to use a representative sample to establish classwide liability will depend on the purpose for which the sample is being introduced and on the underlying cause of action. In FLSA actions, inferring the hours an employee has worked from a study such as Mericle’s has been permitted by the Court so long as the study is otherwise admissible. Mt. Clemens, supra, at 687. Pp. 8–15. (b) Petitioner contends that respondents are required to demonstrate that uninjured class members will not recover damages here. That question is not yet fairly presented by this case, because the damages award has not yet been disbursed and the record does not indicate how it will be disbursed. Petitioner may raise a challenge to the allocation method when the case returns to the District Court for disbursal of the award. Pp. 15–17. 765 F.3d 791, affirmed and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Roberts, C. J., filed a concurring opinion, in which Alito, J., joined as to Part II. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

Following a jury trial, a class of employees recovered $2.9 million in compensatory damages from their employer for a violation of the Fair Labor Standards Act of 1938 (FLSA), 52Stat. 1060, as amended, 29 U. S. C. §201 et seq. The employees’ primary grievance was that they did not receive statutorily mandated overtime pay for time spent donning and doffing protective equipment. The employer seeks to reverse the judgment. It makes two arguments. Both relate to whether it was proper to permit the employees to pursue their claims as a class. First, the employer argues the class should not have been certified because the primary method of proving injury assumed each employee spent the same time donning and doffing protective gear, even though differences in the composition of that gear may have meant that, in fact, employees took different amounts of time to don and doff. Second, the employer argues certification was improper because the damages awarded to the class may be distributed to some persons who did not work any uncompensated overtime. The Court of Appeals for the Eighth Circuit concluded there was no error in the District Court’s decision to cer-tify and maintain the class. This Court granted certiorari. 576 U. S. ___ (2015). I Respondents are employees at petitioner Tyson Foods’ pork processing plant in Storm Lake, Iowa. They work in the plant’s kill, cut, and retrim departments, where hogs are slaughtered, trimmed, and prepared for shipment. Grueling and dangerous, the work requires employees to wear certain protective gear. The exact composition of the gear depends on the tasks a worker performs on a given day. Until 1998, employees at the plant were paid under a system called “gang-time.” This compensated them only for time spent at their workstations, not for the time required to put on and take off their protective gear. In response to a federal-court injunction, and a Department of Labor suit to enforce that injunction, Tyson in 1998 began to pay all its employees for an additional four minutes a day for what it called “K-code time.” The4-minute period was the amount of time Tyson estimated employees needed to don and doff their gear. In 2007, Tyson stopped paying K-code time uniformly to all employees. Instead, it compensated some employees for between four and eight minutes but paid others nothing beyond their gang-time wages. At no point did Tyson record the time each employee spent donning and doffing. Unsatisfied by these changes, respondents filed suit in the United States District Court for the Northern District of Iowa, alleging violations of the FLSA. The FLSA requires that a covered employee who works more than 40 hours a week receive compensation for excess time worked “at a rate not less than one and one-half times the regular rate at which he is employed.” 29 U. S. C. §207(a). In 1947, nine years after the FLSA was first enacted, Congress passed the Portal-to-Portal Act, which clarified that compensable work does not include time spent walking to and from the employee’s workstation or other “preliminary or postliminary activities.” §254(d). The FLSA, however, still requires employers to pay employees for activities “integral and indispensable” to their regular work, even if those activities do not occur at the employee’s workstation. Steiner v. Mitchell, 350 U. S. 247, 249, 255 (1956) . The FLSA also requires an employer to “make, keep, and preserve . . . records of the persons employed by him and of the wages, hours, and other conditions and practices of employment.” §211(c). In their complaint, respondents alleged that donning and doffing protective gear were integral and indispensable to their hazardous work and that petitioner’s policy not to pay for those activities denied them overtime compensation required by the FLSA. Respondents also raised a claim under the Iowa Wage Payment Collection Law. This statute provides for recovery under state law when an employer fails to pay its employees “all wages due,” which includes FLSA-mandated overtime. Iowa Code §91A.3 (2013); cf. Anthony v. State, 632 N. W. 2d 897, 901–902 (Iowa 2001). Respondents sought certification of their Iowa law claims as a class action under Rule 23 of the Federal Rules of Civil Procedure. Rule 23 permits one or more individ-uals to sue as “representative parties on behalf of all mem-bers” of a class if certain preconditions are met. Fed. Rule Civ. Proc. 23(a). Respondents also sought certification of their federal claims as a “collective action” under 29 U. S. C. §216. Section 216 is a provision of the FLSA that permits employees to sue on behalf of “themselves and other employees similarly situated.” §216(b). Tyson objected to the certification of both classes on the same ground. It contended that, because of the variance in protective gear each employee wore, the employees’ claims were not sufficiently similar to be resolved on a classwide basis. The District Court rejected that position. It concluded there were common questions susceptible to classwide resolution, such as “whether the donning and doffing of [protective gear] is considered work under the FLSA, whether such work is integral and [in]dispensable, and whether any compensable work is de minim[i]s.” 564 F. Supp. 2d 870, 899 (ND Iowa 2008). The District Court acknowledged that the workers did not all wear the same protective gear, but found that “when the putative plaintiffs are limited to those that are paid via a gang time system, there are far more factual similarities than dissimilarities.” Id., at 899–900. As a result, the District Court certified the following classes: “All current and former employees of Tyson’s Storm Lake, Iowa, processing facility who have been employed at any time from February 7, 2004 [in the case of the FLSA collective action and February 7, 2005, in the case of the state-law class action], to the present, and who are or were paid under a ‘gang time’ compensation system in the Kill, Cut, or Retrim departments.” Id., at 901. The only difference in definition between the classes was the date at which the class period began. The size of the class certified under Rule 23, however, was larger than that certified under §216. This is because, while a class under Rule 23 includes all unnamed members who fall within the class definition, the “sole consequence of conditional certification [under §216] is the sending of court-approved written notice to employees . . . who in turn become parties to a collective action only by filing written consent with the court.” Genesis HealthCare Corp. v. Symczyk, 569 U. S. ___, ___ (2013) (slip op., at 8). A total of 444 employees joined the collective action, while the Rule 23 class contained 3,344 members. The case proceeded to trial before a jury. The parties stipulated that the employees were entitled to be paid for donning and doffing of certain equipment worn to protect from knife cuts. The jury was left to determine whether the time spent donning and doffing other protective equipment was compensable; whether Tyson was required to pay for donning and doffing during meal breaks; and the total amount of time spent on work that was not compensated under Tyson’s gang-time system. Since the employees’ claims relate only to overtime, each employee had to show he or she worked more than 40 hours a week, inclusive of time spent donning and doffing, in order to recover. As a result of Tyson’s failure to keep records of donning and doffing time, however, the employees were forced to rely on what the parties describe as “representative evidence.” This evidence included employee testimony, video recordings of donning and doffing atthe plant, and, most important, a study performed by an industrial relations expert, Dr. Kenneth Mericle. Mericle conducted 744 videotaped observations and analyzed how long various donning and doffing activities took. He then averaged the time taken in the observations to produce an estimate of 18 minutes a day for the cut and retrim departments and 21.25 minutes for the kill department. Although it had not kept records for time spent donning and doffing, Tyson had information regarding each employee’s gang-time and K-code time. Using this data, the employees’ other expert, Dr. Liesl Fox, was able to estimate the amount of uncompensated work each employee did by adding Mericle’s estimated average donning and doffing time to the gang-time each employee worked and then subtracting any K-code time. For example, if an employee in the kill department had worked 39.125 hours of gang-time in a 6-day workweek and had been paid an hour of K-code time, the estimated number of compensable hours the employee worked would be: 39.125 (individual number of gang-time hours worked) + 2.125 (the average donning and doffing hours for a 6-day week, based on Mericle’s estimated average of 21.25 minutes a day) – 1 (K-code hours) = 40.25. That would mean the employee was being undercompensated by a quarter of an hour of overtime a week, in violation of the FLSA. On the other hand, if the employee’s records showed only 38 hours of gang-time and an hour of K-code time, the calculation would be: 38 + 2.125 – 1 = 39.125. Having worked less than 40 hours, that employee would not be entitled to overtime pay and would not have proved an FLSA violation. Using this methodology, Fox stated that 212 employees did not meet the 40-hour threshold and could not recover. The remaining class members, Fox maintained, had potentially been undercompensated to some degree. Respondents proposed to bifurcate proceedings. They requested that, first, a trial be conducted on the questions whether time spent in donning and doffing was compensable work under the FLSA and how long those activities took to perform on average; and, second, that Fox’s methodology be used to determine which employees suffered an FLSA violation and how much each was entitled to recover. Petitioner insisted upon a single proceeding in whichdamages would be calculated in the aggregate and by the jury. The District Court submitted both issues of liability and damages to the jury. Petitioner did not move for a hearing regarding the statistical validity of respondents’ studies under Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U. S. 579 (1993) , nor did it attempt to discredit the evidence with testimony from a rebuttal expert. Instead, as it had done in its opposition to class certification, petitioner argued to the jury that the varying amounts of time it took employees to don and doff different protective equipment made the lawsuit too speculative for classwide recovery. Petitioner also argued that Mericle’s study overstated the average donning and doffing time. The jury was in-structed that nontestifying members of the class could only recover if the evidence established they “suffered the same harm as a result of the same unlawful decision or policy.” App. 471–472. Fox’s calculations supported an aggregate award of approximately $6.7 million in unpaid wages. The jury returned a special verdict finding that time spent in donning and doffing protective gear at the beginning and end of the day was compensable work but that time during meal breaks was not. The jury more than halved the damages recommended by Fox. It awarded the class about $2.9 million in unpaid wages. That damages award has not yet been disbursed to the individual employees. Tyson moved to set aside the jury verdict, arguing, among other things, that, in light of the variation in donning and doffing time, the classes should not have been certified. The District Court denied Tyson’s motion, and the Court of Appeals for the Eighth Circuit affirmed the judgment and the award. The Court of Appeals recognized that a verdict for the employees “require[d] inference” from their representative proof, but it held that “this inference is allowable under Anderson v. Mt. Clemens Pottery Co., 328 U. S. 680 –688 (1946).” 765 F. 3d 791, 797 (2014). The Court of Appeals rejected petitioner’s challenge to the sufficiency of the evidence for similar reasons, holding that, under the facts of this case, the jury could have drawn “a ‘reasonable inference’ of class-wide liability.” Id., at 799 (quoting Anderson v. Mt. Clemens Pottery Co., 328 U. S. 680, 687 (1946) ). Judge Beam dissented, stating that, in his view, the class should not have been certified. For the reasons that follow, this Court now affirms. II Petitioner challenges the class certification of the state- law claims and the certification of the FLSA collective action. The parties do not dispute that the standard for certifying a collective action under the FLSA is no more stringent than the standard for certifying a class under the Federal Rules of Civil Procedure. This opinion assumes, without deciding, that this is correct. For purposes of this case then, if certification of respondents’ class action under the Federal Rules was proper, certification of the collective action was proper as well. Furthermore, as noted above, Iowa’s Wage Payment Collection Law was used in this litigation as a state-law mechanism for recovery of FLSA-mandated overtime pay. The parties do not dispute that, in order to prove a violation of the Iowa statute, the employees had to do no more than demonstrate a violation of the FLSA. In this opinion, then, no distinction is made between the requirements for the class action raising the state-law claims and the collective action raising the federal claims. A Federal Rule of Civil Procedure 23(b)(3) requires that, before a class is certified under that subsection, a district court must find that “questions of law or fact common to class members predominate over any questions affecting only individual members.” The “predominance inquiry tests whether proposed classes are sufficiently cohesive to warrant adjudication by representation.” Amchem Products, Inc. v. Windsor, 521 U. S. 591, 623 (1997) . This calls upon courts to give careful scrutiny to the relation between common and individual questions in a case. An individual question is one where “members of a proposed class will need to present evidence that varies from member to member,” while a common question is one where “the same evidence will suffice for each member to make a prima facie showing [or] the issue is susceptible to generalized, class-wide proof.” 2 W. Rubenstein, Newberg on Class Actions §4:50, pp. 196–197 (5th ed. 2012) (internal quotation marks omitted). The predominance inquiry “asks whether the common, aggregation-enabling, issues in the case are more prevalent or important than the non-common, aggregation-defeating, individual issues.” Id., §4:49, at 195–196. When “one or more of the central issues in the action are common to the class and can be said to predominate, the action may be considered proper under Rule 23(b)(3) even though other important matters will have to be tried separately, such as damages or some affirmative defenses peculiar to some individual class members.” 7AA C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §1778, pp. 123–124 (3d ed. 2005) (footnotes omitted). Here, the parties do not dispute that there are important questions common to all class members, the most significant of which is whether time spent donning and doffing the required protective gear is compensable work under the FLSA. Cf. IBP, Inc. v. Alvarez, 546 U. S. 21 (2005) (holding that time spent walking between the locker room and the production area after donning protective gear is compensable work under the FLSA). To be entitled to recovery, however, each employee must prove that the amount of time spent donning and doffing, when added to his or her regular hours, amounted to more than 40 hours in a given week. Petitioner argues that these necessarily person-specific inquiries into individual work time predominate over the common questions raised by respondents’ claims, making class certification improper. Respondents counter that these individual inquiries are unnecessary because it can be assumed each employee donned and doffed for the same average time observed in Mericle’s sample. Whether this inference is permissible becomes the central dispute in this case. Petitioner contends that Mericle’s study manufactures predominance by assuming away the very differences that make the case inappropriate for classwide resolution. Reliance on a representative sample, petitioner argues, absolves each employee of the responsibility to prove personal injury, and thus deprives petitioner of any ability to litigate its defenses to individual claims. Calling this unfair, petitioner and various of its amici maintain that the Court should announce a broad rule against the use in class actions of what the parties call representative evidence. A categorical exclusion of that sort, however, would make little sense. A representative or statistical sample, like all evidence, is a means to establish or defend against liability. Its permissibility turns not on the form a proceeding takes—be it a class or individual action—but on the degree to which the evidence is reliable in proving or disproving the elements of the relevant cause of action. See Fed. Rules Evid. 401, 403, and 702. It follows that the Court would reach too far were it to establish general rules governing the use of statistical evidence, or so-called representative evidence, in all class- action cases. Evidence of this type is used in various substantive realms of the law. Brief for Complex Litigation Law Professors as Amici Curiae 5–9; Brief for Economists et al. as Amici Curiae 8–10. Whether and when statistical evidence can be used to establish classwide liability will depend on the purpose for which the evidence is being introduced and on “the elements of the underlying cause of action,” Erica P. John Fund, Inc. v. Halliburton Co., 563 U. S. 804, 809 (2011) . In many cases, a representative sample is “the only practicable means to collect and present relevant data” establishing a defendant’s liability. Manual of Complex Litigation §11.493, p. 102 (4th ed. 2004). In a case where representative evidence is relevant in proving a plaintiff’s individual claim, that evidence cannot be deemed im-proper merely because the claim is brought on behalf of a class. To so hold would ignore the Rules Enabling Act’s pellucid instruction that use of the class device cannot “abridge . . . any substantive right.” 28 U. S. C. §2072(b). One way for respondents to show, then, that the sample relied upon here is a permissible method of proving classwide liability is by showing that each class member could have relied on that sample to establish liability if he or she had brought an individual action. If the sample could have sustained a reasonable jury finding as to hours worked in each employee’s individual action, that sample is a permissible means of establishing the employees’ hours worked in a class action. This Court’s decision in Anderson v. Mt. Clemens explains why Mericle’s sample was permissible in the circumstances of this case. In Mt. Clemens, 7 employees and their union, seeking to represent over 300 others, brought a collective action against their employer for failing to compensate them for time spent walking to and from their workstations. The variance in walking time among workers was alleged to be upwards of 10 minutes a day, which is roughly consistent with the variances in donning and doffing times here. 328 U. S., at 685. The Court in Mt. Clemens held that when employers violate their statutory duty to keep proper records, and employees thereby have no way to establish the time spent doing uncompensated work, the “remedial nature of [the FLSA] and the great public policy which it embodies . . . militate against making” the burden of proving uncompensated work “an impossible hurdle for the employee.” Id., at 687; see also Hoffmann-La Roche Inc. v. Sperling, 493 U. S. 165, 173 (1989) (“The broad remedial goal of the statute should be enforced to the full extent of its terms”). Instead of punishing “the employee by denying him any recovery on the ground that he is unable to prove the precise extent of uncompensated work,” the Court held “an employee has carried out his burden if he proves that he has in fact performed work for which he was improperly compensated and if he produces sufficient evidence to show the amount and extent of that work as a matter of just and reasonable inference.” 328 U. S., at 687. Under these circumstances, “[t]he burden then shifts to the employer to come forward with evidence of the precise amount of work performed or with evidence to negative the reasonableness of the inference to be drawn from the employee’s evidence.” Id., at 687–688. In this suit, as in Mt. Clemens, respondents sought to introduce a representative sample to fill an evidentiary gap created by the employer’s failure to keep adequate records. If the employees had proceeded with 3,344 individual lawsuits, each employee likely would have had to introduce Mericle’s study to prove the hours he or she worked. Rather than absolving the employees from proving individual injury, the representative evidence here was a permissible means of making that very showing. Reliance on Mericle’s study did not deprive petitioner of its ability to litigate individual defenses. Since there were no alternative means for the employees to establish their hours worked, petitioner’s primary defense was to show that Mericle’s study was unrepresentative or inaccurate. That defense is itself common to the claims made by all class members. Respondents’ “failure of proof on th[is] common question” likely would have ended “the litigation and thus [would not have] cause[d] individual questions . . . to overwhelm questions common to the class.” Amgen Inc. v. Connecticut Retirement Plans and Trust Funds, 568 U. S. ___, ___ (2013) (slip op., at 11). When, as here, “the concern about the proposed class is not that it exhibits some fatal dissimilarity but, rather, a fatal similarity—[an alleged] failure of proof as to an element of the plaintiffs’ cause of action—courts should engage that question as a matter of summary judgment, not class certification.” Nagareda, Class Certification in the Age of Aggregate Proof, 84 N. Y. U. L. Rev. 97, 107 (2009). Petitioner’s reliance on Wal-Mart Stores, Inc. v. Dukes, 564 U. S. 338 (2011) , is misplaced. Wal-Mart does not stand for the broad proposition that a representative sample is an impermissible means of establishing classwide liability. Wal-Mart involved a nationwide Title VII class of over 11∕2 million employees. In reversing class certification, this Court did not reach Rule 23(b)(3)’s predominance prong, holding instead that the class failed to meet even Rule 23(a)’s more basic requirement that class members share a common question of fact or law. The plaintiffs in Wal-Mart did not provide significant proof of a common policy of discrimination to which each employee was subject. “The only corporate policy that the plaintiffs’ evidence convincingly establishe[d was] Wal-Mart’s ‘policy’ of allowing discretion by local supervisors over employment matters”; and even then, the plaintiffs could not identify “a common mode of exercising discretion that pervade[d] the entire company.” Id., at 355–356 (emphasis deleted). The plaintiffs in Wal-Mart proposed to use representative evidence as a means of overcoming this absence of a common policy. Under their proposed methodology, a “sample set of the class members would be selected, as to whom liability for sex discrimination and the backpay owing as a result would be determined in depositions supervised by a master.” Id., at 367. The aggregate damages award was to be derived by taking the “percentage of claims determined to be valid” from this sample and applying it to the rest of the class, and then multiplying the “number of (presumptively) valid claims” by “the average backpay award in the sample set.” Ibid. The Court held that this “Trial By Formula” was contrary to the Rules Enabling Act because it “ ‘enlarge[d]’ ” the class members’ “ ‘substantive right[s]’ ” and deprived defendants of their right to litigate statutory defenses to individual claims. Ibid. The Court’s holding in the instant case is in accord with Wal-Mart. The underlying question in Wal-Mart, as here, was whether the sample at issue could have been used to establish liability in an individual action. Since the Court held that the employees were not similarly situated, none of them could have prevailed in an individual suit by relying on depositions detailing the ways in which other employees were discriminated against by their particular store managers. By extension, if the employees had brought 11∕2 million individual suits, there would be little or no role for representative evidence. Permitting the use of that sample in a class action, therefore, would have violated the Rules Enabling Act by giving plaintiffs and defendants different rights in a class proceeding than they could have asserted in an individual action. In contrast, the study here could have been sufficient to sustain a jury finding as to hours worked if it were introduced in each employee’s individual action. While the experiences of the employees in Wal-Mart bore little relationship to one another, in this case each employee worked in the same facility, did similar work, and was paid under the same policy. As Mt. Clemens confirms, under these circumstances the experiences of a subset of employees can be probative as to the experiences of all of them. This is not to say that all inferences drawn from representative evidence in an FLSA case are “just and reason-able.” Mt. Clemens, 328 U. S., at 687. Representativeevidence that is statistically inadequate or based on implausible assumptions could not lead to a fair or accurate estimate of the uncompensated hours an employee has worked. Petitioner, however, did not raise a challenge to respondents’ experts’ methodology under Daubert; and, as a result, there is no basis in the record to conclude it was legal error to admit that evidence. Once a district court finds evidence to be admissible, its persuasiveness is, in general, a matter for the jury. Reasonable minds may differ as to whether the average time Mericle calculated is probative as to the time actually worked by each employee. Resolving that question, however, is the near-exclusive province of the jury. The District Court could have denied class certification on this ground only if it concluded that no reasonable juror could have believed that the employees spent roughly equal time donning and doffing. Cf. Anderson v. Liberty Lobby, Inc., 477 U. S. 242 –252 (1986). The District Court made no such finding, and the record here provides no basis for this Court to second-guess that conclusion. The Court reiterates that, while petitioner, respondents, or their respective amici may urge adoption of broad and categorical rules governing the use of representative and statistical evidence in class actions, this case provides no occasion to do so. Whether a representative sample may be used to establish classwide liability will depend on the purpose for which the sample is being introduced and on the underlying cause of action. In FLSA actions, inferring the hours an employee has worked from a study such as Mericle’s has been permitted by the Court so long as the study is otherwise admissible. Mt. Clemens, supra, at 687; see also Fed. Rules Evid. 402 and 702. The fairness and utility of statistical methods in contexts other than those presented here will depend on facts and circumstances particular to those cases. B In its petition for certiorari petitioner framed its second question presented as whether a class may be certified if it contains “members who were not injured and have no legal right to any damages.” Pet. for Cert. i. In its merits brief, however, petitioner reframes its argument. It now concedes that “[t]he fact that federal courts lack authority to compensate persons who cannot prove injury does not mean that a class action (or collective action) can never be certified in the absence of proof that all class members were injured.” Brief for Petitioner 49. In light of petitioner’s abandonment of its argument from the petition, the Court need not, and does not, address it. Petitioner’s new argument is that, “where class plaintiffs cannot offer” proof that all class members are injured, “they must demonstrate instead that there is some mechanism to identify the uninjured class members prior to judgment and ensure that uninjured members (1) do not contribute to the size of any damage award and (2) cannot recover such damages.” Ibid. Petitioner contends that respondents have not demonstrated any mechanism for ensuring that uninjured class members do not recover damages here. Petitioner’s new argument is predicated on the assumption that the damages award cannot be apportioned so that only those class members who suffered an FLSA violation recover. According to petitioner, because Fox’s mechanism for determining who had worked over 40 hours depended on Mericle’s estimate of donning and doffing time, and because the jury must have rejected Mericle’s estimate when it reduced the damages award by more than half, it will not be possible to know which workers are entitled to share in the award. As petitioner and its amici stress, the question whether uninjured class members may recover is one of great importance. See, e.g., Brief for Consumer Data Industry Association as Amicus Curiae. It is not, however, a question yet fairly presented by this case, because the damages award has not yet been disbursed, nor does the record indicate how it will be disbursed. Respondents allege there remain ways of distributing the award to only those individuals who worked more than 40 hours. For example, by working backwards from the damages award, and assuming each employee donned and doffed for an identical amount of time (an assumption that follows from the jury’s finding that the employees suffered equivalent harm under the policy), it may be possible to calculate the average donning and doffing time the jury necessarily must have found, and then apply this figure to each employee’s known gang-time hours to determine which employees worked more than 40 hours. Whether that or some other methodology will be successful in identifying uninjured class members is a question that, on this record, is premature. Petitioner may raise a challenge to the proposed method of allocation when the case returns to the District Court for disbursal of the award. Finally, it bears emphasis that this problem appears to be one of petitioner’s own making. Respondents proposed bifurcating between the liability and damages phases of this proceeding for the precise reason that it may be difficult to remove uninjured individuals from the class after an award is rendered. It was petitioner who argued against that option and now seeks to profit from the difficulty it caused. Whether, in light of the foregoing, any error should be deemed invited, is a question for the District Court to address in the first instance. * * * The judgment of the Court of Appeals for the Eighth Circuit is affirmed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

578.US.2015_15-290

The Clean Water Act regulates “the discharge of any pollutant” into “the waters of the United States.” 33 U. S. C. §§1311(a), 1362(7), (12). When property contains such waters, landowners who discharge pollutants without a permit from the Army Corps of Engineers risk substantial criminal and civil penalties, §§1319(c), (d), while those who do apply for a permit face a process that is often arduous, expensive, and long. It can be difficult to determine in the first place, however, whether “waters of the United States” are present. During the time period relevant to this case, for example, the Corps defined that term to include all wetlands, the “use, degradation or destruction of which could affect interstate or foreign commerce.” 33 CFR §328.3(a)(3). Because of that difficulty, the Corps allows property owners to obtain a standalone “jurisdictional determination” (JD) specifying whether a particular property contains “waters of the United States.” §331.2. A JD may be either “preliminary,” advising a property owner that such waters “may” be present, or “approved,” definitively “stating the presence or absence” of such waters. Ibid. An “approved” JD is considered an administratively appealable “final agency action,” §§320.1(a)(6), 331.2, and is binding for five years on both the Corps and the Environmental Protection Agency, 33 CFR pt. 331, App. C; EPA, Memorandum of Agreement: Exemptions Under Section 404(F) of the Clean Water Act §VI–A. Respondents, three companies engaged in mining peat, sought a permit from the Corps to discharge material onto wetlands located on property that respondents own and hope to mine. In connection with the permitting process, respondents obtained an approved JD from the Corps stating that the property contained “waters of the United States” because its wetlands had a “significant nexus” to the Red River of the North, located some 120 miles away. After exhausting administrative remedies, respondents sought review of the approved JD in Federal District Court under the Administrative Procedure Act (APA), but the District Court dismissed for want of jurisdiction, holding that the revised JD was not a “final agency action for which there is no other adequate remedy in a court,” 5 U. S. C. §704. The Eighth Circuit reversed. Held: The Corps’ approved JD is a final agency action judicially reviewable under the APA. Pp. 5–10. (a) In general, two conditions must be satisfied for an agency action to be “final” under the APA: “First, the action must mark the consummation of the agency’s decisionmaking process,” and “second, the action must be one by which rights or obligations have been determined, or from which legal consequences will flow.” Bennett v. Spear, 520 U. S. 154 –178. Pp. 5–8. (1) An approved JD satisfies Bennett’s first condition. It clearly “mark[s] the consummation” of the Corps’ decisionmaking on the question whether a particular property does or does not contain “waters of the United States.” It is issued after extensive factfinding by the Corps regarding the physical and hydrological characteristics of the property, see U. S. Army Corps of Engineers, Jurisdictional Determination Form Instructional Guidebook 47–60, and typically remains valid for a period of five years, see 33 CFR pt. 331, App. C. The Corps itself describes approved JDs as “final agency action.” Id. §320.1(a)(6). Pp. 5–6. (2) The definitive nature of approved JDs also gives rise to “direct and appreciable legal consequences,” thereby satisfying Bennett’s second condition as well. 520 U. S., at 178. A “negative” JD—i.e., an approved JD stating that property does not contain jurisdictional waters—creates a five-year safe harbor from civil enforcement proceedings brought by the Government and limits the potential liability a property owner faces for violating the Clean Water Act. See 33 U. S. C. §§1319, 1365(a). Each of those effects is a legal consequence. It follows that an “affirmative” JD, like the one issued here, also has legal consequences: It deprives property owners of the five-year safe harbor that “negative” JDs afford. This conclusion tracks the “pragmatic” approach the Court has long taken to finality. Abbott Laboratories v. Gardner, 387 U. S. 136 . Pp. 6–8. (b) A “final” agency action is reviewable under the APA only if there are no adequate alternatives to APA review in court. The Corps contends that respondents have two such alternatives: They may proceed without a permit and argue in a Government enforcement action that a permit was not required, or they may complete the permit process and then seek judicial review, which, the Corps suggests, is what Congress envisioned. Neither alternative is adequate. Parties need not await enforcement proceedings before challenging final agency action where such proceedings carry the risk of “serious criminal and civil penalties.” Abbott, 387 U. S., at 153. And the permitting process is not only costly and lengthy, but also irrelevant to the finality of the approved JD and its suitability for judicial review. Furthermore, because the Clean Water Act makes no reference to standalone jurisdictional determinations, there is little basis for inferring anything from it concerning their reviewability. Given “the APA’s presumption of reviewability for all final agency action,” Sackett v. EPA, 566 U. S. ___, ___, “[t]he mere fact” that permitting decisions are reviewable is insufficient to imply “exclusion as to other[ ]” agency actions, such as approved JDs, Abbott, 387 U. S., at 141. Pp. 8–10. 782 F. 3d 994, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Thomas, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Kennedy, J., filed a concurring opinion, in which Thomas and Alito, JJ., joined. Kagan, J., filed a concurring opinion. Ginsburg, J., filed an opinion concurring in part and concurring in the judgment.

The Clean Water Act regulates the discharge of pollutants into “the waters of the United States.” 33 U. S. C. §§1311(a), 1362(7), (12). Because it can be difficult to determine whether a particular parcel of property contains such waters, the U. S. Army Corps of Engineers will issue to property owners an “approved jurisdictional determination” stating the agency’s definitive view on that matter. See 33 CFR §331.2 and pt. 331, App. C (2015). The question presented is whether that determination is final agency action judicially reviewable under the Administrative Procedure Act, 5 U. S. C. §704. I A The Clean Water Act prohibits “the discharge of any pollutant” without a permit into “navigable waters,” which it defines, in turn, as “the waters of the United States.” 33 U. S. C. §§1311(a), 1362(7), (12). During the time period relevant to this case, the U. S. Army Corps of Engineers defined the waters of the United States to include land areas occasionally or regularly saturated with water—such as “mudflats, sandflats, wetlands, sloughs, prairie potholes, wet meadows, [and] playa lakes”—the “use, degradation or destruction of which could affect interstate or foreign commerce.” 33 CFR §328.3(a)(3) (2012). The Corps has applied that definition to assert jurisdiction over “270-to-300 million acres of swampy lands in the United States—including half of Alaska and an area the size of California in the lower 48 States.” Rapanos v. United States, 547 U. S. 715, 722 (2006) (plurality opinion).[1] It is often difficult to determine whether a particular piece of property contains waters of the United States, but there are important consequences if it does. The Clean Water Act imposes substantial criminal and civil penalties for discharging any pollutant into waters covered by the Act without a permit from the Corps. See 33 U. S. C. §§1311(a), 1319(c), (d), 1344(a). The costs of obtaining such a permit are significant. For a specialized “individ-ual” permit of the sort at issue in this case, for example, one study found that the average applicant “spends 788 days and $271,596 in completing the process,” without “counting costs of mitigation or design changes.” Rapanos, 547 U. S., at 721. Even more readily available “general” permits took applicants, on average, 313 days and $28,915 to complete. Ibid. See generally 33 CFR §323.2(h) (limiting “general” permits to activities that “cause only minimal individual and cumulative environmental impacts”). The Corps specifies whether particular property contains “waters of the United States” by issuing “jurisdictional determinations” (JDs) on a case-by-case basis. §331.2. JDs come in two varieties: “preliminary” and “approved.” Ibid. While preliminary JDs merely advise a property owner “that there may be waters of the United States on a parcel,” approved JDs definitively “stat[e] the presence or absence” of such waters. Ibid. (emphasis added). Unlike preliminary JDs, approved JDs can be administratively appealed and are defined by regulation to “constitute a Corps final agency action.” §§320.1(a)(6), 331.2. They are binding for five years on both the Corps and the Environmental Protection Agency, which share authority to enforce the Clean Water Act. See 33 U. S. C. §§1319, 1344(s); 33 CFR pt. 331, App. C; EPA, Memorandum of Agreement: Exemptions Under Section 404(F) of the Clean Water Act §VI–A (1989) (Memorandum of Agreement). B Respondents are three companies engaged in mining peat in Marshall County, Minnesota. Peat is an organic material that forms in waterlogged grounds, such as wetlands and bogs. See Xuehui & Jinming, Peat and Peatlands, in 2 Coal, Oil Shale, Natural Bitumen, Heavy Oil and Peat 267–272 (G. Jinsheng ed. 2009) (Peat and Peatlands). It is widely used for soil improvement and burned as fuel. Id., at 277. It can also be used to provide structural support and moisture for smooth, stable greens that leave golfers with no one to blame but themselves for errant putts. See Monteith & Welton, Use of Peat and Other Organic Materials on Golf Courses, 13 Bulletin of the United States Golf Association Green Section 90, 95–100 (1933). At the same time, peat mining can have significant environmental and ecological impacts, see Peat and Peatlands 280–281, and therefore is regulated by both federal and state environmental protection agencies, see, e.g., Minn. Stat. §103G.231 (2014). Respondents own a 530-acre tract near their existing mining operations. The tract includes wetlands, which respondents believe contain sufficient high quality peat, suitable for use in golf greens, to extend their mining operations for 10 to 15 years. App. 8, 14–15, 31. In December 2010, respondents applied to the Corps for a Section 404 permit for the property. Id., at 15. A Section 404 permit authorizes “the discharge of dredged or fill material into the navigable waters at specified disposal sites.” 33 U. S. C. §1344(a). Over the course of several communications with respondents, Corps officials signaled that the permitting process would be very expensive and take years to complete. The Corps also advised respondents that, if they wished to pursue their application, they would have to submit numerous assessments of various features of the property, which respondents estimate would cost more than $100,000. App. 16–17, 31–35. In February 2012, in connection with the permitting process, the Corps issued an approved JD stating that the property contained “water of the United States” because its wetlands had a “significant nexus” to the Red River of the North, located some 120 miles away. Id., at 13, 18, 20. Respondents appealed the JD to the Corps’ Mississippi Valley Division Commander, who remanded for further factfinding. On remand, the Corps reaffirmed its original conclusion and issued a revised JD to that effect. Id., at 18–20; App. to Pet. for Cert. 44a–45a. Respondents then sought judicial review of the revised JD under the Administrative Procedure Act (APA), 5 U. S. C. §500 et seq. The District Court dismissed for want of subject matter jurisdiction, holding that the revised JD was not “final agency action for which there is no other adequate remedy in a court,” as required by the APA prior to judicial review, 5 U. S. C. §704. 963 F. Supp. 2d 868, 872, 878 (Minn. 2013). The Court of Appeals for the Eighth Circuit reversed, 782 F. 3d 994, 1002 (2015), and we granted certiorari, 577 U. S. ___ (2015). II The Corps contends that the revised JD is not “final agency action” and that, even if it were, there are adequate alternatives for challenging it in court. We disagree at both turns. A In Bennett v. Spear, 520 U. S. 154 (1997) , we distilled from our precedents two conditions that generally must be satisfied for agency action to be “final” under the APA. “First, the action must mark the consummation of the agency’s decisionmaking process—it must not be of a merely tentative or interlocutory nature. And second, the action must be one by which rights or obligations have been determined, or from which legal consequences will flow.” Id., at 177–178 (internal quotation marks and citation omitted).[2] The Corps does not dispute that an approved JD satisfies the first Bennett condition. Unlike preliminary JDs— which are “advisory in nature” and simply indicate that “there may be waters of the United States” on a parcel of property, 33 CFR §331.2—an approved JD clearly “mark[s] the consummation” of the Corps’ decisionmaking process on that question, Bennett, 520 U. S., at 178 (internal quotation marks omitted). It is issued after extensive factfinding by the Corps regarding the physical and hydrological characteristics of the property, see U. S. Army Corps of Engineers, Jurisdictional Determination Form Instructional Guidebook 47–60 (2007), and is typically not revisited if the permitting process moves forward. Indeed, the Corps itself describes approved JDs as “final agency action,” see 33 CFR §320.1(a)(6), and specifies that an approved JD “will remain valid for a period of five years,” Corps, Regulatory Guidance Letter No. 05–02, §1(a), p. 1 (June 14, 2005) (2005 Guidance Letter); see also 33 CFR pt. 331, App. C. The Corps may revise an approved JD within the five-year period based on “new information.” 2005 Guidance Letter §1(a), at 1. That possibility, however, is a common characteristic of agency action, and does not make an otherwise definitive decision nonfinal. See Sackett v. EPA, 566 U. S. ___, ___ (2012); see also National Cable & Telecommunications Assn. v. Brand X Internet Services, 545 U. S. 967, 981 (2005) . By issuing respondents an approved JD, the Corps for all practical purposes “has ruled definitively” that respondents’ property contains jurisdictional waters. Sackett, 566 U. S., at ___ (Ginsburg, J., concurring) (slip op., at 1). The definitive nature of approved JDs also gives rise to “direct and appreciable legal consequences,” thereby satisfying the second prong of Bennett. 520 U. S., at 178. Consider the effect of an approved JD stating that a party’s property does not contain jurisdictional waters—a “negative” JD, in Corps parlance. As noted, such a JD will generally bind the Corps for five years. See 33 CFR pt. 331, App. C; 2005 Guidance Letter §1. Under a longstanding memorandum of agreement between the Corps and EPA, it will also be “binding on the Government and represent the Government’s position in any subsequent Federal action or litigation concerning that final determination.” Memorandum of Agreement §§IV–C–2, VI–A. A negative JD thus binds the two agencies authorized to bring civil enforcement proceedings under the Clean Water Act, see 33 U. S. C. §1319, creating a five-year safe harbor from such proceedings for a property owner. Additionally, although the property owner may still face a citizen suit under the Act, such a suit—unlike actions brought by the Government—cannot impose civil liability for wholly past violations. See §§1319(d), 1365(a); Gwaltney of Smithfield, Ltd. v. Chesapeake Bay Foundation, Inc., 484 U. S. 49 –59 (1987). In other words, a negative JD both narrows the field of potential plaintiffs and limits the potential liability a landowner faces for discharging pollutants without a permit. Each of those effects is a “legal consequence[ ]” satisfying the second Bennett prong. 520 U. S., at 178; see also Sackett, 566 U. S., at ___. It follows that affirmative JDs have legal consequences as well: They represent the denial of the safe harbor that negative JDs afford. See 5 U. S. C. §551(13) (defining “agency action” to include an agency “rule, order, license, sanction, relief, or the equivalent,” or the “denial thereof ”). Because “legal consequences . . . flow” from approved JDs, they constitute final agency action. Bennett, 520 U. S., at 178 (internal quotation marks omitted).[3] This conclusion tracks the “pragmatic” approach we have long taken to finality. Abbott Laboratories v. Gardner, 387 U. S. 136, 149 (1967) . For example, in Frozen Food Express v. United States, 351 U. S. 40 (1956) , we considered the finality of an order specifying which commodities the Interstate Commerce Commission believed were exempt by statute from regulation, and which it believed were not. Although the order “had no authority except to give notice of how the Commission interpreted” the relevant statute, and “would have effect only if and when a particular action was brought against a particular carrier,” Abbott, 387 U. S., at 150, we held that the order was nonetheless immediately reviewable, Frozen Food, 351 U. S., at 44–45. The order, we explained, “warns every carrier, who does not have authority from the Commission to transport those commodities, that it does so at the risk of incurring criminal penalties.” Id., at 44. So too here, while no administrative or criminal proceeding can be brought for failure to conform to the approved JD itself, that final agency determination not only deprives respondents of a five-year safe harbor from liability under the Act, but warns that if they discharge pollutants onto their property without obtaining a permit from the Corps, they do so at the risk of significant criminal and civil penalties. B Even if final, an agency action is reviewable under the APA only if there are no adequate alternatives to APA review in court. 5 U. S. C. §704. The Corps contends that respondents have two such alternatives: either discharge fill material without a permit, risking an EPA enforcement action during which they can argue that no permit was required, or apply for a permit and seek judicial review if dissatisfied with the results. Brief for Petitioner 45–51. Neither alternative is adequate. As we have long held, parties need not await enforcement proceedings before challenging final agency action where such proceedings carry the risk of “serious criminal and civil penalties.” Abbott, 387 U. S., at 153. If respondents discharged fill material without a permit, in the mistaken belief that their property did not contain jurisdictional waters, they would expose themselves to civil penalties of up to $37,500 for each day they violated the Act, to say nothing of potential criminal liability. See 33 U. S. C. §§1319(c), (d); Sackett, 566 U. S., at ___, n. 1 (citing 74 Fed. Reg. 626, 627 (2009)). Respondents need not assume such risks while waiting for EPA to “drop the hammer” in order to have their day in court. Sackett, 566 U. S., at ___ (slip op., at 6). Nor is it an adequate alternative to APA review for a landowner to apply for a permit and then seek judicial review in the event of an unfavorable decision. As Corps officials indicated in their discussions with respondents, the permitting process can be arduous, expensive, and long. See Rapanos, 547 U. S., at 721 (plurality opinion). On top of the standard permit application that respondents were required to submit, see 33 CFR §325.1(d) (detailing contents of permit application), the Corps demanded that they undertake, among other things, a “hydrogeologic assessment of the rich fen system including the mineral/nutrient composition and pH of the groundwater; groundwater flow spatially and vertically; discharge and recharge areas”; a “functional/resource assessment of the site including a vegetation survey and identification of native fen plan communities across the site”; an “inven-tory of similar wetlands in the general area (watershed), including some analysis of their quality”; and an “inven-tory of rich fen plant communities that are within sites of High and Outstanding Biodiversity Significance in the area.” App. 33–34. Respondents estimate that undertaking these analyses alone would cost more than $100,000. Id., at 17. And whatever pertinence all this might have to the issuance of a permit, none of it will alter the finality of the approved JD, or affect its suitability for judicial review. The permitting process adds nothing to the JD. The Corps nevertheless argues that Congress made the “evident[ ]” decision in the Clean Water Act that a coverage determination would be made “as part of the permitting process, and that the property owner would obtain any necessary judicial review of that determination at the conclusion of that process.” Brief for Petitioner 46. But as the Corps acknowledges, the Clean Water Act makes no reference to standalone jurisdictional determinations, ibid., so there is little basis for inferring anything from it concerning the reviewability of such distinct final agency action. And given “the APA’s presumption of reviewability for all final agency action,” Sackett, 566 U. S., at ___ (slip op., at 8), “[t]he mere fact” that permitting decisions are “reviewable should not suffice to support an implication of exclusion as to other[ ]” agency actions, such as approved JDs, Abbott, 387 U. S., at 141 (internal quotation marks omitted); see also Sackett, 566 U. S., at ___ (slip op., at 8) (“[I]f the express provision of judicial review in one section of a long and complicated statute were alone enough to overcome the APA’s presumption of reviewability . . . , it would not be much of a presumption at all”). Finally, the Corps emphasizes that seeking review in an enforcement action or at the end of the permitting process would be the only available avenues for obtaining review “[i]f the Corps had never adopted its practice of issuing standalone jurisdictional determinations upon request.” Reply Brief 3; see also id., at 4, 23. True enough. But such a “count your blessings” argument is not an adequate rejoinder to the assertion of a right to judicial review under the APA. The judgment of the Court of Appeals for the Eighth Circuit is affirmed. It is so ordered.Notes 1 In 2015, the Corps adopted a new rule modifying the definition of the scope of waters covered by the Clean Water Act in light of scientific research and decisions of this Court interpreting the Act. See Clean Water Rule: Definition of “Waters of the United States,” 80 Fed. Reg. 37054, 37055–37056. That rule is currently stayed nationwide, pending resolution of claims that the rule is arbitrary, capricious, and contrary to law. See In re EPA, 803 F. 3d 804, 807–809 (CA6 2015). 2 Because we determine that a JD satisfies both prongs of Bennett, we need not consider respondents’ argument that an agency action that satisfies only the first may also constitute final agency action. See Brief for Respondents 19–20. 3 The Corps asserts that the Memorandum of Agreement addresses only “special case” JDs, rather than “mine-run” ones “of the sort at issue here.” Reply Brief 12, n. 3. But the memorandum plainly makes binding “[a]ll final determinations,” whether in “[s]pecial” or “[n]on-special” cases. Memorandum of Agreement §§IV–C, VI–A; see also Corps, Memorandum of Understanding Geographical Jurisdiction of the Section 404 Program, 45 Fed. Reg. 45019, n. 1 (1980) (“[U]nder this [memorandum], except in special cases previously agreed to, the [Corps] is authorized to make a final determination . . . and such determination shall be binding.”).

579.US.2015_15-420

In response to the high incidence of domestic violence against Native American women, Congress enacted a felony offense of domestic assault in Indian country by a habitual offender. 18 U. S. C. §117(a). Section 117(a)(1) provides that any person who “commits a domestic assault within . . . Indian country” and who has at least two prior final convictions for domestic violence rendered “in Federal, State, or Indian tribal court proceedings . . . shall be fined . . . , imprisoned for a term of not more than 5 years, or both . . . .” Having two prior tribal-court convictions for domestic violence crimes is thus a predicate of the new offense. This case raises the question whether §117(a)’s inclusion of tribal-court convictions as predicate offenses is compatible with the Sixth Amendment’s right to counsel. The Sixth Amendment guarantees indigent defendants appointed counsel in any state or federal criminal proceeding in which a term of imprisonment is imposed, Scott v. Illinois, 440 U. S. 367 –374, but it does not apply in tribal-court proceedings, see Plains Commerce Bank v. Long Family Land & Cattle Co., 554 U. S. 316 . The Indian Civil Rights Act of 1968 (ICRA), which governs tribal-court proceedings, accords a range of procedural safeguards to tribal-court defendants “similar, but not identical, to those contained in the Bill of Rights and the Fourteenth Amendment,” Santa Clara Pueblo v. Martinez, 436 U. S. 49 . In particular, ICRA provides indigent defendants with a right to appointed counsel only for sentences exceeding one year. 25 U. S. C. §1302(c)(2). ICRA’s right to counsel therefore is not coextensive with the Sixth Amendment right. This Court has held that a conviction obtained in state or federal court in violation of a defendant’s Sixth Amendment right to counsel cannot be used in a subsequent proceeding “to support guilt or enhance punishment for another offense.” Burgett v. Texas, 389 U. S. 109 . Use of a constitutionally infirm conviction would cause “the accused in effect [to] suffe[r] anew from the [prior] deprivation of [his] Sixth Amendment right.” Ibid. Burgett’s principle was limited by the Court’s holding in Nichols v. United States, 511 U. S. 738 , that “an uncounseled misdemeanor conviction, valid under Scott because no prison term was imposed, is also valid when used to enhance punishment at a subsequent conviction,” id., at 748–749. Respondent Michael Bryant, Jr., has multiple tribal-court convictions for domestic assault. When convicted, Bryant was indigent and was not appointed counsel. For most of his convictions, he was sentenced to terms of imprisonment not exceeding one year’s duration. Because of his short prison terms, the prior tribal-court proceedings complied with ICRA, and his convictions were therefore valid when entered. Based on domestic assaults he committed in 2011, Bryant was indicted on two counts of domestic assault by a habitual offender, in violation of §117(a). Represented in federal court by appointed counsel, he contended that the Sixth Amendment precluded use of his prior, uncounseled, tribal-court misdemeanor convictions to satisfy §117(a)’s predicate-offense element and moved to dismiss the indictment. The District Court denied the motion; Bryant pleaded guilty, reserving the right to appeal. The Ninth Circuit reversed the conviction and directed dismissal of the indictment. It comprehended that Bryant’s uncounseled tribal-court convictions were valid when entered because the Sixth Amendment right to counsel does not apply in tribal-court proceedings. It held, however, that Bryant’s tribal-court convictions could not be used as predicate convictions within §117(a)’s compass because they would have violated the Sixth Amendment had they been rendered in state or federal court. Held: Because Bryant’s tribal-court convictions occurred in proceedings that complied with ICRA and were therefore valid when entered, use of those convictions as predicate offenses in a §117(a) prosecution does not violate the Constitution. Nichols instructs that convictions valid when entered retain that status when invoked in a subsequent proceeding. Nichols reasoned that “[e]nhancement statutes . . . do not change the penalty imposed for the earlier conviction”; rather, repeat-offender laws “penaliz[e] only the last offense committed by the defendant.” 511 U. S., at 747. Bryant’s sentence for violating §117(a) punishes his most recent acts of domestic assault, not his prior crimes prosecuted in tribal court. He was denied no right to counsel in tribal court, and his Sixth Amendment right was honored in federal court. Bryant acknowledges that Nichols would have allowed reliance on uncounseled tribal-court convictions resulting in fines to satisfy §117(a)’s prior-crimes predicate. But there is no cause to distinguish for §117(a) purposes between fine-only tribal-court convictions and valid but uncounseled tribal-court convictions resulting in imprisonment for a term not exceeding one year. Neither violates the Sixth Amendment. Bryant is not aided by Burgett. A defendant convicted in tribal court suffered no Sixth Amendment violation in the first instance, so he cannot “suffe[r] anew” from a prior deprivation in his federal prosecution. Bryant also invokes the Due Process Clause of the Fifth Amendment to support his assertion that tribal-court judgments should not be used as predicate offenses under §117(a). ICRA, however, guarantees “due process of law,” accords other procedural safeguards, and permits a prisoner to challenge the fundamental fairness of tribal court proceedings in federal habeas corpus proceedings. Because proceedings in compliance with ICRA sufficiently ensure the reliability of tribal-court convictions, the use of those convictions in a federal prosecution does not violate a defendant’s due process right. Pp. 12–16. 769 F. 3d 671, reversed and remanded. Ginsburg, J., delivered the opinion for a unanimous Court. Thomas, J., filed a concurring opinion.

In response to the high incidence of domestic violence against Native American women, Congress, in 2005, enacted 18 U. S. C. §117(a), which targets serial offenders. Section 117(a) makes it a federal crime for any person to “commi[t] a domestic assault within . . . Indian country” if the person has at least two prior final convictions for domestic violence rendered “in Federal, State, or Indian tribal court proceedings.” See Violence Against Women and Department of Justice Reauthorization Act of 2005 (VAWA Reauthorization Act), Pub. L. 109–162, §§901, 909, 119Stat. 3077, 3084.[1] Respondent Michael Bryant, Jr., has multiple tribal-court convictions for domestic assault. For most of those convictions, he was sentenced to terms of imprisonment, none of them exceeding one year’s duration. His tribal-court convictions do not count for §117(a) purposes, Bryant maintains, because he was uncounseled in those proceedings. The Sixth Amendment guarantees indigent defendants, in state and federal criminal proceedings, appointed counsel in any case in which a term of imprisonment is imposed. Scott v. Illinois, 440 U. S. 367 –374 (1979). But the Sixth Amendment does not apply to tribal-court proceedings. See Plains Commerce Bank v. Long Family Land & Cattle Co., 554 U. S. 316, 337 (2008) . The Indian Civil Rights Act of 1968 (ICRA), Pub. L. 90–284, 82Stat. 77, 25 U. S. C. §1301 et seq., which governs criminal proceedings in tribal courts, requires appointed counsel only when a sentence of more than one year’s imprisonment is imposed. §1302(c)(2). Bryant’s tribal-court convictions, it is undisputed, were valid when entered. This case presents the question whether those convictions, though uncounseled, rank as predicate offenses within the compass of §117(a). Our answer is yes. Bryant’s tribal-court convictions did not violate the Sixth Amendment when obtained, and they retain their validity when invoked in a §117(a) prosecution. That proceeding generates no Sixth Amendment defect where none previously existed. I A “[C]ompared to all other groups in the United States,” Native American women “experience the highest rates of domestic violence.” 151 Cong. Rec. 9061 (2005) (remarks of Sen. McCain). According to the Centers for Disease Control and Prevention, as many as 46% of American Indian and Alaska Native women have been victims of physical violence by an intimate partner. Centers for Disease Control and Prevention, National Center for Injury Prevention and Control, M. Black et al., National Intimate Partner and Sexual Violence Survey 2010 Summary Report 40 (2011) (Table 4.3), online at http://www.cdc.gov/ViolencePrevention/pdf/NISVS_report2010-a.pdf (all Internet materials as last visited June 9, 2016). American Indian and Alaska Native women “are 2.5 times more likely to be raped or sexually assaulted than women in the United States in general.” Dept. of Justice, Attorney General’s Advisory Committee on American Indian and Alaska Native Children Exposed to Violence, Ending Violence So Children Can Thrive 38 (Nov. 2014), online at https:// www.justice.gov /sites /default/files/defendingchildhood/ pages/attachments/2015/03/23/ending_violence_so_children_can_thrive.pdf. American Indian women experience battery “at a rate of 23.2 per 1,000, compared with 8 per 1,000 among Caucasian women,” and they “experience 7 sexual assaults per 1,000, compared with 4 per 1,000 among Black Americans, 3 per 1,000 among Caucasians, 2 per 1,000 among Hispanic women, and 1 per 1,000 among Asian women.” VAWA Reauthorization Act, §901, 119Stat. 3077. As this Court has noted, domestic abusers exhibit high rates of recidivism, and their violence “often escalates in severity over time.” United States v. Castleman, 572 U. S. ___, ___ (2014) (slip op., at 2). Nationwide, over 75% of female victims of intimate partner violence have been previously victimized by the same offender, Dept. of Justice, Bureau of Justice Statistics, S. Catalano, Intimate Partner Violence 1993–2010, p. 4 (rev. 2015) (Figure 4), online at http://www.bjs.gov/content/pub/pdf/ipv9310.pdf, often multiple times, Dept. of Justice, National Institute of Justice, P. Tjaden & N. Thoennes, Extent, Nature, and Consequences of Intimate Partner Violence, p. iv (2000), online at https://www.ncjrs.gov/pdffiles1/nij/181867.pdf (“[W]omen who were physically assaulted by an intimate partner averaged 6.9 physical assaults by the same partner.”). Incidents of repeating, escalating abuse more than occasionally culminate in a fatal attack. See VAWA Reauthorization Act, §901, 119Stat. 3077–3078 (“[D]uring the period 1979 through 1992, homicide was the third leading cause of death of Indian females aged 15 to 34, and 75 percent were killed by family members or acquaintances.”). The “complex patchwork of federal, state, and tribal law” governing Indian country, Duro v. Reina, 495 U. S. 676 , n. 1 (1990), has made it difficult to stem the tide of domestic violence experienced by Native American women. Although tribal courts may enforce the tribe’s criminal laws against Indian defendants, Congress has curbed tribal courts’ sentencing authority. At the time of §117(a)’s passage, ICRA limited sentences in tribal court to a maximum of one year’s imprisonment. 25 U. S. C. §1302(a)(7) (2006 ed.).[2] Congress has since expanded tribal courts’ sentencing authority, allowing them to impose up to three years’ imprisonment, contingent on adoption of additional procedural safeguards. 124Stat. 2279–2280 (codified at 25 U. S. C. §1302(a)(7)(C), (c)).[3] To date, however, few tribes have employed this enhanced sentencing authority. See Tribal Law and Policy Inst., Implementation Chart: VAWA Enhanced Jurisdiction and TLOA Enhanced Sentencing, online at http://www.tribal-institute.org/download/VAWA/VAWAImplementationChart.pdf.[4] States are unable or unwilling to fill the enforcement gap. Most States lack jurisdiction over crimes committed in Indian country against Indian victims. See United States v. John, 437 U. S. 634, 651 (1978) . In 1953, Congress increased the potential for state action by giving six States “jurisdiction over specified areas of Indian country within the States and provid[ing] for the [voluntary] assumption of jurisdiction by other States.” California v. Cabazon Band of Mission Indians, 480 U. S. 202, 207 (1987) (footnote omitted). See Act of Aug. 15, 1953, Pub. L. 280, 67Stat. 588 (codified, as amended, at 18 U. S. C. §1162 and 25 U. S. C. §§1321–1328, 1360). States so empowered may apply their own criminal laws to “offenses committed by or against Indians within all Indian country within the State.” Cabazon Band of Mission Indians, 480 U. S., at 207; see 18 U. S. C. §1162(a). Even when capable of exercising jurisdiction, however, States have not de-voted their limited criminal justice resources to crimes committed in Indian country. Jimenez & Song, Concurrent Tribal and State Jurisdiction Under Public Law 280, 47 Am. U. L. Rev. 1627, 1636–1637 (1998); Tribal Law and Policy Inst., S. Deer, C. Goldberg, H. Valdez Singleton, & M. White Eagle, Final Report: Focus Group on Public Law 280 and the Sexual Assault of Native Women 7–8 (2007), online at http://www.tribal-institute.org/download/Final%20280%20FG%20Report.pdf. That leaves the Federal Government. Although federal law generally governs in Indian country, Congress has long excluded from federal-court jurisdiction crimes committed by an Indian against another Indian. 18 U. S. C. §1152; see Ex parte Crow Dog, 109 U. S. 556, 572 (1883) (requiring “a clear expression of the intention of Congress” to confer federal jurisdiction over crimes committed by an Indian against another Indian). In the Major Crimes Act, Congress authorized federal jurisdiction over enumerated grave criminal offenses when both perpetrator and victim are Indians, including murder, manslaughter, and felony assault. §1153. At the time of §117(a)’s enactment, felony assault subject to federal prosecution required “serious bodily injury,” §113(a)(6) (2006 ed.), meaning “a substantial risk of death,” “extreme physical pain,” “protracted and obvious disfigurement,” or “protracted loss or impairment of the function of a bodily member, organ, or mental faculty.” §1365(h)(3) (incorporated through §113(b)(2)).[5] In short, when §117(a) was before Congress, Indian perpetrators of domestic violence “escape[d] felony charges until they seriously injure[d] or kill[ed] someone.” 151 Cong. Rec. 9062 (2005) (remarks of Sen. McCain). As a result of the limitations on tribal, state, and federal jurisdiction in Indian country, serial domestic violence offenders, prior to the enactment of §117(a), faced at most a year’s imprisonment per offense—a sentence insufficient to deter repeated and escalating abuse. To ratchet up the punishment of serial offenders, Congress created the federal felony offense of domestic assault in Indian country by a habitual offender. §117(a) (2012 ed.); see No. 12–30177 (CA9, July 6, 2015), App. to Pet. for Cert. 41a (Owens, J., dissenting from denial of rehearing en banc) (“Tailored to the unique problems . . . that American Indian and Alaska Native Tribes face, §117(a) provides felony-level punishment for serial domestic violence offenders, and it represents the first true effort to remove these recidivists from the communities that they repeatedly terrorize.”). The section provides in pertinent part: “Any person who commits a domestic assault within . . . Indian country and who has a final conviction on at least 2 separate prior occasions in Federal, State, or Indian tribal court proceedings for offenses that would be, if subject to Federal jurisdiction any assault, sex-ual abuse, or serious violent felony against a spouse or intimate partner . . . shall be fined . . . , imprisoned for a term of not more than 5 years, or both . . . .” §117(a)(1).[6] Having two prior convictions for domestic violence crimes—including tribal-court convictions—is thus a predicate of the new offense. B This case requires us to determine whether §117(a)’s inclusion of tribal-court convictions is compatible with the Sixth Amendment’s right to counsel. The Sixth Amendment to the U. S. Constitution guarantees a criminal defendant in state or federal court “the Assistance of Counsel for his defence.” See Gideon v. Wainwright, 372 U. S. 335, 339 (1963) . This right, we have held, requires appointment of counsel for indigent defendants whenever a sentence of imprisonment is imposed. Argersinger v. Hamlin, 407 U. S. 25, 37 (1972) . But an indigent defendant has no constitutional right to appointed counsel if his conviction results in a fine or other noncustodial punishment. Scott, 440 U. S., at 373–374. “As separate sovereigns pre-existing the Constitution, tribes have historically been regarded as unconstrained by those constitutional provisions framed specifically as limitations on federal or state authority.” Santa Clara Pueblo v. Martinez, 436 U. S. 49, 56 (1978) . The Bill of Rights, including the Sixth Amendment right to counsel, therefore, does not apply in tribal-court proceedings. See Plains Commerce Bank, 554 U. S., at 337. In ICRA, however, Congress accorded a range of procedural safeguards to tribal-court defendants “similar, but not identical, to those contained in the Bill of Rights and the Fourteenth Amendment.” Martinez, 436 U. S., at 57; see id., at 62–63 (ICRA “modified the safeguards of the Bill of Rights to fit the unique political, cultural, and economic needs of tribal governments”). In addition to other enumerated protections, ICRA guarantees “due process of law,” 25 U. S. C. §1302(a)(8), and allows tribal-court defendants to seek habeas corpus review in federal court to test the legality of their imprisonment, §1303. The right to counsel under ICRA is not coextensive with the Sixth Amendment right. If a tribal court imposes a sentence in excess of one year, ICRA requires the court to accord the defendant “the right to effective assistance of counsel at least equal to that guaranteed by the United States Constitution,” including appointment of counsel for an indigent defendant at the tribe’s expense. §1302(c)(1), (2). If the sentence imposed is no greater than one year, however, the tribal court must allow a defendant only the opportunity to obtain counsel “at his own expense.” §1302(a)(6). In tribal court, therefore, unlike in federal or state court, a sentence of imprisonment up to one year may be imposed without according indigent defendants the right to appointed counsel. The question here presented: Is it permissible to use uncounseled tribal-court convictions—obtained in full compliance with ICRA—to establish the prior-crimes predicate of §117(a)? It is undisputed that a conviction obtained in violation of a defendant’s Sixth Amendment right to counsel cannot be used in a subsequent proceeding “either to support guilt or enhance punishment for another offense.” Burgett v. Texas, 389 U. S. 109, 115 (1967) . In Burgett, we held that an uncounseled felony conviction obtained in state court in violation of the right to counsel could not be used in a subsequent proceeding to prove the prior-felony element of a recidivist statute. To permit such use of a constitutionally infirm conviction, we explained, would cause “the accused in effect [to] suffe[r] anew from the [prior] deprivation of [his] Sixth Amendment right.” Ibid.; see United States v. Tucker, 404 U. S. 443, 448 (1972) (invalid, uncounseled prior convictions could not be relied upon at sentencing to impose a longer term of imprisonment for a subsequent conviction); cf. Loper v. Beto, 405 U. S. 473 –484 (1972) (plurality opinion) (“use of convictions constitutionally invalid under Gideon v. Wainwright to impeach a defendant’s credibility deprives him of due process of law” because the prior convictions “lac[k] reliability”). In Nichols v. United States, 511 U. S. 738 (1994) , we stated an important limitation on the principle recognized in Burgett. In the case under review, Nichols pleaded guilty to a federal felony drug offense. 511 U. S., at 740. Several years earlier, unrepresented by counsel, he had been convicted of driving under the influence (DUI), a state-law misdemeanor, and fined $250 but not imprisoned. Ibid. Nichols’ DUI conviction, under the then-mandatory Sentencing Guidelines, effectively elevated by about two years the sentencing range for Nichols’ federal drug offense. Ibid. We rejected Nichols’ contention that, as his later sentence for the federal drug offense involved imprisonment, use of his uncounseled DUI conviction to elevate that sentence violated the Sixth Amendment. Id., at 746–747. “[C]onsistent with the Sixth and Fourteenth Amendments of the Constitution,” we held, “an uncounseled misdemeanor conviction, valid under Scott because no prison term was imposed, is also valid when used to enhance punishment at a subsequent conviction.” Id., at 748–749. C Respondent Bryant’s conduct is illustrative of the domestic violence problem existing in Indian country. During the period relevant to this case, Bryant, an enrolled member of the Northern Cheyenne Tribe, lived on that Tribe’s reservation in Montana. He has a record of over 100 tribal-court convictions, including several misdemeanor convictions for domestic assault. Specifically, between 1997 and 2007, Bryant pleaded guilty on at least five occasions in Northern Cheyenne Tribal Court to committing domestic abuse in violation of the Northern Cheyenne Tribal Code. On one occasion, Bryant hit his live-in girlfriend on the head with a beer bottle and attempted to strangle her. On another, Bryant beat a different girlfriend, kneeing her in the face, breaking her nose, and leaving her bruised and bloodied. For most of Bryant’s repeated brutal acts of domestic violence, the Tribal Court sentenced him to terms of imprisonment, never exceeding one year. When convicted of these offenses, Bryant was indigent and was not appointed counsel. Because of his short prison terms, Bryant acknowledges, the prior tribal-court proceedings complied with ICRA, and his convictions were therefore valid when entered. Bryant has never challenged his tribal-court convictions in federal court under ICRA’s habeas corpus provision. In 2011, Bryant was arrested yet again for assaulting women. In February of that year, Bryant attacked his then girlfriend, dragging her off the bed, pulling her hair, and repeatedly punching and kicking her. During an interview with law enforcement officers, Bryant admitted that he had physically assaulted this woman five or six times. Three months later, he assaulted another woman with whom he was then living, waking her by yelling that he could not find his truck keys and then choking her until she almost lost consciousness. Bryant later stated that he had assaulted this victim on three separate occasions during the two months they dated. Based on the 2011 assaults, a federal grand jury in Montana indicted Bryant on two counts of domestic assault by a habitual offender, in violation of §117(a). Bryant was represented in federal court by appointed counsel. Contending that the Sixth Amendment precluded use of his prior, uncounseled, tribal-court misdemeanor convictions to satisfy §117(a)’s predicate-offense element, Bryant moved to dismiss the indictment. The District Court denied the motion, App. to Pet. for Cert. 32a, and Bryant entered a conditional guilty plea, reserving the right to appeal that decision. Bryant was sentenced to concurrent terms of 46 months’ imprisonment on each count, to be followed by three years of supervised release. The Court of Appeals for the Ninth Circuit reversed the conviction and directed dismissal of the indictment. 769 F. 3d 671 (2014). Bryant’s tribal-court convictions were not themselves constitutionally infirm, the Ninth Circuit comprehended, because “the Sixth Amendment right to appointed counsel does not apply in tribal court proceedings.” Id., at 675. But, the court continued, had the convictions been obtained in state or federal court, they would have violated the Sixth Amendment because Bryant had received sentences of imprisonment although he lacked the aid of appointed counsel. Adhering to its prior decision in United States v. Ant, 882 F. 2d 1389 (CA9 1989),[7] the Court of Appeals held that, subject to narrow exceptions not relevant here, “tribal court convictions may be used in subsequent [federal] prosecutions only if the tribal court guarantees a right to counsel that is, at minimum, coextensive with the Sixth Amendment right.” 769 F. 3d, at 677. Rejecting the Government’s argument that our decision in Nichols required the opposite result, the Ninth Circuit concluded that Nichols applies only when the prior conviction did comport with the Sixth Amendment, i.e., when no sentence of imprisonment was imposed for the prior conviction. 769 F. 3d, at 677–678. Judge Watford concurred, agreeing that Ant controlled the outcome of this case, but urging reexamination of Ant in light of Nichols. 769 F. 3d, at 679. This Court’s decision in Nichols, Judge Watford wrote, “undermines the notion that uncounseled convictions are, as a categorical matter, too unreliable to be used as a basis for imposing a prison sentence in a subsequent case.” 769 F. 3d, at 679. The Court of Appeals declined to rehear the case en banc over vigorous dissents by Judges Owens and O’Scannlain. In disallowing the use of an uncounseled tribal-court conviction to establish a prior domestic violence conviction within §117(a)’s compass, the Ninth Circuit created a Circuit split. The Eighth and Tenth Circuits have both held that tribal-court “convictions, valid at their inception, and not alleged to be otherwise unreliable, may be used to prove the elements of §117.” United States v. Cavanaugh, 643 F. 3d 592, 594 (CA8 2011); see United States v. Shavanaux, 647 F. 3d 993, 1000 (CA10 2011). To resolve this disagreement, we granted certiorari, 577 U. S. ___ (2016), and now reverse. II Bryant’s tribal-court convictions, he recognizes, infringed no constitutional right because the Sixth Amendment does not apply to tribal-court proceedings. Brief for Respondent 5. Those prior convictions complied with ICRA, he concedes, and therefore were valid when entered. But, had his convictions occurred in state or federal court, Bryant observes, Argersinger and Scott would have rendered them invalid because he was sentenced to incarceration without representation by court-appointed counsel. Essentially, Bryant urges us to treat tribal-court convictions, for §117(a) purposes, as though they had been entered by a federal or state court. We next explain why we decline to do so. As earlier recounted, we held in Nichols that “an uncounseled misdemeanor conviction, valid under Scott because no prison term was imposed, is also valid when used to enhance punishment at a subsequent conviction.” 511 U. S., at 748–749. “Enhancement statutes,” we reasoned, “do not change the penalty imposed for the earlier conviction”; rather, repeat-offender laws “penaliz[e] only the last offense committed by the defendant.” Id., at 747; see United States v. Rodriquez, 553 U. S. 377, 386 (2008) (“When a defendant is given a higher sentence under a recidivism statute . . . 100% of the punishment is for the offense of conviction. None is for the prior convictions or the defendant’s ‘status as a recidivist.’ ”). Nichols thus instructs that convictions valid when entered—that is, those that, when rendered, did not violate the Constitution—retain that status when invoked in a subsequent proceeding. Nichols’ reasoning steers the result here. Bryant’s 46-month sentence for violating §117(a) punishes his most recent acts of domestic assault, not his prior crimes prosecuted in tribal court. Bryant was denied no right to counsel in tribal court, and his Sixth Amendment right was honored in federal court, when he was “adjudicated guilty of the felony offense for which he was imprisoned.” Alabama v. Shelton, 535 U. S. 654, 664 (2002) . It would be “odd to say that a conviction untainted by a violation of the Sixth Amendment triggers a violation of that same amendment when it’s used in a subsequent case where the defendant’s right to appointed counsel is fully respected.” 769 F. 3d, at 679 (Watford, J., concurring).[8] Bryant acknowledges that had he been punished only by fines in his tribal-court proceedings, Nichols would have allowed reliance on his uncounseled convictions to satisfy §117(a)’s prior-crimes predicate. Brief for Respondent 50. We see no cause to distinguish for §117(a) purposes between valid but uncounseled convictions resulting in a fine and valid but uncounseled convictions resulting in imprisonment not exceeding one year. “Both Nichols’s and Bryant’s uncounseled convictions ‘comport’ with the Sixth Amendment, and for the same reason: the Sixth Amendment right to appointed counsel did not apply to either conviction.” App. to Pet. for Cert. 50a (O’Scannlain, J., dissenting from denial of rehearing en banc). In keeping with Nichols, we resist creating a “hybrid” category of tribal-court convictions, “good for the punishment actually imposed but not available for sentence enhancement in a later prosecution.” 511 U. S., at 744. Nichols indicates that use of Bryant’s uncounseled tribal-court convictions in his §117(a) prosecution did not “transform his prior, valid, tribal court convictions into new, invalid, federal ones.” App. to Pet. for Cert. 50a (opinion of O’Scannlain, J.). Our decision in Burgett, which prohibited the subsequent use of a conviction obtained in violation of the right to counsel, does not aid Bryant. Reliance on an invalid conviction, Burgett reasoned, would cause the accused to “suffe[r] anew from the deprivation of [his] Sixth Amendment right.” 389 U. S., at 115. Because a defendant convicted in tribal court suffers no Sixth Amendment violation in the first instance, “[u]se of tribal convictions in a subsequent prosecution cannot violate [the Sixth Amendment] ‘anew.’ ” Shavanaux, 647 F. 3d, at 998. Bryant observes that reliability concerns underlie our right-to-counsel decisions and urges that those concerns remain even if the Sixth Amendment itself does not shelter him. Scott and Nichols, however, counter the argument that uncounseled misdemeanor convictions are categorically unreliable, either in their own right or for use in a subsequent proceeding. Bryant’s recognition that a tribal-court conviction resulting in a fine would qualify as a §117(a) predicate offense, we further note, diminishes the force of his reliability-based argument. There is no reason to suppose that tribal-court proceedings are less reliable when a sentence of a year’s imprisonment is imposed than when the punishment is merely a fine. No evidentiary or procedural variation turns on the sanction—fine only or a year in prison—ultimately imposed. Bryant also invokes the Due Process Clause of the Fifth Amendment in support of his assertion that tribal-court judgments should not be used as predicate offenses. But, as earlier observed, ICRA itself requires tribes to ensure “due process of law,” §1302(a)(8), and it accords defendants specific procedural safeguards resembling those contained in the Bill of Rights and the Fourteenth Amendment. See supra, at 8. Further, ICRA makes habeas review in federal court available to persons incarcerated pursuant to a tribal-court judgment. §1303. By that means, a prisoner may challenge the fundamental fairness of the proceedings in tribal court. Proceedings in compliance with ICRA, Congress determined, and we agree, sufficiently ensure the reliability of tribal-court convictions. Therefore, the use of those convictions in a federal prosecution does not violate a defendant’s right to due process. See Shavanaux, 647 F. 3d, at 1000; cf. State v. Spotted Eagle, 316 Mont. 370, 378–379, 71 P. 3d 1239, 1245–1246 (2003) (principles of comity support recognizing uncounseled tribal-court convictions that complied with ICRA). * * * Because Bryant’s tribal-court convictions occurred in proceedings that complied with ICRA and were therefore valid when entered, use of those convictions as predicate offenses in a §117(a) prosecution does not violate the Constitution. We accordingly reverse the judgment of the Court of Appeals for the Ninth Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 “Indian country” is defined in 18 U. S. C. §1151 to encompass all land within any Indian reservation under federal jurisdiction, all dependent Indian communities, and all Indian allotments, the Indian titles to which have not been extinguished. 2 Until 1986, ICRA permitted sentences of imprisonment up to only six months. See 100Stat. 3207–146. 3 Among the additional safeguards attending longer sentences is the unqualified right of an indigent defendant to appointed counsel. 25 U. S. C. §1302(c)(1), (2). 4 Tribal governments generally lack criminal jurisdiction over non-Indians who commit crimes in Indian country. See Oliphant v. Suquamish Tribe, 435 U. S. 191, 195 (1978) . In the Violence Against Women Reauthorization Act of 2013, Congress amended ICRA to authorize tribal courts to “exercise special domestic violence criminal jurisdiction” over certain domestic violence offenses committed by a non-Indian against an Indian. Pub. L. 113–4, §904, 127Stat. 120–122 (codified at 25 U. S. C. §1304). Tribal courts’ exercise of this jurisdiction requires procedural safeguards similar to those required for imposing on Indian defendants sentences in excess of one year, including the unqualified right of an indigent defendant to appointed counsel. See §1304(d). We express no view on the validity of those provisions. 5 Congress has since expanded the definition of felony assault to include “[a]ssault resulting in substantial bodily injury to a spouse[,] . . . intimate partner, [or] dating partner” and “[a]ssault of a spouse, intimate partner, or dating partner by strangling, suffocating, or attempting to strangle or suffocate.” Violence Against Women Reauthorization Act of 2013, §906, 127Stat. 124 (codified at 18 U. S. C. §113(a)(7), (8)). The “substantial bodily injury” requirement remains difficult to satisfy, as it requires “a temporary but substantial disfigurement” or “a temporary but substantial loss or impairment of the function of any bodily member, organ, or mental faculty.” §113(b)(1). 6 Section 117(a) has since been amended to include as qualifying predicate offenses, in addition to intimate-partner crimes, “assault, sexual abuse, [and] serious violent felony” offenses committed “against a child of or in the care of the person committing the domestic assault.” 18 U. S. C. §117(a) (Supp. II 2014). 7 In United States v. Ant, 882 F. 2d 1389 (1989), the Ninth Circuit proscribed the use of an uncounseled tribal-court guilty plea as evidence of guilt in a subsequent federal prosecution arising out of the same incident. Use of the plea was impermissible, the Court of Appeals reasoned, “because the tribal court guilty plea was made under circumstances which would have violated the United States Constitution were it applicable to tribal proceedings.” Id., at 1390. 8 True, as Bryant points out, we based our decision in Nichols v.United States, 511 U. S. 738, 747 (1994) , in part on the “less exact-ing” nature of sentencing, compared with the heightened burden of proof required for determining guilt. But, in describing the rule we adopted, we said that it encompasses both “criminal history provisions,” applicable at sentencing, and “recidivist statutes,” of which §117(a) is one. Ibid. Moreover, Nichols’ two primary rationales—the validity of the prior conviction and the sentence’s punishment of “only the last offense”—do not rely on a distinction between guilt adjudication and sentencing. Indeed, it is the validity of the prior conviction that distinguishes Nichols from United States v. Tucker, 404 U. S. 443, 448 (1972) , in which we found impermissible the use at sentencing of an invalid, uncounseled prior conviction.

579.US.2015_15-7

Yarushka Rivera, a teenage beneficiary of Massachusetts’ Medicaid program, received counseling services for several years at Arbour Counseling Services, a satellite mental health facility owned and operated by a subsidiary of petitioner Universal Health Services, Inc. She had an adverse reaction to a medication that a purported doctor at Arbour prescribed after diagnosing her with bipolar disorder. Her condition worsened, and she eventually died of a seizure. Respondents, her mother and stepfather, later discovered that few Arbour employees were actually licensed to provide mental health counseling or authorized to prescribe medications or offer counseling services without supervision. Respondents filed a qui tam suit, alleging that Universal Health had violated the False Claims Act (FCA). That Act imposes significant penalties on anyone who “knowingly presents . . . a false or fraudulent claim for payment or approval” to the Federal Government, 31 U. S. C. §3729(a)(1)(A). Respondents sought to hold Universal Health liable under what is commonly referred to as an “implied false certification theory of liability,” which treats a payment request as a claimant’s implied certification of compliance with relevant statutes, regulations, or contract requirements that are material conditions of payment and treats a failure to disclose a violation as a misrepresentation that renders the claim “false or fraudulent.” Specifically, respondents alleged, Universal Health (acting through Arbour) defrauded the Medicaid program by submitting reimbursement claims that made representations about the specific services provided by specific types of professionals, but that failed to disclose serious violations of Massachusetts Medicaid regulations pertaining to staff qualifications and licensing requirements for these services. Universal Health thus allegedly defrauded the program because Universal Health knowingly misrepresented its compliance with mental health facility requirements that are so central to the provision of mental health counseling that the Medicaid program would have refused to pay these claims had it known of these violations. The District Court granted Universal Health’s motion to dismiss. It held that respondents had failed to state a claim under the “implied false certification” theory of liability because none of the regulations violated by Arbour was a condition of payment. The First Circuit reversed in relevant part, holding that every submission of a claim implicitly represents compliance with relevant regulations, and that any undisclosed violation of a precondition of payment (whether or not expressly identified as such) renders a claim “false or fraudulent.” The First Circuit further held that the regulations themselves provided conclusive evidence that compliance was a material condition of payment because the regulations expressly required facilities to adequately supervise staff as a condition of payment. Held: 1. The implied false certification theory can be a basis for FCA liability when a defendant submitting a claim makes specific representations about the goods or services provided, but fails to disclose noncompliance with material statutory, regulatory, or contractual requirements that make those representations misleading with respect to those goods or services. Pp. 8–11. (a) The FCA does not define a “false” or “fraudulent” claim, so the Court turns to the principle that “absent other indication, ‘Congress intends to incorporate the well-settled meaning of the common-law terms it uses,’ ” Sekhar v. United States, 570 U. S. ___, ___. Under the common-law definition of “fraud,” the parties agree, certain misrepresentations by omission can give rise to FCA liability. Respondents and the Government contend that every claim for payment implicitly represents that the claimant is legally entitled to payment, and that failing to disclose violations of material legal requirements renders the claim misleading. Universal Health, on the other hand, argues that submitting a claim involves no representations and that the nondisclosure of legal violations is not actionable absent a special duty of reasonable care to disclose such matters. Today’s decision holds that the claims at issue may be actionable because they do more than merely demand payment; they fall squarely within the rule that representations that state the truth only so far as it goes, while omitting critical qualifying information, can be actionable misrepresentations. Pp. 8–10. (b) By submitting claims for payment using payment codes corresponding to specific counseling services, Universal Health represented that it had provided specific types of treatment. And Arbour staff allegedly made further representations by using National Provider Identification numbers corresponding to specific job titles. By conveying this information without disclosing Arbour’s many violations of basic staff and licensing requirements for mental health facilities, Universal Health’s claims constituted misrepresentations. Pp. 10–11. 2. Contrary to Universal Health’s contentions, FCA liability for failing to disclose violations of legal requirements does not turn upon whether those requirements were expressly designated as conditions of payment. Pp. 11–17. (a) Section 3729(a)(1)(A), which imposes liability on those presenting “false or fraudulent claim[s],” does not limit claims to misrepresentations about express conditions of payment. Nothing in the text supports such a restriction. And under the Act’s materiality requirement, statutory, regulatory, and contractual requirements are not automatically material, even if they are labeled conditions of payment. Nor is the restriction supported by the Act’s scienter requirement. A defendant can have “actual knowledge” that a condition is material even if the Government does not expressly call it a condition of payment. What matters is not the label that the Government attaches to a requirement, but whether the defendant knowingly violated a requirement that the defendant knows is material to the Government’s payment decision. Universal Health’s policy arguments are unavailing, and are amply addressed through strict enforcement of the FCA’s stringent materiality and scienter provisions. Pp. 12–14. (b) A misrepresentation about compliance with a statutory, regulatory, or contractual requirement must be material to the Government’s payment decision in order to be actionable under the FCA. The FCA’s materiality requirement is demanding. An undisclosed fact is material if, for instance, “[n]o one can say with reason that the plaintiff would have signed this contract if informed of the likelihood” of the undisclosed fact. Junius Constr. Co. v. Cohen, 257 N. Y. 393, 400, 178 N. E. 672, 674. When evaluating the FCA’s materiality requirement, the Government’s decision to expressly identify a provision as a condition of payment is relevant, but not automatically dispositive. A misrepresentation cannot be deemed material merely because the Government designates compliance with a particular requirement as a condition of payment. Nor is the Government’s option to decline to pay if it knew of the defendant’s noncompliance sufficient for a finding of materiality. Materiality also cannot be found where noncompliance is minor or insubstantial. Moreover, if the Government pays a particular claim in full despite its actual knowledge that certain requirements were violated, that is very strong evidence that those requirements are not material. The FCA thus does not support the Government’s and First Circuit’s expansive view that any statutory, regulatory, or contractual violation is material so long as the defendant knows that the Government would be entitled to refuse payment were it aware of the violation. Pp. 14–17. 780 F. 3d 504, vacated and remanded. Thomas, J., delivered the opinion for a unanimous Court.

The False Claims Act, 31 U. S. C. §3729 et seq., imposes significant penalties on those who defraud the Government. This case concerns a theory of False Claims Act liability commonly referred to as “implied false certification.” According to this theory, when a defendant submits a claim, it impliedly certifies compliance with all conditions of payment. But if that claim fails to disclose the defendant’s violation of a material statutory, regulatory, or contractual requirement, so the theory goes, the defendant has made a misrepresentation that renders the claim “false or fraudulent” under §3729(a)(1)(A). This case requires us to consider this theory of liability and to clarify some of the circumstances in which the False Claims Act imposes liability. We first hold that, at least in certain circumstances, the implied false certification theory can be a basis for liability. Specifically, liability can attach when the defendant submits a claim for payment that makes specific representations about the goods or services provided, but knowingly fails to disclose the defendant’s noncompliance with a statutory, regulatory, or contractual requirement. In these circumstances, liability may attach if the omission renders those representations misleading. We further hold that False Claims Act liability for failing to disclose violations of legal requirements does not turn upon whether those requirements were expressly designated as conditions of payment. Defendants can be liable for violating requirements even if they were not expressly designated as conditions of payment. Conversely, even when a requirement is expressly designated acondition of payment, not every violation of such a requirement gives rise to liability. What matters is not the label the Government attaches to a requirement, but whether the defendant knowingly violated a requirement that the defendant knows is material to the Government’s payment decision. A misrepresentation about compliance with a statutory, regulatory, or contractual requirement must be material to the Government’s payment decision in order to be actionable under the False Claims Act. We clarify below how that rigorous materiality requirement should be enforced. Because the courts below interpreted §3729(a)(1)(A) differently, we vacate the judgment and remand so that those courts may apply the approach set out in thisopinion. I A Enacted in 1863, the False Claims Act “was originally aimed principally at stopping the massive frauds perpetrated by large contractors during the Civil War.” United States v. Bornstein, 423 U. S. 303, 309 (1976) . “[A] series of sensational congressional investigations” prompted hearings where witnesses “painted a sordid picture of how the United States had been billed for nonexistent or worthless goods, charged exorbitant prices for goods delivered, and generally robbed in purchasing the necessities of war.” United States v. McNinch, 356 U. S. 595, 599 (1958) . Congress responded by imposing civil and criminal liability for 10 types of fraud on the Government, subjecting violators to double damages, forfeiture, and up to five years’ imprisonment. Act of Mar. 2, 1863, ch. 67, 12Stat. 696. Since then, Congress has repeatedly amended the Act, but its focus remains on those who present or directly induce the submission of false or fraudulent claims. See 31 U. S. C. §3729(a) (imposing civil liability on “any person who . . . knowingly presents, or causes to be presented, a false or fraudulent claim for payment or approval”). A “claim” now includes direct requests to the Government for payment as well as reimbursement requests made to the recipients of federal funds under federal benefits programs. See §3729(b)(2)(A). The Act’s scienter requirement defines “knowing” and “knowingly” to mean that a person has “actual knowledge of the information,” “acts in deliberate ignorance of the truth or falsity of the information,” or “acts in reckless disregard of the truth or falsity of the information.” §3729(b)(1)(A). And the Act defines “material” to mean “having a natural tendency to influence, or be capable of influencing, the payment or receipt of money or property.” §3729(b)(4). Congress also has increased the Act’s civil penalties so that liability is “essentially punitive in nature.” Vermont Agency of Natural Resources v. United States ex rel. Stevens, 529 U. S. 765, 784 (2000) . Defendants are subjected to treble damages plus civil penalties of up to $10,000 per false claim. §3729(a); 28 CFR §85.3(a)(9) (2015) (adjusting penalties for inflation). B The alleged False Claims Act violations here arose within the Medicaid program, a joint state-federal program in which healthcare providers serve poor or disabled patients and submit claims for government reimbursement. See generally 42 U. S. C. §1396 et seq. The facts recited in the complaint, which we take as true at this stage, are as follows. For five years, Yarushka Rivera, a teenage beneficiary of Massachusetts’ Medicaid program, received counseling services at Arbour Counseling Services, a satellite mental health facility in Lawrence, Massa-chusetts, owned and operated by a subsidiary of peti-tioner Universal Health Services. Beginning in 2004, when Yarushka started having behavioral problems, five medical professionals at Arbour intermittently treated her. In May 2009, Yarushka had an adverse reaction to a medication that a purported doctor at Arbour prescribed after diagnosing her with bipolar disorder. Her condition worsened; she suffered a seizure that required hospitalization. In October 2009, she suffered another seizure and died. She was 17 years old. Thereafter, an Arbour counselor revealed to respondents Carmen Correa and Julio Escobar—Yarushka’s mother and stepfather—that few Arbour employees were actually licensed to provide mental health counseling and that supervision of them was minimal. Respondents discovered that, of the five professionals who had treated Yarushka, only one was properly licensed. The practitioner who diagnosed Yarushka as bipolar identified herself as a psychologist with a Ph. D., but failed to mention that her degree came from an unaccredited Internet college and that Massachusetts had rejected her application to be licensed as a psychologist. Likewise, the practitioner who prescribed medicine to Yarushka, and who was held out as a psychiatrist, was in fact a nurse who lacked authority to prescribe medications absent supervision. Rather than ensuring supervision of unlicensed staff, the clinic’s director helped to misrepresent the staff’s qualifications. And the problem went beyond those who treated Yarushka. Some 23 Arbour employees lacked licenses to provide mental health services, yet—despite regulatory requirements to the contrary—they counseled patients and prescribed drugs without supervision. When submitting reimbursement claims, Arbour used payment codes corresponding to different services that its staff provided to Yaruskha, such as “Individual Therapy” and “family therapy.” 1 App. 19, 20. Staff members also misrepresented their qualifications and licensing status to the Federal Government to obtain individual National Provider Identification numbers, which are submitted in connection with Medicaid reimbursement claims and correspond to specific job titles. For instance, one Arbour staff member who treated Yaruskha registered for a number associated with “ ‘Social Worker, Clinical,’ ” despite lacking the credentials and licensing required for social workers engaged in mental health counseling. 1 id., at 32. After researching Arbour’s operations, respondents filed complaints with various Massachusetts agencies. Massachusetts investigated and ultimately issued a report detailing Arbour’s violation of over a dozen Massachusetts Medicaid regulations governing the qualifications and supervision required for staff at mental health facili-ties. Arbour agreed to a remedial plan, and two Arbour employees also entered into consent agreements with Massachusetts. In 2011, respondents filed a qui tam suit in federal court, see 31 U. S. C. §3730, alleging that Universal Health had violated the False Claims Act under an implied false certification theory of liability. The operative complaint asserts that Universal Health (acting through Arbour) submitted reimbursement claims that made representations about the specific services provided by specific types of professionals, but that failed to disclose serious violations of regulations pertaining to staff qualifications and licensing requirements for these services.[1] Specifically, the Massachusetts Medicaid program requires satellite facilities to have specific types of clinicians on staff, delineates licensing requirements for particular positions (like psychiatrists, social workers, and nurses), and details supervision requirements for other staff. See 130 Code Mass. Regs. §§429.422–424, 429.439 (2014). Universal Health allegedly flouted these regulations because Arbour employed unqualified, unlicensed, and unsupervised staff. The Massachusetts Medicaid program, unaware of these deficiencies, paid the claims. Universal Health thus allegedly defrauded the program, which would not have reimbursed the claims had it known that it was billed for mental health services that were performed by unlicensed and unsupervised staff. The United States declined to intervene. The District Court granted Universal Health’s motion to dismiss the complaint. Circuit precedent had previously embraced the implied false certification theory of liability. See, e.g., United States ex rel. Hutcheson v. Blackstone Medical, Inc., 647 F. 3d 377, 385–387 (CA1 2011). But the District Court held that respondents had failed to state a claim under that theory because, with one exception not relevant here, none of the regulations that Arbour violated was a condition of payment. See 2014 WL 1271757, *1, *6–*12 (D Mass., Mar. 26, 2014). The United States Court of Appeals for the First Circuit reversed in relevant part and remanded. 780 F. 3d 504, 517 (2015). The court observed that each time a billing party submits a claim, it “implicitly communicate[s] that it conformed to the relevant program requirements, such that it was entitled to payment.” Id., at 514, n. 14. To determine whether a claim is “false or fraudulent” based on such implicit communications, the court explained, it “asks simply whether the defendant, in submitting a claim for reimbursement, knowingly misrepresented compliance with a material precondition of payment.” Id., at 512. In the court’s view, a statutory, regulatory, or contractual requirement can be a condition of payment either by expressly identifying itself as such or by implication. Id., at 512–513. The court then held that Universal Health had violated Massachusetts Medicaid regulations that “clearly impose conditions of payment.” Id., at 513. The court further held that the regulations themselves “constitute[d] dispositive evidence of materiality,” because they identified adequate supervision as an “express and absolute” condition of payment and “repeated[ly] reference[d]” supervision. Id., at 514 (internal quotation marks omitted). We granted certiorari to resolve the disagreement among the Courts of Appeals over the validity and scope of the implied false certification theory of liability. 577 U. S. ___ (2015). The Seventh Circuit has rejected this theory, reasoning that only express (or affirmative) falsehoods can render a claim “false or fraudulent” under 31 U. S. C. §3729(a)(1)(A). United States v. Sanford-Brown, Ltd., 788 F. 3d 696, 711–712 (2015). Other courts have accepted the theory, but limit its application to cases where defendants fail to disclose violations of expressly designated conditions of payment. E.g., Mikes v. Straus, 274 F. 3d 687, 700 (CA2 2011). Yet others hold that conditions of payment need not be expressly designated as such to be a basis for False Claims Act liability. E.g., United States v. Science Applications Int’l Corp., 626 F. 3d 1257, 1269 (CADC 2010) (SAIC). II We first hold that the implied false certification theory can, at least in some circumstances, provide a basis for liability. By punishing defendants who submit “false or fraudulent claims,” the False Claims Act encompasses claims that make fraudulent misrepresentations, which include certain misleading omissions. When, as here, a defendant makes representations in submitting a claim but omits its violations of statutory, regulatory, or contractual requirements, those omissions can be a basis for liability if they render the defendant’s representations misleading with respect to the goods or services provided. To reach this conclusion, “[w]e start, as always, with the language of the statute.” Allison Engine Co. v. United States ex rel. Sanders, 553 U. S. 662, 668 (2008) (brackets in original; internal quotation marks omitted). The False Claims Act imposes civil liability on “any person who . . . knowingly presents, or causes to be presented, a falseor fraudulent claim for payment or approval.” §3729(a)(1)(A). Congress did not define what makes a claim “false” or “fraudulent.” But “[i]t is a settled principle of interpretation that, absent other indication, Congress intends to incorporate the well-settled meaning of the common-law terms it uses.” Sekhar v. United States, 570 U. S. ___, ___ (2013) (slip op., at 3) (internal quotation marks omitted). And the term “fraudulent” is a paradigmatic example of a statutory term that incorporates the common-law meaning of fraud. See Neder v. United States, 527 U. S. 1, 22 (1999) (the term “actionable ‘fraud’” is one with “a well-settled meaning at common law”).[2] Because common-law fraud has long encompassed certain misrepresentations by omission, “false or fraudulent claims” include more than just claims containing express falsehoods. The parties and the Government agree that misrepresentations by omission can give rise to liability. Brief for Petitioner 30–31; Brief for Respondents 22–31; Brief for United States as Amicus Curiae 16–20. The parties instead dispute whether submitting a claim without disclosing violations of statutory, regulatory, or contractual requirements constitutes such an actionable misrepresentation. Respondents and the Government invoke the common-law rule that, while nondisclosure alone ordinarily is not actionable, “[a] representation stating the truth so far as it goes but which the maker knows or believes to be materially misleading because of his failure to state additional or qualifying matter” is actionable. Restatement (Second) of Torts §529, p. 62 (1976). They contend that every submission of a claim for payment implicitly represents that the claimant is legally entitled to payment, and that failing to disclose violations of material legal requirements renders the claim misleading. Universal Health, on the other hand, argues that submitting a claim involves no representations, and that a different common-law rule thus governs: nondisclosure of legal violations is not actionable absent a special “ ‘duty . . . to exercise reasonable care to disclose the matter in question,’ ” which it says is lacking in Government contracting. Brief for Petitioner 31 (quoting Restatement (Second) of Torts §551(1), at 119). We need not resolve whether all claims for payment implicitly represent that the billing party is legally entitled to payment. The claims in this case do more than merely demand payment. They fall squarely within the rule that half-truths—representations that state the truth only so far as it goes, while omitting critical qualifying information—can be actionable misrepresentations.[3] A classic example of an actionable half-truth in contract law is the seller who reveals that there may be two new roads near a property he is selling, but fails to disclose that a third potential road might bisect the property. See Junius Constr. Co. v. Cohen, 257 N. Y. 393, 400, 178 N. E. 672, 674 (1931) (Cardozo, J.). “The enumeration of two streets, described as unopened but projected, was a tacit representation that the land to be conveyed was subject to no others, and certainly subject to no others materially affecting the value of the purchase.” Ibid. Likewise, an applicant for an adjunct position at a local college makes an actionable misrepresentation when his resume lists prior jobs and then retirement, but fails to disclose that his “retirement” was a prison stint for perpetrating a $12 million bank fraud. See 3 D. Dobbs, P. Hayden, & H. Bublick, Law of Torts §682, pp. 702–703, and n. 14 (2d ed. 2011) (citing Sarvis v. Vermont State Colleges, 172 Vt. 76, 78, 80–82, 772 A. 2d 494, 496, 497–499 (2001)). So too here, by submitting claims for payment using payment codes that corresponded to specific counseling services, Universal Health represented that it had pro-vided individual therapy, family therapy, preventive medica-tion counseling, and other types of treatment. Moreover, Arbour staff members allegedly made further representations in submitting Medicaid reimbursement claims by using National Provider Identification numbers corresponding to specific job titles. And these representations were clearly misleading in context. Anyone informed that a social worker at a Massachusetts mental health clinic provided a teenage patient with individual counseling services would probably—but wrongly—conclude that the clinic had complied with core Massachusetts Medicaid requirements (1) that a counselor “treating children [is] required to have specialized training and experience in children’s services,” 130 Code Mass. Regs. §429.422, and also (2) that, at a minimum, the social worker possesses the prescribed qualifications for the job, §429.424(C). By using payment and other codes that conveyed this information without disclosing Arbour’s many violations of basic staff and licensing requirements for mental health facilities, Universal Health’s claims constituted misrepresentations. Accordingly, we hold that the implied certification theory can be a basis for liability, at least where two conditions are satisfied: first, the claim does not merely request payment, but also makes specific representations about the goods or services provided; and second, the defendant’s failure to disclose noncompliance with material statutory, regulatory, or contractual requirements makes those representations misleading half-truths.[4] III The second question presented is whether, as Universal Health urges, a defendant should face False Claims Act liability only if it fails to disclose the violation of a contractual, statutory, or regulatory provision that the Government expressly designated a condition of payment. We conclude that the Act does not impose this limit on liability. But we also conclude that not every undisclosed violation of an express condition of payment automatically triggers liability. Whether a provision is labeled a condition of payment is relevant to but not dispositive of the materiality inquiry. A Nothing in the text of the False Claims Act sup-ports Universal Health’s proposed restriction. Section 3729(a)(1)(A) imposes liability on those who present “false or fraudulent claims” but does not limit such claims to misrepresentations about express conditions of payment. See SAIC, 626 F. 3d, at 1268 (rejecting any textual basis for an express-designation rule). Nor does the common-law meaning of fraud tether liability to violating an express condition of payment. A statement that misleadingly omits critical facts is a misrepresentation irrespective of whether the other party has expressly signaled the importance of the qualifying information. Supra, at 9–11. The False Claims Act’s materiality requirement also does not support Universal Health. Under the Act, the misrepresentation must be material to the other party’s course of action. But, as discussed below, see infra, at 15–17, statutory, regulatory, and contractual requirements are not automatically material, even if they are labeled conditions of payment. Cf. Matrixx Initiatives, Inc. v. Siracusano, 563 U. S. 27, 39 (2011) (materiality cannot rest on “a single fact or occurrence as always determinative” (internal quotation marks omitted)). Nor does the Act’s scienter requirement, §3729(b)(1)(A), support Universal Health’s position. A defendant can have “actual knowledge” that a condition is material without the Government expressly calling it a condition of payment. If the Government failed to specify that guns it orders must actually shoot, but the defendant knows that the Government routinely rescinds contracts if the guns do not shoot, the defendant has “actual knowledge.” Likewise, because a reasonable person would realize the imperative of a functioning firearm, a defendant’s failure to appreciate the materiality of that condition would amount to “deliberate ignorance” or “reckless disregard” of the “truth or falsity of the information” even if the Government did not spell this out. Universal Health nonetheless contends that False Claims Act liability should be limited to undisclosed violations of expressly designated conditions of payment to provide defendants with fair notice and to cabin liability. But policy arguments cannot supersede the clear statutory text. Kloeckner v. Solis, 568 U. S. ___, ___–___, n. 4 (2012) (slip op., at 13–14, n. 4). In any event, Universal Health’s approach risks undercutting these policy goals. The Government might respond by designating every legal requirement an express condition of payment. But billing parties are often subject to thousands of complex statutory and regulatory provisions. Facing False Claims Act liability for violating any of them would hardly help would-be defendants anticipate and prioritize compliance obligations. And forcing the Government to expressly designate a provision as a condition of payment would create further arbitrariness. Under Universal Health’s view, misrepresenting compliance with a requirement that the Government expressly identified as a condition of payment could expose a defendant to liability. Yet, under this theory, misrepresenting compliance with a condition of eligibility to even participate in a federal program when submitting a claim would not. Moreover, other parts of the False Claims Act allay Universal Health’s concerns. “[I]nstead of adopting a circumscribed view of what it means for a claim to be false or fraudulent,” concerns about fair notice and open-ended liability “can be effectively addressed through strict enforcement of the Act’s materiality and scienter requirements.” SAIC, supra, at 1270. Those requirements are rigorous. B As noted, a misrepresentation about compliance with a statutory, regulatory, or contractual requirement must be material to the Government’s payment decision in order to be actionable under the False Claims Act. We now clarify how that materiality requirement should be enforced. Section 3729(b)(4) defines materiality using language that we have employed to define materiality in other federal fraud statutes: “[T]he term ‘material’ means having a natural tendency to influence, or be capable of influencing, the payment or receipt of money or property.” See Neder, 527 U. S., at 16 (using this definition to interpret the mail, bank, and wire fraud statutes); Kungys v. United States, 485 U. S. 759, 770 (1988) (same for fraudulent statements to immigration officials). This materiality requirement descends from “common-law antecedents.” Id., at 769. Indeed, “the common law could not have conceived of ‘fraud’ without proof of materiality.” Neder, supra, at 22; see also Brief for United States as Amicus Curiae 30 (describing common-law principles and arguing that materiality under the False Claims Act should involve a “similar approach”). We need not decide whether §3729(a)(1)(A)’s materiality requirement is governed by §3729(b)(4) or derived directly from the common law. Under any understanding of the concept, materiality “look[s] to the effect on the likely or actual behavior of the recipient of the alleged misrepresentation.” 26 R. Lord, Williston on Contracts §69:12, p. 549 (4th ed. 2003) (Williston). In tort law, for instance, a “matter is material” in only two circumstances: (1) “[if ] a reasonable man would attach importance to [it] in determining his choice of action in the transaction”; or (2) if the defendant knew or had reason to know that the recipient of the representation attaches importance to the specific matter “in determining his choice of action,” even though a reasonable person would not. Restatement (Second) of Torts §538, at 80. Materiality in contract law is substantially similar. See Restatement (Second) of Contracts §162(2), and Comment c, pp. 439, 441 (1979) (“[A] misrepresentation is material” only if it would “likely . . . induce a reasonable person to manifest his assent,” or the defendant “knows that for some special reason [the representation] is likely to induce the particular recipient to manifest his assent” to the transaction).[5] The materiality standard is demanding. The False Claims Act is not “an all-purpose antifraud statute,” Allison Engine, 553 U. S., at 672, or a vehicle for punishing garden-variety breaches of contract or regulatory violations. A misrepresentation cannot be deemed material merely because the Government designates compliance with a particular statutory, regulatory, or contractual requirement as a condition of payment. Nor is it sufficient for a finding of materiality that the Government would have the option to decline to pay if it knew of the defendant’s noncompliance. Materiality, in addition, cannot be found where noncompliance is minor or insubstantial. See United States ex rel. Marcus v. Hess, 317 U. S. 537, 543 (1943) (contractors’ misrepresentation that they satisfied a non-collusive bidding requirement for federal program contracts violated the False Claims Act because “[t]he government’s money would never have been placed in the joint fund for payment to respondents had its agents known the bids were collusive”); see also Junius Constr., 257 N. Y., at 400, 178 N. E., at 674 (an undisclosed fact was material because “[n]o one can say with reason that the plaintiff would have signed this contract if informed of the likelihood” of the undisclosed fact). In sum, when evaluating materiality under the False Claims Act, the Government’s decision to expressly iden-tify a provision as a condition of payment is relevant, but not automatically dispositive. Likewise, proof of materiality can include, but is not necessarily limited to, evidence that the defendant knows that the Government consis-tently refuses to pay claims in the mine run of cases based on noncompliance with the particular statutory, regulatory, or contractual requirement. Conversely, if the Government pays a particular claim in full despite its actual knowledge that certain requirements were violated, that is very strong evidence that those requirements are not material. Or, if the Government regularly pays a particular type of claim in full despite actual knowledge that certain requirements were violated, and has signaled no change in position, that is strong evidence that the requirements are not material.[6] These rules lead us to disagree with the Government’s and First Circuit’s view of materiality: that any statutory, regulatory, or contractual violation is material so long as the defendant knows that the Government would be entitled to refuse payment were it aware of the violation. See Brief for United States as Amicus Curiae 30; Tr. of Oral Arg. 43 (Government’s “test” for materiality “is whether the person knew that the government could lawfully withhold payment”); 780 F. 3d, at 514; see also Tr. of Oral Arg. 26, 29 (statements by respondents’ counsel endorsing this view). At oral argument, the United States explained the implications of its position: If the Government contracts for health services and adds a requirement that contractors buy American-made staplers, anyone who submits a claim for those services but fails to disclose its use of foreign staplers violates the False Claims Act. To the Government, liability would attach if the defendant’s use of foreign staplers would entitle the Government not to pay the claim in whole or part—irrespective of whether the Government routinely pays claims despite knowing that foreign staplers were used. Id., at 39–45. Likewise, if the Government required contractors to aver their compliance with the entire U. S. Code and Code of Federal Regulations, then under this view, failing to mention noncompliance with any of those requirements would always be material. The False Claims Act does not adopt such an extraordinarily expansive view of liability. * * * Because both opinions below assessed respondents’ complaint based on interpretations of §3729(a)(1)(A) that differ from ours, we vacate the First Circuit’s judgment and remand the case for reconsideration of whether respondents have sufficiently pleaded a False Claims Act violation. See Omnicare, Inc. v. Laborers Dist. Council Constr. Industry Pension Fund, 575 U. S. ___, ___ (2015) (slip op., at 19). We emphasize, however, that the False Claims Act is not a means of imposing treble damages and other penalties for insignificant regulatory or contractual violations. This case centers on allegations of fraud, not medical malpractice. Respondents have alleged that Universal Health misrepresented its compliance with mental health facility requirements that are so central to the provision of mental health counseling that the Medicaid program would not have paid these claims had it known of these violations. Respondents may well have adequately pleaded a violation of §3729(a)(1)(A). But we leave it to the courts below to resolve this in the first instance. The judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Although Universal Health submitted some of the claims at issue before 2009, we assume—as the parties have done—that the 2009 amendments to the False Claims Act apply here. Universal Health does not argue, and we thus do not consider, whether pre-2009 conduct should be treated differently. 2 The False Claims Act abrogates the common law in certain respects. For instance, the Act’s scienter requirement “require[s] no proof of specific intent to defraud.” 31 U. S. C. §3729(b)(1)(B). But we presume that Congress retained all other elements of common-law fraud that are consistent with the statutory text because there are no textual indicia to the contrary. See Neder, 527 U. S., at 24–25. 3 This rule recurs throughout the common law. In tort law, for example, “if the defendant does speak, he must disclose enough to prevent his words from being misleading.” W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts §106, p. 738 (5th ed. 1984). Contract law also embraces this principle. See, e.g., Restatement (Second) of Contracts §161, Comment a, p. 432 (1979). And we have used this definition in other statutory contexts. See, e.g., Matrixx Initiatives, Inc. v. Siracusano, 563 U. S. 27, 44 (2011) (securities law). 4 As an alternative argument, Universal Health asserts that misleading partial disclosures constitute fraudulent misrepresentations only when the initial statement partially disclosed unfavorable information. Not so. “[A] statement that contains only favorable matters and omits all reference to unfavorable matters is as much a false representation as if all the facts stated were untrue.” Restatement (Second) of Torts, §529, Comment a, pp. 62–63 (1976). 5 Accord, Williston §69:12, pp. 549–550 (“most popular” understanding is “that a misrepresentation is material if it concerns a matter to which a reasonable person would attach importance in determining his or her choice of action with respect to the transaction involved: which will induce action by a complaining party[,] knowledge of which would have induced the recipient to act differently” (footnote omitted)); id., at 550 (noting rule that “a misrepresentation is material if, had it not been made, the party complaining of fraud would not have taken the action alleged to have been induced by the misrepresentation”); Junius Constr. Co. v. Cohen, 257 N. Y. 393, 400, 178 N. E. 672, 674 (1931) (a misrepresentation is material if it “went to the very essence of the bargain”); cf. Neder v. United States, 527 U. S. 1 , n. 5 (1999) (relying on “ ‘natural tendency to influence’ ” standard and citing Restatement (Second) of Torts §538 definition of materiality). 6 We reject Universal Health’s assertion that materiality is too fact intensive for courts to dismiss False Claims Act cases on a motion to dismiss or at summary judgment. The standard for materiality that we have outlined is a familiar and rigorous one. And False Claims Act plaintiffs must also plead their claims with plausibility and particular-ity under Federal Rules of Civil Procedure 8 and 9(b) by, for instance, pleading facts to support allegations of materiality.

579.US.2015_14-1373

Narcotics detective Douglas Fackrell conducted surveillance on a South Salt Lake City residence based on an anonymous tip about drug activity. The number of people he observed making brief visits to the house over the course of a week made him suspicious that the occupants were dealing drugs. After observing respondent Edward Strieff leave the residence, Officer Fackrell detained Strieff at a nearby parking lot, identifying himself and asking Strieff what he was doing at the house. He then requested Strieff’s identification and relayed the information to a police dispatcher, who informed him that Strieff had an outstanding arrest warrant for a traffic violation. Officer Fackrell arrested Strieff, searched him, and found methamphetamine and drug paraphernalia. Strieff moved to suppress the evidence, arguing that it was derived from an unlawful investigatory stop. The trial court denied the motion, and the Utah Court of Appeals affirmed. The Utah Supreme Court reversed, however, and ordered the evidence suppressed. Held: The evidence Officer Fackrell seized incident to Strieff’s arrest is admissible based on an application of the attenuation factors from Brown v. Illinois, 422 U. S. 590 . In this case, there was no flagrant police misconduct. Therefore, Officer Fackrell’s discovery of a valid, pre-existing, and untainted arrest warrant attenuated the connection between the unconstitutional investigatory stop and the evidence seized incident to a lawful arrest. Pp. 4–10. (a) As the primary judicial remedy for deterring Fourth Amendment violations, the exclusionary rule encompasses both the “primary evidence obtained as a direct result of an illegal search or seizure” and, relevant here, “evidence later discovered and found to be derivative of an illegality.” Segura v. United States, 468 U. S. 796 . But to ensure that those deterrence benefits are not outweighed by the rule’s substantial social costs, there are several exceptions to the rule. One exception is the attenuation doctrine, which provides for admissibility when the connection between unconstitutional police conduct and the evidence is sufficiently remote or has been interrupted by some intervening circumstance. See Hudson v. Michigan, 547 U. S. 586 . Pp. 4–5. (b) As a threshold matter, the attenuation doctrine is not limited to the defendant’s independent acts. The doctrine therefore applies here, where the intervening circumstance is the discovery of a valid, pre-existing, and untainted arrest warrant. Assuming, without deciding, that Officer Fackrell lacked reasonable suspicion to stop Strieff initially, the discovery of that arrest warrant attenuated the connection between the unlawful stop and the evidence seized from Strieff incident to his arrest. Pp. 5–10. (1) Three factors articulated in Brown v. Illinois, 422 U. S. 590 , lead to this conclusion. The first, “temporal proximity” between the initially unlawful stop and the search, id., at 603, favors suppressing the evidence. Officer Fackrell discovered drug contraband on Strieff only minutes after the illegal stop. In contrast, the second factor, “the presence of intervening circumstances, id., at 603–604, strongly favors the State. The existence of a valid warrant, predating the investigation and entirely unconnected with the stop, favors finding sufficient attenuation between the unlawful conduct and the discovery of evidence. That warrant authorized Officer Fackrell to arrest Strieff, and once the arrest was authorized, his search of Strieff incident to that arrest was undisputedly lawful. The third factor, “the purpose and flagrancy of the official misconduct,” id., at 604, also strongly favors the State. Officer Fackrell was at most negligent, but his errors in judgment hardly rise to a purposeful or flagrant violation of Strieff’s Fourth Amendment rights. After the unlawful stop, his conduct was lawful, and there is no indication that the stop was part of any systemic or recurrent police misconduct. Pp. 6–9. (2) Strieff’s counterarguments are unpersuasive. First, neither Officer Fackrell’s purpose nor the flagrancy of the violation rises to a level of misconduct warranting suppression. Officer Fackrell’s purpose was not to conduct a suspicionless fishing expedition but was to gather information about activity inside a house whose occupants were legitimately suspected of dealing drugs. Strieff conflates the standard for an illegal stop with the standard for flagrancy, which requires more than the mere absence of proper cause. Second, it is unlikely that the prevalence of outstanding warrants will lead to dragnet searches by police. Such misconduct would expose police to civil liability and, in any event, is already accounted for by Brown’s “purpose and flagrancy” factor. Pp. 9–10. 2015 UT 2, 357 P. 3d 532, reversed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Alito, JJ., joined. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, J., joined as to Parts I, II, and III. Kagan, J., filed a dissenting opinion, in which Ginsburg, J., joined.

To enforce the Fourth Amendment’s prohibition against “unreasonable searches and seizures,” this Court has at times required courts to exclude evidence obtained by unconstitutional police conduct. But the Court has also held that, even when there is a Fourth Amendment violation, this exclusionary rule does not apply when the costs of exclusion outweigh its deterrent benefits. In some cases, for example, the link between the unconstitutional conduct and the discovery of the evidence is too attenuated to justify suppression. The question in this case is whether this attenuation doctrine applies when an officer makesan unconstitutional investigatory stop; learns during that stop that the suspect is subject to a valid arrest warrant; and proceeds to arrest the suspect and seize incriminating evidence during a search incident to that arrest. We hold that the evidence the officer seized as part of the search incident to arrest is admissible because the officer’s discovery of the arrest warrant attenuated the connection between the unlawful stop and the evidence seized incident to arrest.IThis case began with an anonymous tip. In December 2006, someone called the South Salt Lake City police’s drug-tip line to report “narcotics activity” at a particular residence. App. 15. Narcotics detective Douglas Fackrell investigated the tip. Over the course of about a week, Officer Fackrell conducted intermittent surveillance of the home. He observed visitors who left a few minutes after arriving at the house. These visits were sufficiently frequent to raise his suspicion that the occupants were dealing drugs.One of those visitors was respondent Edward Strieff. Officer Fackrell observed Strieff exit the house and walk toward a nearby convenience store. In the store’s parking lot, Officer Fackrell detained Strieff, identified himself, and asked Strieff what he was doing at the residence.As part of the stop, Officer Fackrell requested Strieff’s identification, and Strieff produced his Utah identification card. Officer Fackrell relayed Strieff’s information to a police dispatcher, who reported that Strieff had an outstanding arrest warrant for a traffic violation. Officer Fackrell then arrested Strieff pursuant to that warrant. When Officer Fackrell searched Strieff incident to the arrest, he discovered a baggie of methamphetamine and drug paraphernalia.The State charged Strieff with unlawful possession of methamphetamine and drug paraphernalia. Strieff moved to suppress the evidence, arguing that the evidence was inadmissible because it was derived from an unlawful investigatory stop. At the suppression hearing, the prosecutor conceded that Officer Fackrell lacked reasonable suspicion for the stop but argued that the evidence should not be suppressed because the existence of a valid arrest warrant attenuated the connection between the unlawful stop and the discovery of the contraband.The trial court agreed with the State and admitted the evidence. The court found that the short time between the illegal stop and the search weighed in favor of suppressing the evidence, but that two countervailing considerations made it admissible. First, the court considered the presence of a valid arrest warrant to be an “ ‘extraordinary intervening circumstance.’ ” App. to Pet. for Cert. 102 (quoting United States v. Simpson, 439 F. 3d 490, 496 (CA8 2006). Second, the court stressed the absence of flagrant misconduct by Officer Fackrell, who was conducting a legitimate investigation of a suspected drug house.Strieff conditionally pleaded guilty to reduced charges of attempted possession of a controlled substance and possession of drug paraphernalia, but reserved his right to appeal the trial court’s denial of the suppression motion. The Utah Court of Appeals affirmed. 2012 UT App 245, 286 P. 3d 317.The Utah Supreme Court reversed. 2015 UT 2, 357 P. 3d 532. It held that the evidence was inadmissible because only “a voluntary act of a defendant’s free will (as in a confession or consent to search)” sufficiently breaks the connection between an illegal search and the discovery of evidence. Id., at 536. Because Officer Fackrell’s discovery of a valid arrest warrant did not fit this description, the court ordered the evidence suppressed. Ibid.We granted certiorari to resolve disagreement about how the attenuation doctrine applies where an unconstitutional detention leads to the discovery of a valid arrest warrant. 576 U. S. ___ (2015). Compare, e.g., United States v. Green, 111 F. 3d 515, 522–523 (CA7 1997) (holding that discovery of the warrant is a dispositive intervening circumstance where police misconduct was not flagrant), with, e.g., State v. Moralez, 297 Kan. 397, 415, 300 P. 3d 1090, 1102 (2013) (assigning little significance to the discovery of the warrant). We now reverse.IIAThe Fourth Amendment protects “[t]he right of the people to be secure in their persons, houses, papers, and effects, against unreasonable searches and seizures.” Because officers who violated the Fourth Amendment were traditionally considered trespassers, individuals subject to unconstitutional searches or seizures histori-cally enforced their rights through tort suits or self-help. Davies, Recovering the Original Fourth Amendment, 98 Mich. L. Rev. 547, 625 (1999). In the 20th century, however, the exclusionary rule—the rule that often requires trial courts to exclude unlawfully seized evidence in a criminal trial—became the principal judicial remedy to deter Fourth Amendment violations. See, e.g., Mapp v. Ohio,367 U. S. 643,655 (1961).Under the Court’s precedents, the exclusionary rule encompasses both the “primary evidence obtained as a direct result of an illegal search or seizure” and, relevant here, “evidence later discovered and found to be derivative of an illegality,” the so-called “ ‘fruit of the poisonous tree.’ ” Segura v. United States,468 U. S. 796,804 (1984). But the significant costs of this rule have led us to deem it “applicable only . . . where its deterrence benefits outweigh its substantial social costs.” Hudson v. Michigan,547 U. S. 586,591 (2006) (internal quotation marks omitted). “Suppression of evidence . . . has always been our last resort, not our first impulse.” Ibid.We have accordingly recognized several exceptions to the rule. Three of these exceptions involve the causal relationship between the unconstitutional act and the discovery of evidence. First, the independent source doctrine allows trial courts to admit evidence obtained in an unlawful search if officers independently acquired it from a separate, independent source. See Murray v. United States,487 U. S. 533,537 (1988). Second, the inevitable discovery doctrine allows for the admission of evidence that would have been discovered even without the unconstitutional source. See Nix v. Williams,467 U. S. 431–444 (1984). Third, and at issue here, is the attenuation doctrine: Evidence is admissible when the connection between unconstitutional police conduct and the evidence is remote or has been interrupted by some intervening circumstance, so that “the interest protected by the constitutional guarantee that has been violated would not be served by suppression of the evidence obtained.” Hudson, supra, at 593.BTurning to the application of the attenuation doctrine to this case, we first address a threshold question: whether this doctrine applies at all to a case like this, where the intervening circumstance that the State relies on is the discovery of a valid, pre-existing, and untainted arrest warrant. The Utah Supreme Court declined to apply the attenuation doctrine because it read our precedents as applying the doctrine only “to circumstances involving an independent act of a defendant’s ‘free will’ in confessing to a crime or consenting to a search.” 357 P. 3d, at 544. In this Court, Strieff has not defended this argument, and we disagree with it, as well. The attenuation doctrine evaluates the causal link between the government’s unlawful act and the discovery of evidence, which often has nothing to do with a defendant’s actions. And the logic of our prior attenuation cases is not limited to independent acts by the defendant.It remains for us to address whether the discovery of a valid arrest warrant was a sufficient intervening event to break the causal chain between the unlawful stop and the discovery of drug-related evidence on Strieff’s person. The three factors articulated in Brown v. Illinois,422 U. S. 590 (1975), guide our analysis. First, we look to the “temporal proximity” between the unconstitutional conduct and the discovery of evidence to determine how closely the discovery of evidence followed the unconstitutional search. Id., at 603. Second, we consider “the presence of intervening circumstances.” Id., at 603–604. Third, and “particularly” significant, we examine “the purpose and flagrancy of the official misconduct.” Id., at 604. In evaluating these factors, we assume without deciding (because the State conceded the point) that Officer Fackrell lacked reasonable suspicion to initially stop Strieff. And, because we ultimately conclude that the warrant breaks the causal chain, we also have no need to decide whether the warrant’s existence alone would make the initial stop constitutional even if Officer Fackrell was unaware of its existence.1The first factor, temporal proximity between the ini-tially unlawful stop and the search, favors suppressing the evidence. Our precedents have declined to find that this factor favors attenuation unless “substantial time” elapses between an unlawful act and when the evidence is obtained. Kaupp v. Texas,538 U. S. 626,633 (2003) (per curiam). Here, however, Officer Fackrell discovered drug contraband on Strieff’s person only minutes after the illegal stop. See App. 18–19. As the Court explained in Brown, such a short time interval counsels in favor of suppression; there, we found that the confession should be suppressed, relying in part on the “less than two hours” that separated the unconstitutional arrest and the confession. 422 U. S., at 604.In contrast, the second factor, the presence of intervening circumstances, strongly favors the State. In Segura,468 U. S. 796, the Court addressed similar facts to those here and found sufficient intervening circumstances to allow the admission of evidence. There, agents had probable cause to believe that apartment occupants were dealing cocaine. Id., at 799–800. They sought a warrant. In the meantime, they entered the apartment, arrested an occupant, and discovered evidence of drug activity during a limited search for security reasons. Id., at 800–801. The next evening, the Magistrate Judge issued the search warrant. Ibid. This Court deemed the evidence admissible notwithstanding the illegal search because the information supporting the warrant was “wholly unconnected with the [arguably illegal] entry and was known to the agents well before the initial entry.” Id., at 814.Segura, of course, applied the independent source doctrine because the unlawful entry “did not contribute in any way to discovery of the evidence seized under the warrant.” Id., at 815. But the Segura Court suggested that the existence of a valid warrant favors finding that the connection between unlawful conduct and the discovery of evidence is “sufficiently attenuated to dissipate the taint.” Ibid. That principle applies here.In this case, the warrant was valid, it predated Officer Fackrell’s investigation, and it was entirely unconnected with the stop. And once Officer Fackrell discovered the warrant, he had an obligation to arrest Strieff. “A warrant is a judicial mandate to an officer to conduct a search or make an arrest, and the officer has a sworn duty to carry out its provisions.” United States v. Leon,468 U. S. 897,920, n.21 (1984) (internal quotation marks omitted). Officer Fackrell’s arrest of Strieff thus was a ministerial act that was independently compelled by the pre-existing warrant. And once Officer Fackrell was authorized to arrest Strieff, it was undisputedly lawful to search Strieff as an incident of his arrest to protect Officer Fackrell’s safety. See Arizona v. Gant,556 U. S. 332,339 (2009) (explaining the permissible scope of searches incident to arrest).Finally, the third factor, “the purpose and flagrancy of the official misconduct,” Brown, supra, at 604, also strongly favors the State. The exclusionary rule exists to deter police misconduct. Davis v. United States,564 U. S. 229–237 (2011). The third factor of the attenuation doctrine reflects that rationale by favoring exclusion only when the police misconduct is most in need of deterrence—that is, when it is purposeful or flagrant.Officer Fackrell was at most negligent. In stopping Strieff, Officer Fackrell made two good-faith mistakes. First, he had not observed what time Strieff entered the suspected drug house, so he did not know how long Strieff had been there. Officer Fackrell thus lacked a sufficient basis to conclude that Strieff was a short-term visitor who may have been consummating a drug transaction. Second, because he lacked confirmation that Strieff was a short-term visitor, Officer Fackrell should have asked Strieff whether he would speak with him, instead of demanding that Strieff do so. Officer Fackrell’s stated purpose was to “find out what was going on [in] the house.” App. 17. Nothing prevented him from approaching Strieff simply to ask. See Florida v. Bostick,501 U. S. 429,434 (1991) (“[A] seizure does not occur simply because a police officer approaches an individual and asks a few questions”). But these errors in judgment hardly rise to a purposeful or flagrant violation of Strieff’s Fourth Amendment rights.While Officer Fackrell’s decision to initiate the stop was mistaken, his conduct thereafter was lawful. The officer’s decision to run the warrant check was a “negligibly burdensome precautio[n]” for officer safety. Rodriguez v. United States, 575 U. S. ___, ___ (2015) (slip op., at 7). And Officer Fackrell’s actual search of Strieff was a lawful search incident to arrest. See Gant, supra, at 339.Moreover, there is no indication that this unlawful stop was part of any systemic or recurrent police misconduct. To the contrary, all the evidence suggests that the stop was an isolated instance of negligence that occurred in connection with a bona fide investigation of a suspected drug house. Officer Fackrell saw Strieff leave a suspected drug house. And his suspicion about the house was based on an anonymous tip and his personal observations.Applying these factors, we hold that the evidence discovered on Strieff’s person was admissible because the unlawful stop was sufficiently attenuated by the pre-existing arrest warrant. Although the illegal stop was close in time to Strieff’s arrest, that consideration is outweighed by two factors supporting the State. The outstanding arrest warrant for Strieff’s arrest is a critical intervening circumstance that is wholly independent of the illegal stop. The discovery of that warrant broke the causal chain between the unconstitutional stop and the discovery of evidence by compelling Officer Fackrell to arrest Strieff. And, it is especially significant that there is no evidence that Officer Fackrell’s illegal stop reflected flagrantly unlawful police misconduct.2We find Strieff’s counterarguments unpersuasive.First, he argues that the attenuation doctrine should not apply because the officer’s stop was purposeful and flagrant. He asserts that Officer Fackrell stopped him solely to fish for evidence of suspected wrongdoing. But Officer Fackrell sought information from Strieff to find out what was happening inside a house whose occupants were legitimately suspected of dealing drugs. This was not a suspicionless fishing expedition “in the hope that something would turn up.” Taylor v. Alabama,457 U. S. 687,691 (1982).Strieff argues, moreover, that Officer Fackrell’s conduct was flagrant because he detained Strieff without the necessary level of cause (here, reasonable suspicion). But that conflates the standard for an illegal stop with the standard for flagrancy. For the violation to be flagrant, more severe police misconduct is required than the mere absence of proper cause for the seizure. See, e.g., Kaupp, 538 U. S., at 628, 633 (finding flagrant violation where a warrantless arrest was made in the arrestee’s home after police were denied a warrant and at least some officers knew they lacked probable cause). Neither the officer’s alleged purpose nor the flagrancy of the violation rise to a level of misconduct to warrant suppression.Second, Strieff argues that, because of the prevalence of outstanding arrest warrants in many jurisdictions, police will engage in dragnet searches if the exclusionary rule is not applied. We think that this outcome is unlikely. Such wanton conduct would expose police to civil liability. See42 U. S. C. §1983; Monell v. New York City Dept. of Social Servs.,436 U. S. 658,690 (1978); see also Segura, 468 U. S., at 812. And in any event, the Brown factors take account of the purpose and flagrancy of police misconduct. Were evidence of a dragnet search presented here, the application of the Brown factors could be different. But there is no evidence that the concerns that Strieff raises with the criminal justice system are present in South Salt Lake City, Utah.* * *We hold that the evidence Officer Fackrell seized as part of his search incident to arrest is admissible because his discovery of the arrest warrant attenuated the connection between the unlawful stop and the evidence seized from Strieff incident to arrest. The judgment of the Utah Supreme Court, accordingly, is reversed.It is so ordered.

579.US.2015_14-10154

In an effort to “close [a] dangerous loophole” in the gun control laws, United States v. Castleman, 572 U. S. ___, ___, Congress extended the federal prohibition on firearms possession by convicted felons to persons convicted of a “misdemeanor crime of domestic violence,” 18 U. S. C. §922(g)(9). Section 921(a)(33)(A) defines that phrase to include a misdemeanor under federal, state, or tribal law, committed against a domestic relation that necessarily involves the “use . . . of physical force.” In Castleman, this Court held that a knowing or intentional assault qualifies as such a crime, but left open whether the same was true of a reckless assault. Petitioner Stephen Voisine pleaded guilty to assaulting his girlfriend in violation of §207 of the Maine Criminal Code, which makes it a misdemeanor to “intentionally, knowingly or recklessly cause[ ] bodily injury” to another. When law enforcement officials later investigated Voisine for killing a bald eagle, they learned that he owned a rifle. After a background check turned up Voisine’s prior conviction under §207, the Government charged him with violating §922(g)(9). Petitioner William Armstrong pleaded guilty to assaulting his wife in violation of a Maine domestic violence law making it a misdemeanor to commit an assault prohibited by §207 against a family or household member. While searching Armstrong’s home as part of a narcotics investigation a few years later, law enforcement officers discovered six guns and a large quantity of ammunition. Armstrong was also charged under §922(g)(9). Both men argued that they were not subject to §922(g)(9)’s prohibition because their prior convictions could have been based on reckless, rather than knowing or intentional, conduct and thus did not quality as misdemeanor crimes of domestic violence. The District Court rejected those claims, and each petitioner pleaded guilty. The First Circuit affirmed, holding that “an offense with a mens rea of recklessness may qualify as a ‘misdemeanor crime of violence’ under §922(g)(9).” Voisine and Armstrong filed a joint petition for certiorari, and their case was remanded for further consideration in light of Castleman. The First Circuit again upheld the convictions on the same ground. Held: A reckless domestic assault qualifies as a “misdemeanor crime of domestic violence” under §922(g)(9). Pp. 4–12. (a) That conclusion follows from the statutory text. Nothing in the phrase “use. . . of physical force” indicates that §922(g)(9) distinguishes between domestic assaults committed knowingly or intentionally and those committed recklessly. Dictionaries consistently define the word “use” to mean the “act of employing” something. Accordingly, the force involved in a qualifying assault must be volitional; an involuntary motion, even a powerful one, is not naturally described as an active employment of force. See Castleman, 572 U. S., at ___. But nothing about the definition of “use” demands that the person applying force have the purpose or practical certainty that it will cause harm, as compared with the understanding that it is substantially likely to do so. Nor does Leocal v. Ashcroft, 543 U. S. 1 , which held that the “use” of force excludes accidents. Reckless conduct, which requires the conscious disregard of a known risk, is not an accident: It involves a deliberate decision to endanger another. The relevant text thus supports prohibiting petitioners, and others with similar criminal records, from possessing firearms. Pp. 5–8. (b) So too does the relevant history. Congress enacted §922(g)(9) in 1996 to bar those domestic abusers convicted of garden-variety assault or battery misdemeanors—just like those convicted of felonies—from owning guns. Then, as now, a significant majority of jurisdictions—34 States plus the District of Columbia—defined such misdemeanor offenses to include the reckless infliction of bodily harm. In targeting those laws, Congress thus must have known it was sweeping in some persons who had engaged in reckless conduct. See, e.g., United States v. Bailey, 9 Pet. 238, 256. Indeed, that was part of the point: to apply the federal firearms restriction to those abusers, along with all others, covered by the States’ ordinary misdemeanor assault laws. Petitioners’ reading risks rendering §922(g)(9) broadly inoperative in the 35 jurisdictions with assault laws extending to recklessness. Consider Maine’s law, which criminalizes “intentionally, knowingly or recklessly” injuring another. Assuming that statute defines a single crime, petitioners’ view that §921(a)(33)(A) requires at least a knowing mens rea would mean that no conviction obtained under that law could qualify as a “misdemeanor crime of domestic violence.” Descamps v. United States, 570 U. S. ___, ___. In Castleman, the Court declined to construe §921(a)(33)(A) so as to render §922(g)(9) ineffective in 10 States. All the more so here, where petitioners’ view would jeopardize §922(g)(9)’s force in several times that many. Pp. 8–11. 778 F. 3d 176, affirmed. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Alito, JJ., joined. Thomas, J., filed a dissenting opinion, in which Sotomayor, J., joined as to Parts I and II.

Federal law prohibits any person convicted of a “misdemeanor crime of domestic violence” from possessing a firearm. 18 U. S. C. §922(g)(9). That phrase is defined to include any misdemeanor committed against a domestic relation that necessarily involves the “use . . . of physical force.” §921(a)(33)(A). The question presented here is whether misdemeanor assault convictions for reckless (as contrasted to knowing or intentional) conduct trigger the statutory firearms ban. We hold that they do. I Congress enacted §922(g)(9) some 20 years ago to “close [a] dangerous loophole” in the gun control laws. United States v. Castleman, 572 U. S. ___, ___ (2014) (slip op., at 2) (quoting United States v. Hayes, 555 U. S. 415, 426 (2009) ). An existing provision already barred convicted felons from possessing firearms. See §922(g)(1) (1994 ed.). But many perpetrators of domestic violence are charged with misdemeanors rather than felonies, notwithstanding the harmfulness of their conduct. See Castleman, 572 U. S., at ___ (slip op., at 2). And “[f]irearms and domestic strife are a potentially deadly combination.” Hayes, 555 U. S., at 427. Accordingly, Congress added §922(g)(9) to prohibit any person convicted of a “misdemeanor crime of domestic violence” from possessing any gun or ammunition with a connection to interstate commerce. And it defined that phrase, in §921(a)(33)(A), to include a misdemeanor under federal, state, or tribal law, committed by a person with a specified domestic relationship with the victim, that “has, as an element, the use or attempted use of physical force.” Two Terms ago, this Court considered the scope of that definition in a case involving a conviction for a knowing or intentional assault. See Castleman, 572 U. S., at ___–___ (slip op., at 4–13). In Castleman, we initially held that the word “force” in §921(a)(33)(A) bears its common-law meaning, and so is broad enough to include offensive touching. See id., at ___ (slip op., at 4). We then determined that “the knowing or intentional application of [such] force is a ‘use’ of force.” Id., at ___ (slip op., at 13). But we expressly left open whether a reckless assault also qualifies as a “use” of force—so that a misdemeanor conviction for such conduct would trigger §922(g)(9)’s firearms ban. See id., at ___, n. 8 (slip op., at 11, n. 8). The two cases before us now raise that issue. Petitioner Stephen Voisine pleaded guilty in 2004 to assaulting his girlfriend in violation of §207 of the Maine Criminal Code, which makes it a misdemeanor to “intentionally, knowingly or recklessly cause[ ] bodily injury or offensive physical contact to another person.” Me. Rev. Stat. Ann., Tit. 17–A, §207(1)(A). Several years later, Voisine again found himself in legal trouble, this time for killing a bald eagle. See 16 U. S. C. §668(a). While investigating that crime, law enforcement officers learned that Voisine owned a rifle. When a background check turned up his prior misdemeanor conviction, the Government charged him with violating 18 U. S. C. §922(g)(9).[1] Petitioner William Armstrong pleaded guilty in 2008 to assaulting his wife in violation of a Maine domestic violence law making it a misdemeanor to commit an assault prohibited by §207 (the general statute under which Voisine was convicted) against a family or household member. See Me. Rev. Stat. Ann., Tit. 17–A, §207–A(1)(A). A few years later, law enforcement officers searched Armstrong’s home as part of a narcotics investigation. They discovered six guns, plus a large quantity of ammunition. Like Voisine, Armstrong was charged under §922(g)(9) for unlawfully possessing firearms. Both men argued that they were not subject to §922(g)(9)’s prohibition because their prior convictions (as the Government conceded) could have been based on reckless, rather than knowing or intentional, conduct. The District Court rejected those claims. Each petitioner then entered a guilty plea conditioned on the right to appeal the District Court’s ruling. The Court of Appeals for the First Circuit affirmed the two convictions, holding that “an offense with a mens rea of recklessness may qualify as a ‘misdemeanor crime of violence’ under §922(g)(9).” United States v. Armstrong, 706 F. 3d 1, 4 (2013); see United States v. Voisine, 495 Fed. Appx. 101, 102 (2013) (per curiam). Voisine and Armstrong filed a joint petition for certiorari, and shortly after issuing Castleman, this Court (without opinion) vacated the First Circuit’s judgments and remanded the cases for further consideration in light of that decision. See Armstrong v. United States, 572 U. S. ___ (2014). On remand, the Court of Appeals again upheld the convictions, on the same ground. See 778 F. 3d 176, 177 (2015). We granted certiorari, 577 U. S. ___ (2015), to resolve a Circuit split over whether a misdemeanor conviction for recklessly assaulting a domestic relation disqualifies an individual from possessing a gun under §922(g)(9).[2] We now affirm. II The issue before us is whether §922(g)(9) applies to reckless assaults, as it does to knowing or intentional ones. To commit an assault recklessly is to take that action with a certain state of mind (or mens rea)—in the dominant formulation, to “consciously disregard[ ]” a substantial risk that the conduct will cause harm to another. ALI, Model Penal Code §2.02(2)(c) (1962); Me. Rev. Stat. Ann., Tit. 17–A, §35(3) (Supp. 2015) (adopting that definition); see Farmer v. Brennan, 511 U. S. 825 –837 (1994) (noting that a person acts recklessly only when he disregards a substantial risk of harm “of which he is aware”). For purposes of comparison, to commit an assault knowingly or intentionally (the latter, to add yet another adverb, sometimes called “purposefully”) is to act with another state of mind respecting that act’s consequences—in the first case, to be “aware that [harm] is practically certain” and, in the second, to have that result as a “conscious object.” Model Penal Code §§2.02 (2)(a)–(b); Me. Rev. Stat. Ann., Tit. 17–A, §§35(1)–(2). Statutory text and background alike lead us to conclude that a reckless domestic assault qualifies as a “misdemeanor crime of domestic violence” under §922(g)(9). Congress defined that phrase to include crimes that necessarily involve the “use . . . of physical force.” §921(a)(33)(A). Reckless assaults, no less than the knowing or intentional ones we addressed in Castleman, satisfy that definition. Further, Congress enacted §922(g)(9) in order to prohibit domestic abusers convicted under run-of-the-mill misdemeanor assault and battery laws from possessing guns. Because fully two-thirds of such state laws extend to recklessness, construing §922(g)(9) to exclude crimes committed with that state of mind would substantially undermine the provision’s design. A Nothing in the word “use”—which is the only statutory language either party thinks relevant—indicates that §922(g)(9) applies exclusively to knowing or intentional domestic assaults. Recall that under §921(a)(33)(A), an offense counts as a “misdemeanor crime of domestic violence” only if it has, as an element, the “use” of force. Dictionaries consistently define the noun “use” to mean the “act of employing” something. Webster’s New International Dictionary 2806 (2d ed. 1954) (“[a]ct of employing anything”); Random House Dictionary of the English Language 2097 (2d ed. 1987) (“act of employing, using, or putting into service”); Black’s Law Dictionary 1541 (6th ed. 1990) (“[a]ct of employing,” “application”).[3] On that common understanding, the force involved in a qualifying assault must be volitional; an involuntary motion, even a powerful one, is not naturally described as an active employment of force. See Castleman, 572 U. S., at ___ (slip op., at 13) (“[T]he word ‘use’ conveys the idea that the thing used (here, ‘physical force’) has been made the user’s instrument” (some internal quotation marks omitted)). But the word “use” does not demand that the person applying force have the purpose or practical certainty that it will cause harm, as compared with the understanding that it is substantially likely to do so. Or, otherwise said, that word is indifferent as to whether the actor has the mental state of intention, knowledge, or recklessness with respect to the harmful consequences of his volitional conduct. Consider a couple of examples to see the ordinary meaning of the word “use” in this context. If a person with soapy hands loses his grip on a plate, which then shatters and cuts his wife, the person has not “use[d]” physical force in common parlance. But now suppose a person throws a plate in anger against the wall near where his wife is standing. That hurl counts as a “use” of force even if the husband did not know for certain (or have as an object), but only recognized a substantial risk, that a shard from the plate would ricochet and injure his wife. Similarly, to spin out a scenario discussed at oral argument, if a person lets slip a door that he is trying to hold open for his girlfriend, he has not actively employed (“used”) force even though the result is to hurt her. But if he slams the door shut with his girlfriend following close behind, then he has done so—regardless of whether he thinks it absolutely sure or only quite likely that he will catch her fingers in the jamb. See Tr. of Oral Arg. 10–11 (counsel for petitioners acknowledging that this example involves “the use of physical force”). Once again, the word “use” does not exclude from §922(g)(9)’s compass an act of force carried out in conscious disregard of its substantial risk of causing harm. And contrary to petitioners’ view, nothing in Leocal v. Ashcroft, 543 U. S. 1 (2004) , suggests a different conclusion—i.e., that “use” marks a dividing line between reckless and knowing conduct. See Brief for Petitioners 18–22. In that decision, this Court addressed a statutory definition similar to §921(a)(33)(A): there, “the use . . . of physical force against the person or property of another.” 18 U. S. C. §16. That provision excludes “merely accidental” conduct, Leocal held, because “it is [not] natural to say that a person actively employs physical force against another person by accident.” 543 U. S., at 9. For example, the Court stated, one “would not ordinarily say a person ‘use[s] . . . physical force against’ another by stumbling and falling into him.” Ibid. That reasoning fully accords with our analysis here. Conduct like stumbling (or in our hypothetical, dropping a plate) is a true accident, and so too the injury arising from it; hence the difficulty of describing that conduct as the “active employment” of force. Ibid. But the same is not true of reckless behavior—acts undertaken with awareness of their substantial risk of causing injury (in our contrasting hypo, hurling the plate). The harm such conduct causes is the result of a deliberate decision to endanger another—no more an “accident” than if the “substantial risk” were “practically certain.” See supra, at 4 (comparing reckless and knowing acts). And indeed, Leocal itself recognized the distinction between accidents and recklessness, specifically reserving the issue whether the definition in §16 embraces reckless conduct, see 543 U. S., at 13—as we now hold §921(a)(33)(A) does.[4] In sum, Congress’s definition of a “misdemeanor crime of violence” contains no exclusion for convictions based on reckless behavior. A person who assaults another recklessly “use[s]” force, no less than one who carries out that same action knowingly or intentionally. The relevant text thus supports prohibiting petitioners, and others with similar criminal records, from possessing firearms. B So too does the relevant history. As explained earlier, Congress enacted §922(g)(9) in 1996 to bar those domestic abusers convicted of garden-variety assault or battery misdemeanors—just like those convicted of felonies—from owning guns. See supra, at 1–2; Castleman, 572 U. S., at ___, ___ (slip op., at 2, 6); Hayes, 555 U. S., at 426–427. Then, as now, a significant majority of jurisdictions—34 States plus the District of Columbia—defined such misdemeanor offenses to include the reckless infliction of bodily harm. See Brief for United States 7a–19a (collecting statutes). That agreement was no coincidence. Sev-eral decades earlier, the Model Penal Code had taken the position that a mens rea of recklessness should generally suffice to establish criminal liability, including for assault. See §2.02(3), Comments 4–5, at 243–244 (“purpose, knowledge, and recklessness are properly the basis for” such liability); §211.1 (defining assault to include “purposely, knowingly, or recklessly caus[ing] bodily injury”). States quickly incorporated that view into their misdemeanor assault and battery statutes. So in linking §922(g)(9) to those laws, Congress must have known it was sweeping in some persons who had engaged in reckless conduct. See, e.g., United States v. Bailey, 9 Pet. 238, 256 (1835) (Story, J.) (“Congress must be presumed to have legislated under this known state of the laws”). And indeed, that was part of the point: to apply firearms restrictions to those abusers, along with all others, whom the States’ ordinary misdemeanor assault laws covered. What is more, petitioners’ reading risks rendering §922(g)(9) broadly inoperative in the 35 jurisdictions with assault laws extending to recklessness—that is, inapplicable even to persons who commit that crime knowingly or intentionally. Consider Maine’s statute, which (in typical fashion) makes it a misdemeanor to “intentionally, knowingly or recklessly” injure another. Me. Rev. Stat. Ann., Tit. 17–A, §207(1)(A). Assuming that provision defines a single crime (which happens to list alternative mental states)—and accepting petitioners’ view that §921(a)(33)(A) requires at least a knowing mens rea—then, under Descamps v. United States, 570 U. S. ___ (2013), no conviction obtained under Maine’s statute could qualify as a “misdemeanor crime of domestic violence.” See id., at ___ (slip op., at 5) (If a state crime “sweeps more broadly” than the federally defined one, a conviction for the state offense “cannot count” as a predicate, no matter what mens rea the defendant actually had). So in the 35 jurisdictions like Maine, petitioners’ reading risks allowing domestic abusers of all mental states to evade §922(g)(9)’s firearms ban. In Castleman, we declined to construe §921(a)(33)(A) so as to render §922(g)(9) ineffective in 10 States. See 572 U. S., at ___ (slip op., at 9). All the more so here, where petitioners’ view would jeopardize §922(g)(9)’s force in several times that many. Petitioners respond that we should ignore the assault and battery laws actually on the books when Congress enacted §922(g)(9). In construing the statute, they urge, we should look instead to how the common law defined those crimes in an earlier age. See Brief for Petitioners 13–15. And that approach, petitioners claim, would necessitate reversing their convictions because the common law “required a mens rea greater than recklessness.” Id., at 17. But we see no reason to wind the clock back so far. Once again: Congress passed §922(g)(9) to take guns out of the hands of abusers convicted under the misdemeanor assault laws then in general use in the States. See supra, at 1–2, 8. And by that time, a substantial majority of jurisdictions, following the Model Penal Code’s lead, had abandoned the common law’s approach to mens rea in drafting and interpreting their assault and battery statutes. Indeed, most had gone down that road decades before. That was the backdrop against which Congress was legislating. Nothing suggests that, in enacting §922(g)(9), Congress wished to look beyond that real world to a common-law precursor that had largely expired. To the contrary, such an approach would have undermined Congress’s aim by tying the ban on firearms possession not to the laws under which abusers are prosecuted but instead to a legal anachronism.[5] And anyway, we would not know how to resolve whether recklessness sufficed for a battery conviction at common law. Recklessness was not a word in the common law’s standard lexicon, nor an idea in its conceptual framework; only in the mid- to late-1800’s did courts begin to address reckless behavior in those terms. See Hall, Assault and Battery by the Reckless Motorist, 31 J. Crim. L. & C. 133, 138–139 (1940). The common law traditionally used a variety of overlapping and, frankly, confusing phrases to describe culpable mental states—among them, specific intent, general intent, presumed intent, willfulness, and malice. See, e.g., Morissette v. United States, 342 U. S. 246, 252 (1952) ; Model Penal Code §2.02, Comment 1, at 230. Whether and where conduct that we would today describe as reckless fits into that obscure scheme is anyone’s guess: Neither petitioners’ citations, nor the Government’s competing ones, have succeeded in resolving that counterfactual question. And that indeterminacy confirms our conclusion that Congress had no thought of incorporating the common law’s treatment of mens rea into §921(a)(33)(A). That provision instead corresponds to the ordinary misdemeanor assault and battery laws used to prosecute domestic abuse, regardless of how their mental state requirements might—or, then again, might not—conform to the common law’s.[6] III The federal ban on firearms possession applies to any person with a prior misdemeanor conviction for the “use . . . of physical force” against a domestic relation. §921(a)(33)(A). That language, naturally read, encompasses acts of force undertaken recklessly—i.e., with conscious disregard of a substantial risk of harm. And the state-law backdrop to that provision, which included misdemeanor assault statutes covering reckless conduct in a significant majority of jurisdictions, indicates that Congress meant just what it said. Each petitioner’s possession of a gun, following a conviction under Maine law for abusing a domestic partner, therefore violates §922(g)(9). We accordingly affirm the judgment of the Court of Appeals. It is so ordered.Notes 1 In United States v. Hayes, 555 U. S. 415, 418 (2009) , this Court held that a conviction under a general assault statute like §207 (no less than one under a law targeting only domestic assault) can serve as the predicate offense for a §922(g)(9) prosecution. When that is so, the Government must prove in the later, gun possession case that the perpetrator and the victim of the assault had one of the domestic relationships specified in §921(a)(33)(A). See id., at 426. 2 Compare 778 F. 3d 176 (CA1 2015) (case below) with United States v. Nobriga, 474 F. 3d 561 (CA9 2006) (per curiam) (holding that a conviction for a reckless domestic assault does not trigger §922(g)(9)’s ban). 3 In cases stretching back over a century, this Court has followed suit, although usually discussing the verb form of the word. See, e.g., Bailey v. United States, 516 U. S. 137, 145 (1995) (to use means “ ‘[t]o convert to one’s service,’ ‘to employ,’ [or] ‘to avail oneself of’ ”); Smith v. United States, 508 U. S. 223, 229 (1993) (to use means “ ‘[t]o convert to one’s service’ or ‘to employ’ ”); Astor v. Merritt, 111 U. S. 202, 213 (1884) (to use means “to employ [or] to derive service from”). 4 Like Leocal, our decision today concerning §921(a)(33)(A)’s scope does not resolve whether §16 includes reckless behavior. Courts have sometimes given those two statutory definitions divergent readings in light of differences in their contexts and purposes, and we do not foreclose that possibility with respect to their required mental states. Cf. United States v. Castleman, 572 U. S. ___, ___, n. 4 (2014) (slip op., at 6, n. 4) (interpreting “force” in §921(a)(33)(A) to encompass any offensive touching, while acknowledging that federal appeals courts have usually read the same term in §16 to reach only “violent force”). All we say here is that Leocal’s exclusion of accidental conduct from a definition hinging on the “use” of force is in no way inconsistent with our inclusion of reckless conduct in a similarly worded provision. 5 As petitioners observe, this Court looked to the common law in Castleman to define the term “force” in §921(a)(33)(A). See 572 U. S., at ___–___ (slip op., at 4–5); Brief for Petitioners 13–15. But we did so for reasons not present here. “Force,” we explained, was “a common-law term of art” with an “established common-law meaning.” 572 U. S., at ___ (slip op., at 5) (internal quotation marks omitted). And we thought that Congress meant to adhere to that meaning given its “perfect[ ]” fit with §922(g)(9)’s goal. Ibid. By contrast, neither party pretends that the statutory term “use”—the only one identified as potentially relevant here—has any particular common-law definition. And as explained above, the watershed change in how state legislatures thought of mens rea after the Model Penal Code makes the common law a bad match for the ordinary misdemeanor assault and battery statutes in Congress’s sightline. 6 Petitioners make two last arguments for reading §921(a)(33)(A) their way, but they do not persuade us. First, petitioners contend that we should adopt their construction to avoid creating a question about whether the Second Amendment permits imposing a lifetime firearms ban on a person convicted of a misdemeanor involving reckless conduct. See Brief for Petitioners 32–36. And second, petitioners assert that the rule of lenity requires accepting their view. See id., at 31–32. But neither of those arguments can succeed if the statute is clear. See Pennsylvania Dept. of Corrections v. Yeskey, 524 U. S. 206, 212 (1998) (noting that “the doctrine of constitutional doubt . . . enters in only where a statute is susceptible of two constructions” (internal quotation marks omitted)); Abramski v. United States, 573 U. S. ___, ___, n. 10 (2014) (slip op., at 18, n. 10) (stating that the rule of lenity applies only in cases of genuine ambiguity). And as we have shown, §921(a)(33)(A) plainly encompasses reckless assaults.

578.US.2015_15-6418

Federal law makes the possession of a firearm by a felon a crime punishable by a prison term of up to 10 years, 18 U. S. C. §§922(g), 924(a)(2), but the Armed Career Criminal Act of 1984 increases that sentence to a mandatory 15 years to life if the offender has three or more prior convictions for a “serious drug offense” or a “violent felony,” §924(e)(1). The definition of “violent felony” includes the so-called residual clause, covering any felony that “otherwise involves conduct that presents a serious potential risk of physical injury to another.” §924(e)(2)(B)(ii). In Johnson v. United States, 576 U. S. ___, this Court held that clause unconstitutional under the void-for-vagueness doctrine. Petitioner Welch was sentenced under the Armed Career Criminal Act before Johnson was decided. On direct review, the Eleventh Circuit affirmed his sentence, holding that Welch’s prior Florida conviction for robbery qualified as a “violent felony” under the residual clause. After his conviction became final, Welch sought collateral relief under 28 U. S. C. §2255, which the District Court denied. The Eleventh Circuit then denied Welch a certificate of appealability. Three weeks later, this Court decided Johnson. Welch now seeks the retroactive application of Johnson to his case. Held: Johnson announced a new substantive rule that has retroactive effect in cases on collateral review. Pp. 6–15. (a) An applicant seeking a certificate of appealability in a §2255 proceeding must make “a substantial showing of the denial of a constitutional right.” §2253(c)(2). That standard is met when “reasonable jurists could debate whether . . . the petition should have been resolved in a different manner.” Slack v. McDaniel, 529 U. S. 473 . The question whether Welch met that standard implicates a broader legal issue: whether Johnson is a substantive decision with retroactive effect in cases on collateral review. If so, then on the present record reasonable jurists could at least debate whether Welch should obtain relief in his collateral challenge to his sentence. Pp. 6–7. (b) New constitutional rules of criminal procedure generally do not apply retroactively to cases on collateral review, but new substantive rules do apply retroactively. Teague v. Lane, 489 U. S. 288 ; Schriro v. Summerlin, 542 U. S. 348 . Substantive rules alter “the range of conduct or the class of persons that the law punishes,” id., at 353. Procedural rules, by contrast, “regulate only the manner of determining the defendant’s culpability.” Ibid. Under this framework, Johnson is substantive. Before Johnson, the residual clause could cause an offender to face a prison sentence of at least 15 years instead of at most 10. Since Johnson made the clause invalid, it can no longer mandate or authorize any sentence. By the same logic, Johnson is not procedural, since it had nothing to do with the range of permissible methods a court might use to determine whether a defendant should be sentenced under the Act, see Schriro, supra, at 353. Pp. 7–9. (c) The counterarguments made by Court-appointed amicus are unpersuasive. She contends that Johnson is a procedural decision because the void-for-vagueness doctrine is based on procedural due process. But the Teague framework turns on whether the function of the rule is substantive or procedural, not on the rule’s underlying constitutional source. Amicus’ approach would lead to results that cannot be squared with prior precedent. Precedent also does not support amicus’ claim that a rule must limit Congress’ power to be substantive, see, e.g., Bousley v. United States, 523 U. S. 614 , or her claim that statutory construction cases are an ad hoc exception to that principle and are substantive only because they implement the intent of Congress. The separation-of-powers argument raised by amicus is also misplaced, for regardless of whether a decision involves statutory interpretation or statutory invalidation, a court lacks the power to exact a penalty that has not been authorized by any valid criminal statute. Pp. 10–15. Vacated and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion.

Last Term, this Court decided Johnson v. United States, 576 U. S. ___ (2015). Johnson considered the residual clause of the Armed Career Criminal Act of 1984, 18 U. S. C. §924(e)(2)(B)(ii). The Court held that provision void for vagueness. The present case asks whether Johnson is a substantive decision that is retroactive in cases on collateral review. I Federal law prohibits any felon—meaning a person who has been convicted of a crime punishable by more than a year in prison—from possessing a firearm. 18 U. S. C. §922(g). A person who violates that restriction can be sentenced to prison for up to 10 years. §924(a)(2). For some felons, however, the Armed Career Criminal Act imposes a much more severe penalty. Under the Act, a person who possesses a firearm after three or more convictions for a “serious drug offense” or a “violent felony” is subject to a minimum sentence of 15 years and a maximum sentence of life in prison. §924(e)(1). Because the ordinary maximum sentence for a felon in possession of a firearm is 10 years, while the minimum sentence under the Armed Career Criminal Act is 15 years, a person sentenced under the Act will receive a prison term at least five years longer than the law otherwise would allow. The Act defines “violent felony” as “any crime punishable by imprisonment for a term exceeding one year . . . that— “(i) has as an element the use, attempted use, or threatened use of physical force against the person of another; or “(ii) is burglary, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.” §924(e)(2)(B). Subsection (i) of this definition is known as the elements clause. The end of subsection (ii)—“or otherwise involves conduct that presents a serious potential risk of physical injury to another”—is known as the residual clause. See Johnson, supra, at ___ (slip op., at 2). It is the residual clause that Johnson held to be vague and invalid. The text of the residual clause provides little guidance on how to determine whether a given offense “involves conduct that presents a serious potential risk of physical injury.” This Court sought for a number of years to de-velop the boundaries of the residual clause in a more pre-cise fashion by applying the statute to particular cases. See James v. United States, 550 U. S. 192 (2007) (residual clause covers Florida offense of attempted burglary); Begay v. United States, 553 U. S. 137 (2008) (residual clause does not cover New Mexico offense of driving under the influence of alcohol); Chambers v. United States, 555 U. S. 122 (2009) (residual clause does not cover Illinois offense of failure to report to a penal institution); Sykes v. United States, 564 U. S. 1 (2011) (residual clause covers Indiana offense of vehicular flight from a law-enforcement officer). In Johnson, a majority of this Court concluded that those decisions did not bring sufficient clarity to the scope of the residual clause, noting that the federal courts remained mired in “pervasive disagreement” over how the clause should be interpreted. Johnson, 576 U. S., at ___ (slip op., at 9). The Johnson Court held the residual clause unconstitutional under the void-for-vagueness doctrine, a doctrine that is mandated by the Due Process Clauses of the Fifth Amendment (with respect to the Federal Government) and the Fourteenth Amendment (with respect to the States). The void-for-vagueness doctrine prohibits the government from imposing sanctions “under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement.” Id., at ___ (slip op., at 3). Johnson determined that the residual clause could not be reconciled with that prohibition. The vagueness of the residual clause rests in large part on its operation under the categorical approach. The categorical approach is the framework the Court has applied in deciding whether an offense qualifies as a violent felony under the Armed Career Criminal Act. See id., at ___ (slip op., at 4). Under the categorical approach, “a court assesses whether a crime qualifies as a violent fel-ony ‘in terms of how the law defines the offense and not in terms of how an individual offender might have committed it on a particular occasion.’ ” Ibid. (quoting Begay, supra, at 141). For purposes of the residual clause, then, courts were to determine whether a crime involved a “serious potential risk of physical injury” by considering not the defendant’s actual conduct but an “idealized ordinary case of the crime.” 576 U. S., at ___ (slip op., at 12). The Court’s analysis in Johnson thus cast no doubt on the many laws that “require gauging the riskiness of conduct in which an individual defendant engages on a particular occasion.” Ibid. The residual clause failed not because it adopted a “serious potential risk” standard but because applying that standard under the categorical approach required courts to assess the hypothetical risk posed by an abstract generic version of the offense. In the Johnson Court’s view, the “indeterminacy of the wide-ranging inquiry” made the residual clause more unpredictable and arbitrary in its application than the Consti-tution allows. Id., at ___ (slip op., at 5). “Invoking so shapeless a provision to condemn someone to prison for 15 years to life,” the Court held, “does not comport with the Constitution’s guarantee of due process.” Id., at ___ (slip op., at 10). II Petitioner Gregory Welch is one of the many offenders sentenced under the Armed Career Criminal Act before Johnson was decided. Welch pleaded guilty in 2010 to one count of being a felon in possession of a firearm. The Probation Office prepared a presentence report finding that Welch had three prior violent felony convictions, including a Florida conviction for a February 1996 “strong-arm robbery.” The relevant Florida statute prohibits taking property from the person or custody of another with “the use of force, violence, assault, or putting in fear.” Fla. Stat. §812.13(1) (1994). The charging document from the 1996 Florida case tracked that statutory language. App. 187a. The 2010 federal presentence report provides more detail. It states that, according to the robbery victim, Welch punched the victim in the mouth and grabbed a gold bracelet from his wrist while another attacker grabbed a gold chain from his neck. Welch objected to the presentence report, arguing (as relevant here) that this conviction was not a violent felony conviction under the Armed Career Criminal Act. The District Court overruled the objection. It concluded that the Florida offense of strong-arm robbery qualified as a violent felony both under the elements clause, 18 U. S. C. §924(e)(2)(B)(i), and the residual clause, §924(e)(2)(B)(ii). The District Court proceeded to sentence Welch to the Act’s mandatory minimum sentence of 15 years in prison. The Court of Appeals for the Eleventh Circuit affirmed. That court did not decide whether the conviction at issue could qualify as a violent felony under the elements clause. Instead, it held only that the conviction qualified under the residual clause. This Court denied certiorari, see Welch v. United States, 568 U. S. ___ (2013), and Welch’s conviction became final. In December 2013, Welch appeared pro se before the District Court and filed a collateral challenge to his conviction and sentence through a motion under 28 U. S. C. §2255. He argued, among other points, that his strong-arm robbery conviction itself was “vague” and that his counsel was ineffective for allowing him to be sentenced as an armed career criminal. The District Court denied the motion and denied a certificate of appealability. Still proceeding pro se, Welch applied to the Court of Appeals for a certificate of appealability. His application noted that Johnson was pending before this Court. Welch argued, in part, that his “armed career offender status is unconstitutional and violate[s] [his] Fifth Amendment right to notice of the state priors.” App. 20a. Two months later, Welch filed a motion asking the Court of Appeals to hold his case in abeyance until Johnson could be decided, “based on the fact he was sentenced under the [residual clause].” App. 15a. In June 2015, the Court of Appeals entered a brief single-judge order denying the motion for a certificate of appealability. Less than three weeks later, this Court issued its decision in Johnson holding, as already noted, that the residual clause is void for vagueness. Welch filed a motion asking the Court of Appeals for additional time to seek reconsideration of its decision in light of Johnson, but the court returned that motion unfiled because Welch’s time to seek reconsideration already had expired. Welch then filed a pro se petition for certiorari. His petition presented two questions: whether the District Court erred in denying his §2255 motion because his Florida robbery conviction does not qualify as a violent felony conviction under the Armed Career Criminal Act; and whether Johnson announced a substantive rule that has retroactive effect in cases on collateral review. Pet. for Cert. i. This Court granted the petition. 577 U. S. ___ (2016). Because the United States, as respondent, agrees with Welch that Johnson is retroactive, the Court appointed Helgi C. Walker as amicus curiae in support of the judgment of the Court of Appeals. She has ably discharged her responsibilities. III A This case comes to the Court in a somewhat unusual procedural posture. Under the Antiterrorism and Effective Death Penalty Act of 1996, there can be no appeal from a final order in a §2255 proceeding unless a circuit justice or judge issues a certificate of appealability. 28 U. S. C. §2253(c)(1). A certificate of appealability may issue “only if the applicant has made a substantial showing of the denial of a constitutional right.” §2253(c)(2). That standard is met when “reasonable jurists could debate whether (or, for that matter, agree that) the petition should have been resolved in a different manner.” Slack v. McDaniel, 529 U. S. 473, 484 (2000) . Obtaining a certificate of appealability “does not require a showing that the appeal will succeed,” and “a court of appeals should not decline the application . . . merely because it believes the applicant will not demonstrate an entitlement to relief.” Miller-El v. Cockrell, 537 U. S. 322, 337 (2003) . The decision under review here is the single-judge order in which the Court of Appeals denied Welch a certificate of appealability. Under the standard described above, that order determined not only that Welch had failed to show any entitlement to relief but also that reasonable jurists would consider that conclusion to be beyond all debate. See Slack, supra, at 484. The narrow question here is whether the Court of Appeals erred in making that determination. That narrow question, however, implicates a broader legal issue: whether Johnson is a substantive decision with retroactive effect in cases (like Welch’s) on collateral review. If so, then on the present record reasonable jurists could at least debate whether Welch should obtain relief in his collateral challenge to his sentence. On these premises, the Court now proceeds to decide whether Johnson is retroactive. B The normal framework for determining whether a new rule applies to cases on collateral review stems from the plurality opinion in Teague v. Lane, 489 U. S. 288 (1989) . That opinion in turn drew on the approach outlined by the second Justice Harlan in his separate opinions in Mackey v. United States, 401 U. S. 667 (1971) , and Desist v. United States, 394 U. S. 244 (1969) . The parties here assume that the Teague framework applies in a federal collateral challenge to a federal conviction as it does in a federal collateral challenge to a state conviction, and we proceed on that assumption. See Chaidez v. United States, 568 U. S. ___, ___, n. 16 (2013); Danforth v. Minnesota, 552 U. S. 264 , n. 4 (2008). Under Teague, as a general matter, “new constitutional rules of criminal procedure will not be applicable to those cases which have become final before the new rules are announced.” 489 U. S., at 310. Teague and its progeny recognize two categories of decisions that fall outside this general bar on retroactivity for procedural rules. First, “[n]ew substantive rules generally apply retroactively.” Schriro v. Summerlin, 542 U. S. 348, 351 (2004) ; see Montgomery v. Louisiana, 577 U. S. ___, ___ (2016) (slip op., at 6); Teague, supra, at 307, 311. Second, new “ ‘watershed rules of criminal procedure,’ ” which are proce-dural rules “implicating the fundamental fairness and accu-racy of the criminal proceeding,” will also have retroactive effect. Saffle v. Parks, 494 U. S. 484, 495 (1990) ; see Teague, supra, at 311–313. It is undisputed that Johnson announced a new rule. See Teague, supra, at 301 (“[A] case announces a new rule if the result was not dictated by precedent existing at the time the defendant’s conviction became final”). The question here is whether that new rule falls within one of the two categories that have retroactive effect under Teague. The parties agree that Johnson does not fall into the limited second category for watershed procedural rules. Welch and the United States contend instead that Johnson falls into the first category because it announced a substantive rule. “A rule is substantive rather than procedural if it alters the range of conduct or the class of persons that the law punishes.” Schriro, 542 U. S., at 353. “This includes decisions that narrow the scope of a criminal statute by interpreting its terms, as well as constitutional determinations that place particular conduct or persons covered by the statute beyond the State’s power to punish.” Id., at 351–352 (citation omitted); see Montgomery, supra, at ___ (slip op., at 6). Procedural rules, by contrast, “regulate only the manner of determining the defendant’s culpability.” Schriro, 542 U. S., at 353. Such rules alter “the range of permissible methods for determining whether a defendant’s conduct is punishable.” Ibid. “They do not produce a class of persons convicted of conduct the law does not make criminal, but merely raise the possibility that someone convicted with use of the invalidated procedure might have been acquitted otherwise.” Id., at 352. Under this framework, the rule announced in Johnson is substantive. By striking down the residual clause as void for vagueness, Johnson changed the substantive reach of the Armed Career Criminal Act, altering “the range of conduct or the class of persons that the [Act] punishes.” Schriro, supra, at 353. Before Johnson, the Act applied to any person who possessed a firearm after three violent felony convictions, even if one or more of those convictions fell under only the residual clause. An offender in that situation faced 15 years to life in prison. After Johnson, the same person engaging in the same conduct is no longer subject to the Act and faces at most 10 years in prison. The residual clause is invalid under Johnson, so it can no longer mandate or authorize any sentence. Johnson establishes, in other words, that “even the use of impeccable factfinding procedures could not legitimate” a sentence based on that clause. United States v. United States Coin & Currency, 401 U. S. 715, 724 (1971) . It follows that Johnson is a substantive decision. By the same logic, Johnson is not a procedural decision. Johnson had nothing to do with the range of permissible methods a court might use to determine whether a defendant should be sentenced under the Armed Career Criminal Act. See Schriro, 542 U. S., at 353. It did not, for example, “allocate decisionmaking authority” between judge and jury, ibid., or regulate the evidence that the court could consider in making its decision, see Whorton v. Bockting, 549 U. S. 406 –414, 417 (2007); Mackey, supra, at 700–701 (opinion of Harlan, J.). Unlike those cases, Johnson affected the reach of the underlying statute rather than the judicial procedures by which the statute is applied. Johnson is thus a substantive decision and so has retroactive effect under Teague in cases on collateral review. C Amicus urges the Court to adopt a different understanding of the Teague framework. She contends courts should apply that framework by asking whether the constitu-tional right underlying the new rule is substantive or procedural. Under that approach, amicus concludes that Johnson is a procedural decision because the void-for-vagueness doctrine that Johnson applied is based, she asserts, on procedural due process. Neither Teague nor its progeny support that approach. As described above, this Court has determined whether a new rule is substantive or procedural by considering the function of the rule, not its underlying constitutional source. See, e.g., Schriro, supra, at 351–353. That is for good reason. The Teague framework creates a balance between, first, the need for finality in criminal cases, and second, the countervailing imperative to ensure that criminal punishment is imposed only when authorized by law. That balance turns on the function of the rule at issue, not the constitutional guarantee from which the rule derives. If a new rule regulates only the procedures for determining culpability, the Teague balance generally tips in favor of finality. The chance of a more accurate outcome under the new procedure normally does not justify the cost of vacating a conviction whose only flaw is that its procedures “conformed to then-existing constitutional standards.” Teague, supra, at 310. On the other hand, if a new rule changes the scope of the underlying criminal proscription, the balance is different. A change of that character will “necessarily carry a significant risk that a defendant stands convicted of ‘an act that the law does not make criminal.’ ” Bousley v. United States, 523 U. S. 614, 620 (1998) (quoting Davis v. United States, 417 U. S. 333, 346 (1974) ). By extension, where the conviction or sentence in fact is not authorized by substantive law, then finality interests are at their weakest. As Justice Harlan wrote, “[t]here is little societal interest in permitting the criminal process to rest at a point where it ought properly never to repose.” Mackey, 401 U. S., at 693 (opinion of Harlan, J.). The Teague balance thus does not depend on whether the underlying constitutional guarantee is characterized as procedural or substantive. It depends instead on whether the new rule itself has a procedural function or a substantive function—that is, whether it alters only the procedures used to obtain the conviction, or alters instead the range of conduct or class of persons that the law punishes. See Schriro, supra, at 353; Montgomery, 577 U. S., at ___ (slip op., at 14). The emphasis by amicus on the constitutional guarantee behind the new rule, then, would untether the Teague framework from its basic purpose. The approach amicus suggests also would lead to results that cannot be squared with prior precedent. Decisions from this Court show that a rule that is procedural for Teague purposes still can be grounded in a substantive constitutional guarantee. For instance, the Court has adopted certain rules that regulate capital sentencing procedures in order to enforce the substantive guarantees of the Eighth Amendment. The consistent position has been that those rules are procedural, even though their ultimate source is substantive. See, e.g., Beard v. Banks, 542 U. S. 406 –417 (2004); Sawyer v. Smith, 497 U. S. 227 –242 (1990). From the converse perspective, there also can be substantive rules based on constitutional protections that, on the theory amicus advances, likely would be described as procedural. For instance, a decision that invalidates as void for vagueness a statute prohibiting “conduct annoying to persons passing by,” cf. Coates v. Cincinnati, 402 U. S. 611, 612, 614 (1971) , would doubtless alter the range of conduct that the law prohibits. That would make it a substantive decision under our precedent, see Schriro, 542 U. S., at 353, even if the reasons for holding that statute invalid could be characterized as procedural. Amicus next relies on language from this Court’s cases describing substantive decisions as those that “place particular conduct or persons . . . beyond the State’s power to punish,” id., at 352, or that “prohibi[t] a certain category of punishment for a class of defendants because of their status or offense,” Saffle, 494 U. S., at 494 (internal quotation marks omitted). Cases such as these, in which the Constitution deprives the Government of the power to impose the challenged punishment, “represen[t] the clearest instance” of substantive rules for which retroactive application is appropriate. Mackey, supra, at 693 (opinion of Harlan, J.). Drawing on those decisions, amicus argues that Johnson is not substantive because it does not limit Congress’ power: Congress is free to enact a new version of the residual clause that imposes the same punishment on the same persons for the same conduct, provided the new statute is precise enough to satisfy due process. Although this Court has put great emphasis on substantive decisions that place certain conduct, classes of persons, or punishments beyond the legislative power of Congress, the Court has also recognized that some substantive decisions do not impose such restrictions. The clearest example comes from Bousley, supra. In Bousley, the Court was asked to determine what retroactive effect should be given to its decision in Bailey v. United States, 516 U. S. 137 (1995) . Bailey considered the “use” prong of 18 U. S. C. §924(c)(1), which imposes increased penalties on the use of a firearm in relation to certain crimes. The Court held as a matter of statutory interpretation that the “use” prong punishes only “active employment of the firearm” and not mere possession. 516 U. S., at 144. The Court in Bousley had no difficulty concluding that Bailey was substantive, as it was a decision “holding that a substantive federal criminal statute does not reach certain conduct.” Bousley, supra, at 620; see Schriro, supra, at 354 (“A decision that modifies the elements of an offense is normally substantive rather than procedural”). The Court reached that conclusion even though Congress could (and later did) reverse Bailey by amending the statute to cover possession as well as use. See United States v. O’Brien, 560 U. S. 218 –233 (2010) (discussing statutory amendment known as the “Bailey fix”). Bousley thus contradicts the contention that the Teague inquiry turns only on whether the decision at issue holds that Congress lacks some substantive power. Amicus recognizes that Bousley does not fit the theory that, in her view, should control this case. She instead proposes an ad hoc exception, contending that Bousley “recognized a separate subcategory of substantive rules” for decisions that interpret statutes (but not those, like Johnson, that invalidate statutes). Brief for Court-Appointed Amicus Curiae in Support of Judgment Below 40. For support, amicus looks to the separation-of-powers doctrine. Her argument is that statutory construction cases are substantive because they define what Congress always intended the law to mean—unlike Johnson, which struck down the residual clause regardless of Congress’ intent. That argument is not persuasive. Neither Bousley nor any other case from this Court treats statutory interpretation cases as a special class of decisions that are substantive because they implement the intent of Congress. Instead, decisions that interpret a statute are substantive if and when they meet the normal criteria for a substantive rule: when they “alte[r] the range of conduct or the class of persons that the law punishes.” Schriro, supra, at 353. The separation-of-powers argument that amicus raises is also misplaced. Bousley noted that the separation of powers prohibits a court from imposing criminal punishment beyond what Congress meant to enact. 523 U. S., at 620–621 (“[I]t is only Congress, and not the courts, which can make conduct criminal”). But a court likewise is prohibited from imposing criminal punishment beyond what Congress in fact has enacted by a valid law. In either case a court lacks the power to exact a penalty that has not been authorized by any valid criminal statute. Treating decisions as substantive if they involve statutory interpretation, but not if they involve statutory invalidation, would produce unusual outcomes. “It has long been our practice . . . before striking a federal statute as impermissibly vague, to consider whether the prescription is amenable to a limiting construction.” Skilling v. United States, 561 U. S. 358 –406 (2010). Amicus acknowledges that a decision that saves a vague statute by adopting a limiting construction is substantive, so anyone who falls outside the limiting construction can use that decision to seek relief on collateral review. But amicus also contends that, if a court takes the further step of striking down the whole statute as vague, that decision is procedural, so no one can use it to seek relief on collateral review. That arbitrary distinction has no place in the Teague framework. It should be noted, of course, that not every decision striking down a statute is ipso facto a substantive decision. A decision that strikes down a procedural statute—for example, a statute regulating the types of evidence that can be presented at trial—would itself be a proce-dural decision. It would affect only the “manner of determining the defendant’s culpability,” not the conduct or persons to be punished. Schriro, 542 U. S., at 353 (emphasis deleted). A decision of this kind would have no retroactive effect under Teague unless it could be considered a “watershed” procedural rule. See Teague, 489 U. S., at 311–313. Johnson, however, struck down part of a criminal statute that regulates conduct and prescribes punishment. It thereby altered “the range of conduct or the class of persons that the law punishes.” Schriro, supra, at 353. It follows that Johnson announced a substantive rule that has retroactive effect in cases on collateral review. * * * It may well be that the Court of Appeals on remand will determine on other grounds that the District Court was correct to deny Welch’s motion to amend his sentence. For instance, the parties continue to dispute whether Welch’s strong-arm robbery conviction qualifies as a violent felony under the elements clause of the Act, which would make Welch eligible for a 15-year sentence regardless of Johnson. On the present record, however, and in light of today’s holding that Johnson is retroactive in cases on collateral review, reasonable jurists at least could debate whether Welch is entitled to relief. For these reasons, the judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

579.US.2015_15-274

A “State has a legitimate interest in seeing to it that abortion . . . is performed under circumstances that insure maximum safety for the patient.” Roe v. Wade, 410 U. S. 113 . But “a statute which, while furthering [a] valid state interest, has the effect of placing a substantial obstacle in the path of a woman’s choice cannot be considered a permissible means of serving its legitimate ends,” Planned Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833 (plurality opinion), and “[u]nnecessary health regulations that have the purpose or effect of presenting a substantial obstacle to a woman seeking an abortion impose an undue burden on the right,” id., at 878. In 2013, the Texas Legislature enacted House Bill 2 (H. B. 2), which contains the two provisions challenged here. The “admitting-privileges requirement” provides that a “physician performing or inducing an abortion . . . must, on the date [of service], have active admitting privileges at a hospital . . . located not further than 30 miles from the” abortion facility. The “surgical-center requirement” requires an “abortion facility” to meet the “minimum standards . . . for ambulatory surgical centers” under Texas law. Before the law took effect, a group of Texas abortion providers filed the Abbott case, in which they lost a facial challenge to the constitutionality of the admitting-privileges provision. After the law went into effect, petitioners, another group of abortion providers (including some Abbott plaintiffs), filed this suit, claiming that both the admitting-privileges and the surgical-center provisions violated the Fourteenth Amendment, as interpreted in Casey. They sought injunctions preventing enforcement of the admitting-privileges provision as applied to physicians at one abortion facility in McAllen and one in El Paso and prohibiting enforcement of the surgical-center provision throughout Texas. Based on the parties’ stipulations, expert depositions, and expert and other trial testimony, the District Court made extensive findings, including, but not limited to: as the admitting-privileges requirement began to be enforced, the number of facilities providing abortions dropped in half, from about 40 to about 20; this decrease in geographical distribution means that the number of women of reproductive age living more than 50 miles from a clinic has doubled, the number living more than 100 miles away has increased by 150%, the number living more than 150 miles away by more than 350%, and the number living more than 200 miles away by about 2,800%; the number of facilities would drop to seven or eight if the surgical-center provision took effect, and those remaining facilities would see a significant increase in patient traffic; facilities would remain only in five metropolitan areas; before H. B. 2’s passage, abortion was an extremely safe procedure with very low rates of complications and virtually no deaths; it was also safer than many more common procedures not subject to the same level of regulation; and the cost of compliance with the surgical-center requirement would most likely exceed $1.5 million to $3 million per clinic. The court enjoined enforcement of the provisions, holding that the surgical-center requirement imposed an undue burden on the right of women in Texas to seek previability abortions; that, together with that requirement, the admitting-privileges requirement imposed an undue burden in the Rio Grande Valley, El Paso, and West Texas; and that the provisions together created an “impermissible obstacle as applied to all women seeking a previability abortion.” The Fifth Circuit reversed in significant part. It concluded that res judicata barred the District Court from holding the admitting-privileges requirement unconstitutional statewide and that res judicata also barred the challenge to the surgical-center provision. Reasoning that a law is “constitutional if (1) it does not have the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus and (2) it is reasonably related to . . . a legitimate state interest,” the court found that both requirements were rationally related to a compelling state interest in protecting women’s health. Held: 1. Petitioners’ constitutional claims are not barred by res judicata. Pp. 10–18. (a) Res judicata neither bars petitioners’ challenges to the admitting-privileges requirement nor prevents the Court from awarding facial relief. The fact that several petitioners had previously brought the unsuccessful facial challenge in Abbott does not mean that claim preclusion, the relevant aspect of res judicata, applies. Claim preclusion prohibits “successive litigation of the very same claim,” New Hampshire v. Maine, 532 U. S. 742 , but petitioners’ as-applied postenforcement challenge and the Abbott plaintiffs’ facial preenforcement challenge do not present the same claim. Changed circumstances showing that a constitutional harm is concrete may give rise to a new claim. Abbott rested upon facts and evidence presented before enforcement of the admitting-privileges requirement began, when it was unclear how clinics would be affected. This case rests upon later, concrete factual developments that occurred once enforcement started and a significant number of clinics closed. Res judicata also does not preclude facial relief here. In addition to requesting as-applied relief, petitioners asked for other appropriate relief, and their evidence and arguments convinced the District Court of the provision’s unconstitutionality across the board. Federal Rule of Civil Procedure 54(c) provides that a “final judgment should grant the relief to which each party is entitled, even if the party has not demanded that relief in its pleadings,” and this Court has held that if the arguments and evidence show that a statutory provision is unconstitutional on its face, an injunction prohibiting its enforcement is “proper,” Citizens United v. Federal Election Comm’n, 558 U. S. 310 . Pp. 10–15. (b) Claim preclusion also does not bar petitioners’ challenge to the surgical-center requirement. In concluding that petitioners should have raised this claim in Abbott, the Fifth Circuit did not take account of the fact that the surgical-center provision and the admitting-privileges provision are separate provisions with two different and independent regulatory requirements. Challenges to distinct regulatory requirements are ordinarily treated as distinct claims. Moreover, the surgical-center provision’s implementing regulations had not even been promulgated at the time Abbott was filed, and the relevant factual circumstances changed between the two suits. Pp. 16–18. 2. Both the admitting-privileges and the surgical-center requirements place a substantial obstacle in the path of women seeking a previability abortion, constitute an undue burden on abortion access, and thus violate the Constitution. Pp. 19–39. (a) The Fifth Circuit’s standard of review may be read to imply that a district court should not consider the existence or nonexistence of medical benefits when deciding the undue burden question, but Casey requires courts to consider the burdens a law imposes on abortion access together with the benefits those laws confer, see 505 U. S., at 887–898. The Fifth Circuit’s test also mistakenly equates the judicial review applicable to the regulation of a constitutionally protected personal liberty with the less strict review applicable to, e.g., economic legislation. And the court’s requirement that legislatures resolve questions of medical uncertainty is inconsistent with this Court’s case law, which has placed considerable weight upon evidence and argument presented in judicial proceedings when determining the constitutionality of laws regulating abortion procedures. See id., at 888–894. Explicit legislative findings must be considered, but there were no such findings in H. B. 2. The District Court applied the correct legal standard here, considering the evidence in the record—including expert evidence—and then weighing the asserted benefits against the burdens. Pp. 19–21. (b) The record contains adequate legal and factual support for the District Court’s conclusion that the admitting-privileges requirement imposes an “undue burden” on a woman’s right to choose. The requirement’s purpose is to help ensure that women have easy access to a hospital should complications arise during an abortion procedure, but the District Court, relying on evidence showing extremely low rates of serious complications before H. B. 2’s passage, found no significant health-related problem for the new law to cure. The State’s record evidence, in contrast, does not show how the new law advanced the State’s legitimate interest in protecting women’s health when compared to the prior law, which required providers to have a “working arrangement” with doctors who had admitting privileges. At the same time, the record evidence indicates that the requirement places a “substantial obstacle” in a woman’s path to abortion. The dramatic drop in the number of clinics means fewer doctors, longer waiting times, and increased crowding. It also means a significant increase in the distance women of reproductive age live from an abortion clinic. Increased driving distances do not always constitute an “undue burden,” but they are an additional burden, which, when taken together with others caused by the closings, and when viewed in light of the virtual absence of any health benefit, help support the District Court’s “undue burden” conclusion. Pp. 21–28. (c) The surgical-center requirement also provides few, if any, health benefits for women, poses a substantial obstacle to women seeking abortions, and constitutes an “undue burden” on their constitutional right to do so. Before this requirement was enacted, Texas law required abortion facilities to meet a host of health and safety requirements that were policed by inspections and enforced through administrative, civil, and criminal penalties. Record evidence shows that the new provision imposes a number of additional requirements that are generally unnecessary in the abortion clinic context; that it provides no benefit when complications arise in the context of a medical abortion, which would generally occur after a patient has left the facility; that abortions taking place in abortion facilities are safer than common procedures that occur in outside clinics not subject to Texas’ surgical-center requirements; and that Texas has waived no part of the requirement for any abortion clinics as it has done for nearly two-thirds of other covered facilities. This evidence, along with the absence of any contrary evidence, supports the District Court’s conclusions, including its ultimate legal conclusion that requirement is not necessary. At the same time, the record provides adequate evidentiary support for the District Court’s conclusion that the requirement places a substantial obstacle in the path of women seeking an abortion. The court found that it “strained credulity” to think that the seven or eight abortion facilities would be able to meet the demand. The Fifth Circuit discounted expert witness Dr. Grossman’s testimony that the surgical-center requirement would cause the number of abortions performed by each remaining clinic to increase by a factor of about 5. But an expert may testify in the “form of an opinion” as long as that opinion rests upon “sufficient facts or data” and “reliable principles and methods.” Fed. Rule Evid. 702. Here, Dr. Grossman’s opinion rested upon his participation, together with other university researchers, in research tracking the number of facilities providing abortion services, using information from, among other things, the state health services department and other public sources. The District Court acted within its legal authority in finding his testimony admissible. Common sense also suggests that a physical facility that satisfies a certain physical demand will generally be unable to meet five times that demand without expanding physically or otherwise incurring significant costs. And Texas presented no evidence at trial suggesting that expansion was possible. Finally, the District Court’s finding that a currently licensed abortion facility would have to incur considerable costs to meet the surgical-center requirements supports the conclusion that more surgical centers will not soon fill the gap left by closed facilities. Pp. 28–36. (d) Texas’ three additional arguments are unpersuasive. Pp. 36–39. 790 F. 3d 563 and 598, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Ginsburg, J., filed a concurring opinion. Thomas, J., filed a dissenting opinion. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas, J., joined.

In Planned Parenthood of Southeastern Pa. v. Casey, 505 U. S. 833, 878 (1992) , a plurality of the Court con-cluded that there “exists” an “undue burden” on a woman’s right to decide to have an abortion, and consequently a provision of law is constitutionally invalid, if the “purpose or effect” of the provision “is to place a substantial obstacle in the path of a woman seeking an abortion before the fetus attains viability.” (Emphasis added.) The plurality added that “[u]nnecessary health regulations that have the purpose or effect of presenting a substantial obstacle to a woman seeking an abortion impose an undue burden on the right.” Ibid. We must here decide whether two provisions of Texas’ House Bill 2 violate the Federal Constitution as inter-preted in Casey. The first provision, which we shall call the “admitting-privileges requirement,” says that “[a] physician performing or inducing an abortion . . . must, on the date the abortion is performed or induced, have active admitting privileges at a hospital that . . . is located not further than 30 miles from the location at which the abortion is performed or induced.” Tex. Health & Safety Code Ann. §171.0031(a) (West Cum. Supp. 2015). This provision amended Texas law that had previously required an abortion facility to maintain a written protocol “for managing medical emergencies and the transfer of patients requiring further emergency care to a hospital.” 38 Tex. Reg. 6546 (2013). The second provision, which we shall call the “surgical-center requirement,” says that “the minimum standards for an abortion facility must be equivalent to the minimum standards adopted under [the Texas Health and Safety Code section] for ambulatory surgical centers.” Tex. Health & Safety Code Ann. §245.010(a). We conclude that neither of these provisions offers medical benefits sufficient to justify the burdens upon access that each imposes. Each places a substantial obstacle in the path of women seeking a previability abortion, each constitutes an undue burden on abortion access, Casey, supra, at 878 (plurality opinion), and each violates the Federal Constitution. Amdt. 14, §1. I A In July 2013, the Texas Legislature enacted House Bill 2 (H. B. 2 or Act). In September (before the new law took effect), a group of Texas abortion providers filed an action in Federal District Court seeking facial invalidation ofthe law’s admitting-privileges provision. In late October, the District Court granted the injunction. Planned Parenthood of Greater Tex. Surgical Health Servs. v. Abbott, 951 F. Supp. 2d 891, 901 (WD Tex. 2013). But three days later, the Fifth Circuit vacated the injunction,thereby permitting the provision to take effect. Planned Parenthood of Greater Tex. Surgical Health Servs. v. Abbott, 734 F. 3d 406, 419 (2013). The Fifth Circuit subsequently upheld the provision, and set forth its reasons in an opinion released late the following March. In that opinion, the Fifth Circuit pointed to evidence introduced in the District Court the previous October. It noted that Texas had offered evidence designed to show that the admitting-privileges requirement “will reduce the delay in treatment and decrease health risk for abortion patients with critical complications,” and that it would “ ‘screen out’ untrained or incompetent abortion providers.” Planned Parenthood of Greater Tex. Surgical Health Servs. v. Abbott, 748 F. 3d 583, 592 (2014) (Abbott). The opinion also explained that the plaintiffs had not provided sufficient evidence “that abortion practitioners will likely be unable to comply with the privileges requirement.” Id., at 598. The court said that all “of the major Texas cities, including Austin, Corpus Christi, Dallas, El Paso, Houston, and San Antonio,” would “continue to have multiple clinics where many physicians will have or obtain hospital admitting privileges.” Ibid. The Abbott plaintiffs did not file a petition for certiorari in this Court. B On April 6, one week after the Fifth Circuit’s decision, petitioners, a group of abortion providers (many of whom were plaintiffs in the previous lawsuit), filed the present lawsuit in Federal District Court. They sought an injunction preventing enforcement of the admitting-privileges provision as applied to physicians at two abortion facilities, one operated by Whole Woman’s Health in McAllen and the other operated by Nova Health Systems in El Paso. They also sought an injunction prohibiting enforcement of the surgical-center provision anywhere in Texas. They claimed that the admitting-privileges provision and the surgical-center provision violated the Constitution’s Fourteenth Amendment, as interpreted in Casey. The District Court subsequently received stipulations from the parties and depositions from the parties’ experts. The court conducted a 4-day bench trial. It heard, among other testimony, the opinions from expert witnesses for both sides. On the basis of the stipulations, deposi-tions, and testimony, that court reached the following conclusions: 1. Of Texas’ population of more than 25 million people, “approximately 5.4 million” are “women” of “reproductive age,” living within a geographical area of “nearly 280,000 square miles.” Whole Woman’s Health v. Lakey, 46 F. Supp. 3d 673, 681 (2014); see App. 244. 2. “In recent years, the number of abortions reported in Texas has stayed fairly consistent at approximately 15–16% of the reported pregnancy rate, for a total number of approximately 60,000–72,000 legal abortions performed annually.” 46 F. Supp. 3d, at 681; see App. 238. 3. Prior to the enactment of H. B. 2, there were more than 40 licensed abortion facilities in Texas, which “number dropped by almost half leading up to and in thewake of enforcement of the admitting-privileges requirement that went into effect in late-October 2013.” 46 F. Supp. 3d, at 681; App. 228–231. 4. If the surgical-center provision were allowed to take effect, the number of abortion facilities, after September 1, 2014, would be reduced further, so that “only seven fa-cilities and a potential eighth will exist in Texas.” 46 F. Supp. 3d, at 680; App. 182–183. 5. Abortion facilities “will remain only in Houston, Austin, San Antonio, and the Dallas/Fort Worth metropolitan region.” 46 F. Supp. 3d, at 681; App. 229–230. These include “one facility in Austin, two in Dallas, one in Fort Worth, two in Houston, and either one or two in San Antonio.” 46 F. Supp. 3d, at 680; App. 229–230. 6. “Based on historical data pertaining to Texas’s average number of abortions, and assuming perfectly equal distribution among the remaining seven or eight providers, this would result in each facility serving between 7,500 and 10,000 patients per year. Accounting for the seasonal variations in pregnancy rates and a slightly unequal distribution of patients at each clinic, it is foreseeable that over 1,200 women per month could be vying for counseling, appointments, and follow-up visits at some of these facilities.” 46 F. Supp. 3d, at 682; cf. App. 238. 7. The suggestion “that these seven or eight providers could meet the demand of the entire state stretches credulity.” 46 F. Supp. 3d, at 682; see App. 238. 8. “Between November 1, 2012 and May 1, 2014,” that is, before and after enforcement of the admitting-privileges requirement, “the decrease in geographical distribution of abortion facilities” has meant that the number of women of reproductive age living more than 50 miles from a clinic has doubled (from 800,000 to over 1.6 million); those living more than 100 miles has increased by 150% (from 400,000 to 1 million); those living more than 150 miles has increased by more than 350% (from 86,000 to 400,000); and those living more than 200 miles has increased by about 2,800% (from 10,000 to 290,000). After September 2014, should the surgical-center requirement go into effect, the number of women of reproductive age living significant distances from an abortion provider will increase as follows: 2 million women of reproductive age will live more than 50 miles from an abortion provider; 1.3 million will live more than 100 miles from an abortion provider; 900,000 will live more than 150 miles from an abortion provider; and 750,000 more than 200 miles from an abortion provider. 46 F. Supp. 3d, at 681–682; App. 238–242. 9. The “two requirements erect a particularly high barrier for poor, rural, or disadvantaged women.” 46 F. Supp. 3d, at 683; cf. App. 363–370. 10. “The great weight of evidence demonstrates that, before the act’s passage, abortion in Texas was extremely safe with particularly low rates of serious complications and virtually no deaths occurring on account of the procedure.” 46 F. Supp. 3d, at 684; see, e.g., App. 257–259, 538; see also id., at 200–202, 253–257. 11. “Abortion, as regulated by the State before the enactment of House Bill 2, has been shown to be much safer, in terms of minor and serious complications, than many common medical procedures not subject to such intense regulation and scrutiny.” 46 F. Supp. 3d, at 684; see, e.g., App. 223–224 (describing risks in colonoscopies), 254 (discussing risks in vasectomy and endometrial biopsy, among others), 275–277 (discussing complication rate in plastic surgery). 12. “Additionally, risks are not appreciably lowered for patients who undergo abortions at ambulatory surgical centers as compared to nonsurgical-center facilities.” 46 F. Supp. 3d, at 684; App. 202–206, 257–259. 13. “[W]omen will not obtain better care or experience more frequent positive outcomes at an ambulatory surgical center as compared to a previously licensed facility.” 46 F. Supp. 3d, at 684; App. 202–206. 14. “[T]here are 433 licensed ambulatory surgical centers in Texas,” of which “336 . . . are apparently either ‘grandfathered’ or enjo[y] the benefit of a waiver of some or all” of the surgical-center “requirements.” 46 F. Supp. 3d, at 680–681; App. 184. 15. The “cost of coming into compliance” with thesurgical-center requirement “for existing clinics is significant,” “undisputedly approach[ing] 1 million dollars,” and “most likely exceed[ing] 1.5 million dollars,” with “[s]ome . . . clinics” unable to “comply due to physical size limitations of their sites.” 46 F. Supp. 3d, at 682. The “cost of acquiring land and constructing a new compliant clinic will likely exceed three million dollars.” Ibid. On the basis of these and other related findings, the District Court determined that the surgical-center requirement “imposes an undue burden on the right of women throughout Texas to seek a previability abortion,” and that the “admitting-privileges requirement, . . . in conjunction with the ambulatory-surgical-center requirement, imposes an undue burden on the right of women in the Rio Grande Valley, El Paso, and West Texas to seek a previability abortion.” Id., at 687. The District Court concluded that the “two provisions” would cause “the closing of almost all abortion clinics in Texas that were operating legally in the fall of 2013,” and thereby create a constitutionally “impermissible obstacle as applied to all women seeking a previability abortion” by “restricting access to previously available legal facilities.” Id., at 687–688. On August 29, 2014, the court enjoined the enforcement of the two provisions. Ibid. C On October 2, 2014, at Texas’ request, the Court of Appeals stayed the District Court’s injunction. Whole Woman’s Health v. Lakey, 769 F. 3d 285, 305. Within the next two weeks, this Court vacated the Court of Appeals’ stay (in substantial part) thereby leaving in effect the District Court’s injunction against enforcement of the surgical-center provision and its injunction against enforcement of the admitting-privileges requirement as applied to the McAllen and El Paso clinics. Whole Woman’s Health v. Lakey, 574 U. S. ___ (2014). The Court of Appeals then heard Texas’ appeal. On June 9, 2015, the Court of Appeals reversed the District Court on the merits. With minor exceptions, it found both provisions constitutional and allowed them to take effect. Whole Women’s Health v. Cole, 790 F. 3d 563, 567 (per curiam), modified, 790 F. 3d 598 (CA5 2015). Because the Court of Appeals’ decision rests upon alternative grounds and fact-related considerations, we set forth its basic reasoning in some detail. The Court of Appeals concluded: • The District Court was wrong to hold the admitting-privileges requirement unconstitutional because (except for the clinics in McAllen and El Paso) the providers had not asked them to do so, and principles of res judicata barred relief. Id., at 580–583. • Because the providers could have brought their constitutional challenge to the surgical-center provision in their earlier lawsuit, principles of res judicata also barred that claim. Id., at 581–583. • In any event, a state law “regulating previability abortion is constitutional if: (1) it does not have the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus; and (2) it is reasonably related to (or designed to further) a legitimate state interest.” Id., at 572. • “[B]oth the admitting privileges requirement and” the surgical-center requirement “were rationally related to a legitimate state interest,” namely, “rais[ing] the standard and quality of care for women seeking abortions and . . . protect[ing] the health and welfare of women seeking abortions.” Id., at 584. • The “[p]laintiffs” failed “to proffer competent evidence contradicting the legislature’s statement of a legitimate purpose.” Id., at 585. • “[T]he district court erred by substituting its own judgment [as to the provisions’ effects] for that of the legislature, albeit . . . in the name of the undue burden inquiry.” Id., at 587. • Holding the provisions unconstitutional on their face is improper because the plaintiffs had failed to show that either of the provisions “imposes an undue burden on a large fraction of women.” Id., at 590. • The District Court erred in finding that, if the surgical-center requirement takes effect, there will be too few abortion providers in Texas to meet the demand. That factual determination was based upon the finding of one of plaintiffs’ expert witnesses (Dr. Grossman) that abortion providers in Texas “ ‘will not be able to go from providing approximately 14,000 abortions annually, as they currently are, to providing the 60,000 to 70,000 abortions that are done each year in Texas once all’ ” of the clinics failing to meet the surgical-center requirement “ ‘are forced to close.’ ” Id., at 589–590. But Dr. Grossman’s opinion is (in the Court of Appeals’ view) “ ‘ipse dixit’ ”; the “ ‘record lacks any actual evidence regarding the current or future capacity of the eight clinics’ ”; and there is no “evidence in the record that” the providers that currently meet the surgical-center requirement “are operating at full capacity or that they cannot increase capacity.” Ibid. For these and related reasons, the Court of Appeals reversed the District Court’s holding that the admitting-privileges requirement is unconstitutional and its holding that the surgical-center requirement is unconstitutional. The Court of Appeals upheld in part the District Court’s more specific holding that the requirements are unconstitutional as applied to the McAllen facility and Dr. Lynn (a doctor at that facility), but it reversed the District Court’s holding that the surgical-center requirement is unconstitutional as applied to the facility in El Paso. In respect to this last claim, the Court of Appeals said that women in El Paso wishing to have an abortion could use abortion providers in nearby New Mexico. II Before turning to the constitutional question, we must consider the Court of Appeals’ procedural grounds for holding that (but for the challenge to the provisions of H. B. 2 as applied to McAllen and El Paso) petitioners were barred from bringing their constitutional challenges. A Claim Preclusion—Admitting-Privileges Requirement The Court of Appeals held that there could be no facial challenge to the admitting-privileges requirement. Because several of the petitioners here had previously brought an unsuccessful facial challenge to that requirement (namely, Abbott, 748 F. 3d, at 605; see supra, at 2–3), the Court of Appeals thought that “the principle of res judicata” applied. 790 F. 3d, at 581. The Court of Appeals also held that res judicata prevented the District Court from granting facial relief to petitioners, concluding that it was improper to “facially invalidat[e] the admitting privileges requirement,” because to do so would “gran[t] more relief than anyone requested or briefed.” Id., at 580. We hold that res judicata neither bars petitioners’ challenges to the admitting-privileges requirement nor prevents us from awarding facial relief. For one thing, to the extent that the Court of Appeals concluded that the principle of res judicata bars any facial challenge to the admitting-privileges requirement, see ibid., the court misconstrued petitioners’ claims. Petitioners did not bring a facial challenge to the admitting-privileges requirement in this case but instead challenged that requirement as applied to the clinics in McAllen and El Paso. The question is whether res judicata bars petitioners’ particular as-applied claims. On this point, the Court of Appeals concluded that res judicata was no bar, see 790 F. 3d, at 592, and we agree. The doctrine of claim preclusion (the here-relevant aspect of res judicata) prohibits “successive litigation of the very same claim” by the same parties. New Hampshire v. Maine, 532 U. S. 742, 748 (2001) . Petitioners’ postenforcement as-applied challenge is not “the very same claim” as their preenforcement facial challenge. The Restatement of Judgments notes that development of new material facts can mean that a new case and an otherwise similar previous case do not present the same claim. See Restatement (Second) of Judgments §24, Comment f (1980) (“Material operative facts occurring after the decision of an action with respect to the same subject matter may in themselves, or taken in conjunction with the antecedent facts, comprise a transaction which may be made the basis of a second action not precluded by the first”); cf. id., §20(2) (“A valid and final personal judgment for the defendant, which rests on the prematurity of the action or on the plaintiff’s failure to satisfy a precondition to suit, does not bar another action by the plaintiff instituted after the claim has matured, or the precondition has been satisfied”); id., §20, Comment k (discussing relationship of this rule with §24, Comment f ). The Courts of Appeals have used similar rules to determine the contours of a new claim for purposes of preclusion. See, e.g., Morgan v. Covington, 648 F. 3d 172, 178 (CA3 2011) (“[R]es judicata does not bar claims that are predicated on events that postdate the filing of the initial complaint”); Ellis v. CCA of Tenn. LLC, 650 F. 3d 640, 652 (CA7 2011); Bank of N. Y. v. First Millennium, Inc., 607 F. 3d 905, 919 (CA2 2010); Smith v. Potter, 513 F. 3d 781, 783 (CA7 2008); Rawe v. Liberty Mut. Fire Ins. Co., 462 F. 3d 521, 529 (CA6 2006); Manning v. Auburn, 953 F. 2d 1355, 1360 (CA11 1992). The Restatement adds that, where “important human values—such as the lawfulness of continuing personal disability or restraint—are at stake, even a slight change of circumstances may afford a sufficient basis for concluding that a second action may be brought.” §24, Comment f; see Bucklew v. Lombardi, 783 F. 3d 1120, 1127 (CA8 2015) (allowing as-applied challenge to exe-cution method to proceed notwithstanding prior facial challenge). We find this approach persuasive. Imagine a group of prisoners who claim that they are being forced to drink contaminated water. These prisoners file suit against the facility where they are incarcerated. If at first their suit is dismissed because a court does not believe that the harm would be severe enough to be unconstitutional, it would make no sense to prevent the same prisoners from bringing a later suit if time and experience eventually showed that prisoners were dying from contaminated water. Such circumstances would give rise to a new claim that the prisoners’ treatment violates the Constitution. Factual developments may show that constitutional harm, which seemed too remote or speculative to afford relief at the time of an earlier suit, was in fact indisputable. In our view, such changed circumstances will give rise to a new constitutional claim. This approach is sensible, and it is consistent with our precedent. See Abie State Bank v. Bryan, 282 U. S. 765, 772 (1931) (where “suit was brought immediately upon the enactment of the law,” “decision sustaining the law cannot be regarded as precluding a subsequent suit for the purpose of testing [its] validity . . . in the lights of the later actual experience”); cf. Lawlor v. National Screen Service Corp., 349 U. S. 322, 328 (1955) (judgment that “precludes recovery on claims arising prior to its entry” nonetheless “cannot be given the effect of extinguishing claims which did not even then exist”); United States v. Carolene Products Co., 304 U. S. 144, 153 (1938) (“[T]he constitutionality of a statute predicated upon the existence of a particular state of facts may be challenged by showing to the court that those facts have ceased to exist”); Nashville, C. & St. L. R. Co. v. Walters, 294 U. S. 405, 415 (1935) (“A statute valid as to one set of facts may be invalid as to another. A statute valid when enacted may become invalid by change in the conditions to which it is applied” (footnote omitted)); Third Nat. Bank of Louisville v. Stone 174 U. S. 432, 434 (1899) (“A question cannot be held to have been adjudged before an issue on the subject could possibly have arisen”). Justice Alito’s dissenting opinion is simply wrong that changed circumstances showing that a challenged law has an unconstitutional effect can never give rise to a new claim. See post, at 14–15 (hereinafter the dissent). Changed circumstances of this kind are why the claim presented in Abbott is not the same claim as petitioners’ claim here. The claims in both Abbott and the present case involve “important human values.” Restatement (Second) of Judgments §24, Comment f. We are concerned with H. B. 2’s “effect . . . on women seeking abortions.” Post, at 30 (Alito, J., dissenting). And that effect has changed dramatically since petitioners filed their first lawsuit. Abbott rested on facts and evidence presented to the District Court in October 2013. 748 F. 3d, at 599, n. 14 (declining to “consider any arguments” based on “developments since the conclusion of the bench trial”). Petitioners’ claim in this case rests in significant part upon later, concrete factual developments. Those developments matter. The Abbott plaintiffs brought their facial challenge to the admitting-privileges requirement prior to its enforcement—before many abortion clinics had closed and while it was still unclear how many clinics would be affected. Here, petitioners bring an as-applied challenge to the requirement after its enforcement—and after a large number of clinics have in fact closed. The postenforcement consequences of H. B. 2 were unknowable before it went into effect. The Abbott court itself recognized that “[l]ater as-applied challenges can always deal with subsequent, concrete constitutional issues.” Id., at 589. And the Court of Appeals in this case properly decided that new evidence presented by petitioners had given rise to a new claim and that petitioners’ as-applied challenges are not precluded. See 790 F. 3d, at 591 (“We now know with certainty that the non-[surgical-center] abortion facilities have actually closed and physicians have been unable to obtain admitting privileges after diligent effort”). When individuals claim that a particular statute will produce serious constitutionally relevant adverse consequences before they have occurred—and when the courts doubt their likely occurrence—the factual difference that those adverse consequences have in fact occurred can make all the difference. Compare the Fifth Circuit’s opinion in the earlier case, Abbott, supra, at 598 (“All of the major Texas cities . . . continue to have multiple clinics where many physicians will have or obtain hospital admitting privileges”), with the facts found in this case, 46 F. Supp. 3d, at 680 (the two provisions will leave Texas with seven or eight clinics). The challenge brought in this case and the one in Abbott are not the “very same claim,” and the doctrine of claim preclusion consequently doesnot bar a new challenge to the constitutionality of the admitting-privileges requirement. New Hampshire v. Maine, 532 U. S., at 748. That the litigants in Abbott did not seek review in this Court, as the dissent suggests they should have done, see post, at 10, does not prevent them from seeking review of new claims that have arisen after Abbott was decided. In sum, the Restatement, cases from the Courts of Appeals, our own precedent, and simple logic combine to convince us that res judicata does not bar this claim. The Court of Appeals also concluded that the award of facial relief was precluded by principles of res judicata. 790 F. 3d, at 581. The court concluded that the District Court should not have “granted more relief than anyone requested or briefed.” Id., at 580. But in addition to asking for as-applied relief, petitioners asked for “such other and further relief as the Court may deem just, proper, and equitable.” App. 167. Their evidence and argu-ments convinced the District Court that the provision was unconstitutional across the board. The Federal Rules of Civil Procedure state that (with an exception not relevant here) a “final judgment should grant the relief to which each party is entitled, even if the party has not demanded that relief in its pleadings.” Rule 54(c). And we have held that, if the arguments and evidence show that a statutory provision is unconstitutional on its face, an injunction prohibiting its enforcement is “proper.” Citizens United v. Federal Election Comm’n, 558 U. S. 310, 333 (2010) ; see ibid. (in “the exercise of its judicial responsibility” it may be “necessary . . . for the Court to consider the facial validity” of a statute, even though a facial challenge was not brought); cf. Fallon, As-Applied and Facial Challenges and Third-Party Standing, 113 Harv. L. Rev. 1321, 1339 (2000) (“[O]nce a case is brought, no general categorical line bars a court from making broader pronouncements of invalidity in properly ‘as-applied’ cases”). Nothing prevents this Court from awarding facial relief as the appropriate rem-edy for petitioners’ as-applied claims. B Claim Preclusion—Surgical-Center Requirement The Court of Appeals also held that claim preclusion barred petitioners from contending that the surgical-center requirement is unconstitutional. 790 F. 3d, at 583. Although it recognized that petitioners did not bring this claim in Abbott, it believed that they should have done so. The court explained that petitioners’ constitutional challenge to the surgical-center requirement and the chal-lenge to the admitting-privileges requirement mounted in Abbott “arise from the same ‘transactio[n] or series of connected transactions.’ . . . The challenges involve the same parties and abortion facilities; the challenges are governed by the same legal standards; the provisions at issue were enacted at the same time as part of the same act; the provisions were motivated by a common purpose; the provisions are administered by the same state officials; and the challenges form a convenient trial unit because they rely on a common nucleus of operative facts.” 790 F. 3d, at 581. For all these reasons, the Court of Appeals held petitioners’ challenge to H. B. 2’s surgical-center requirement was precluded. The Court of Appeals failed, however, to take account of meaningful differences. The surgical-center provision and the admitting-privileges provision are separate, distinct provisions of H. B. 2. They set forth two different, independent requirements with different enforcement dates. This Court has never suggested that challenges to two different statutory provisions that serve two different functions must be brought in a single suit. And lower courts normally treat challenges to distinct regulatory requirements as “separate claims,” even when they are part of one overarching “[g]overnment regulatory scheme.” 18 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §4408, p. 52 (2d ed. 2002, Supp. 2015); see Hamilton’s Bogarts, Inc. v. Michigan, 501 F. 3d 644, 650 (CA6 2007). That approach makes sense. The opposite approach adopted by the Court of Appeals would require treating every statutory enactment as a single transaction which a given party would only be able to challenge one time, in one lawsuit, in order to avoid the effects of claim preclusion. Such a rule would encourage a kitchen-sink approach to any litigation challenging the validity of statutes. That outcome is less than optimal—not only for litigants, but for courts. There are other good reasons why petitioners should not have had to bring their challenge to the surgical-center provision at the same time they brought their first suit. The statute gave the Texas Department of State Health Services authority to make rules implementing the surgical-center requirement. H. B. 2, §11(a), App. to Pet. forCert. 201a. At the time petitioners filed Abbott, that state agency had not yet issued any such rules. Cf. EPA v. Brown, 431 U. S. 99, 104 (1977) (per curiam); 13B Wright, supra, §3532.6, at 629 (3d ed. 2008) (most courts will not “undertake review before rules have been adopted”); Natural Resources Defense Council, Inc. v. EPA, 859 F. 2d 156, 204 (CADC 1988). Further, petitioners might well have expected that those rules when issued would contain provisions grandfathering some then-existing abortion facilities and granting full or partial waivers to others. After all, more than three quarters of non-abortion-related surgical centers had benefited from that kind of provision. See 46 F. Supp. 3d, at 680–681 (336 of 433 existing Texas surgical centers have been grandfathered or otherwise enjoy a waiver of some of the surgical-center requirements); see also App. 299–302, 443–447, 468–469. Finally, the relevant factual circumstances changed between Abbott and the present lawsuit, as we previously described. See supra, at 14–15. The dissent musters only one counterargument. According to the dissent, if statutory provisions “impos[e] the same kind of burden . . . on the same kind of right” and have mutually reinforcing effects, “it is evident that” they are “part of the same transaction” and must be challenged together. Post, at 20, 22. But for the word “evident,” the dissent points to no support for this conclusion, and we find it unconvincing. Statutes are often voluminous, with many related, yet distinct, provisions. Plaintiffs, in order to preserve their claims, need not challenge each such provision of, say, the USA PATRIOT Act, the Bipartisan Campaign Reform Act of 2002, the National Labor Relations Act, the Clean Water Act, the Antiterrorism and Effective Death Penalty Act of 1996, or the Patient Protection and Affordable Care Act in their first lawsuit. For all of these reasons, we hold that the petitioners did not have to bring their challenge to the surgical-center provision when they challenged the admitting-privileges provision in Abbott. We accordingly hold that the doctrine of claim preclusion does not prevent them from bringing that challenge now. * * * None of petitioners’ claims are barred by res judicata. Five experts in civil procedure argued, in a brief supporting petitioners’ request for certiorari, that “the panel’s procedural ruling is so clearly incorrect” that there was no reason to decline review. Brief for Professor Michael Dorf et al. as Amici Curiae 22. For all of the reasons described above, we agree that the Court of Appeals’ procedural ruling was incorrect. We consequently proceed to consider the merits of petitioners’ claims. III Undue Burden—Legal Standard We begin with the standard, as described in Casey. We recognize that the “State has a legitimate interest in seeing to it that abortion, like any other medical procedure, is performed under circumstances that insure maximum safety for the patient.” Roe v. Wade, 410 U. S. 113, 150 (1973) . But, we added, “a statute which, while furthering [a] valid state interest, has the effect of placing a substantial obstacle in the path of a woman’s choice cannot be considered a permissible means of serving its legitimate ends.” Casey, 505 U. S., at 877 (plurality opinion). Moreover, “[u]nnecessary health regulations that have the purpose or effect of presenting a substantial obstacle to a woman seeking an abortion impose an undue burden on the right.” Id., at 878. The Court of Appeals wrote that a state law is “constitutional if: (1) it does not have the purpose or effect of placing a substantial obstacle in the path of a woman seeking an abortion of a nonviable fetus; and (2) it is reasonably related to (or designed to further) a legitimate state interest.” 790 F. 3d, at 572. The Court of Appeals went on to hold that “the district court erred by substituting its own judgment for that of the legislature” when it conducted its “undue burden inquiry,” in part because “medical uncertainty underlying a statute is for resolution by legislatures, not the courts.” Id., at 587 (citing Gonzales v.Carhart, 550 U. S. 124, 163 (2007) ). The Court of Appeals’ articulation of the relevant standard is incorrect. The first part of the Court of Appeals’ test may be read to imply that a district court should not consider the existence or nonexistence of medical benefits when considering whether a regulation of abortion constitutes an undue burden. The rule announced in Casey, however, requires that courts consider the burdens a law imposes on abortion access together with the benefits those laws confer. See 505 U. S., at 887–898 (opinion of the Court) (performing this balancing with respect to a spousal notification provision); id., at 899–901 (joint opinion of O’Connor, Kennedy, and Souter, JJ.) (same balancing with respect to a parental notification provision). And the second part of the test is wrong to equate the judicial review applicable to the regulation of a constitutionally protected personal liberty with the less strict review applicable where, for example, economic legislation is at issue. See, e.g., Williamson v. Lee Optical of Okla., Inc., 348 U. S. 483, 491 (1955) . The Court of Appeals’ approach simply does not match the standard that this Court laid out in Casey, which asks courts to consider whether any burden imposed on abortion access is “undue.” The statement that legislatures, and not courts, must resolve questions of medical uncertainty is also inconsistent with this Court’s case law. Instead, the Court, when determining the constitutionality of laws regulating abortion procedures, has placed considerable weight upon evidence and argument presented in judicial proceedings. In Casey, for example, we relied heavily on the District Court’s factual findings and the research-based submissions of amici in declaring a portion of the law at issue unconstitutional. 505 U. S., at 888–894 (opinion of the Court) (discussing evidence related to the prevalence of spousal abuse in determining that a spousal notification provision erected an undue burden to abortion access). And, in Gonzales the Court, while pointing out that we must review legislative “factfinding under a deferential standard,” added that we must not “place dispositive weight” on those “findings.” 550 U. S., at 165. Gonzales went on to point out that the “Court retains an independent constitutional duty to review factual findings where constitutional rights are at stake.” Ibid. (emphasis added). Although there we upheld a statute regulating abortion, we did not do so solely on the basis of legislative findings explicitly set forth in the statute, noting that “evidence presented in the District Courts contradicts” some of the legislative findings. Id., at 166. In these circumstances, we said, “[u]ncritical deference to Congress’ factual findings . . . is inappropriate.” Ibid. Unlike in Gonzales, the relevant statute here does not set forth any legislative findings. Rather, one is left to infer that the legislature sought to further a constitutionally acceptable objective (namely, protecting women’s health). Id., at 149–150. For a district court to give significant weight to evidence in the judicial record in these circumstances is consistent with this Court’s case law. As we shall describe, the District Court did so here. It did not simply substitute its own judgment for that of the legislature. It considered the evidence in the record—including expert evidence, presented in stipulations, depositions, and testimony. It then weighed the asserted benefits against the burdens. We hold that, in so doing, the District Court applied the correct legal standard. IV Undue Burden—Admitting-Privileges Requirement Turning to the lower courts’ evaluation of the evidence, we first consider the admitting-privileges requirement. Before the enactment of H. B. 2, doctors who provided abortions were required to “have admitting privileges or have a working arrangement with a physician(s) who has admitting privileges at a local hospital in order to ensure the necessary back up for medical complications.” Tex. Admin. Code, tit. 25, §139.56 (2009) (emphasis added). The new law changed this requirement by requiring that a “physician performing or inducing an abortion . . . must, on the date the abortion is performed or induced, have active admitting privileges at a hospital that . . . is located not further than 30 miles from the location at which the abortion is performed or induced.” Tex. Health & Safety Code Ann. §171.0031(a). The District Court held that the legislative change imposed an “undue burden” on a woman’s right to have an abortion. We conclude that there is adequate legal and factual support for the District Court’s conclusion. The purpose of the admitting-privileges requirement is to help ensure that women have easy access to a hospital should complications arise during an abortion procedure. Brief for Respondents 32–37. But the District Court found that it brought about no such health-related benefit. The court found that “[t]he great weight of evidence demonstrates that, before the act’s passage, abortion in Texas was extremely safe with particularly low rates of serious complications and virtually no deaths occurring on account of the procedure.” 46 F. Supp. 3d, at 684. Thus, there was no significant health-related problem that the new law helped to cure. The evidence upon which the court based this conclusion included, among other things: • A collection of at least five peer-reviewed studies on abortion complications in the first trimester, showing that the highest rate of major complications—including those complications requiring hospital admission—was less than one-quarter of 1%. See App. 269–270. • Figures in three peer-reviewed studies showing that the highest complication rate found for the much rarer second trimester abortion was less than one-half of 1% (0.45% or about 1 out of about 200). Id., at 270. • Expert testimony to the effect that complications rarely require hospital admission, much less immediate transfer to a hospital from an outpatient clinic. Id., at 266–267 (citing a study of complications occurring within six weeks after 54,911 abortions that had been paid for by the fee-for-service California Medicaid Program finding that the incidence of complications was 2.1%, the incidence of complications requiring hospital admission was 0.23%, and that of the 54,911 abortion patients included in the study, only 15 required immediate transfer to the hospital on the day of the abortion). • Expert testimony stating that “it is extremely unlikely that a patient will experience a serious complication at the clinic that requires emergent hospitalization” and “in the rare case in which [one does], the quality of care that the patient receives is not affected by whether the abortion provider has admitting privileges at the hospital.” Id., at 381. • Expert testimony stating that in respect to surgical abortion patients who do suffer complications requiring hospitalization, most of these complications occur in the days after the abortion, not on the spot. See id., at 382; see also id., at 267. • Expert testimony stating that a delay before the onset of complications is also expected for medical abortions, as “abortifacient drugs take time to exert their effects, and thus the abortion itself almost always occurs after the patient has left the abortion facility.” Id., at 278. • Some experts added that, if a patient needs a hospital in the day or week following her abortion, she will likely seek medical attention at the hospital nearest her home. See, e.g., id., at 153. We have found nothing in Texas’ record evidence that shows that, compared to prior law (which required a “working arrangement” with a doctor with admitting privileges), the new law advanced Texas’ legitimate interest in protecting women’s health. We add that, when directly asked at oral argument whether Texas knew of a single instance in which the new requirement would have helped even one woman obtain better treatment, Texas admitted that there was no evidence in the record of such a case. See Tr. of Oral Arg. 47. This answer is consistent with the findings of the other Federal District Courts that have considered the health benefits of other States’ similar admitting-privileges laws. See Planned Parenthood of Wis., Inc. v. Van Hollen, 94 F. Supp. 3d 949, 953 (WD Wis. 2015), aff’d sub nom. Planned Parenthood of Wis., Inc. v. Schimel, 806 F. 3d 908 (CA7 2015); Planned Parenthood Southeast, Inc. v. Strange, 33 F. Supp. 3d 1330, 1378 (MD Ala. 2014). At the same time, the record evidence indicates that the admitting-privileges requirement places a “substantial obstacle in the path of a woman’s choice.” Casey, 505 U. S., at 877 (plurality opinion). The District Court found, as of the time the admitting-privileges requirement began to be enforced, the number of facilities providing abortions dropped in half, from about 40 to about 20. 46 F. Supp. 3d, at 681. Eight abortion clinics closed in the months leading up to the requirement’s effective date. See App. 229–230; cf. Brief for Planned Parenthood Federation of America et al. as Amici Curiae 14 (noting that abortion facilities in Waco, San Angelo, and Midland no longer operate because Planned Parenthood is “unable to find local physicians in those communities with privileges who are willing to provide abortions due to the size of those communities and the hostility that abortion providers face”). Eleven more closed on the day the admitting-privileges requirement took effect. See App. 229–230; Tr. of Oral Arg. 58. Other evidence helps to explain why the new requirement led to the closure of clinics. We read that other evidence in light of a brief filed in this Court by the Soci-ety of Hospital Medicine. That brief describes the undisputed general fact that “hospitals often condition admitting privileges on reaching a certain number of admissions per year.” Brief for Society of Hospital Medicine et al. as Amici Curiae 11. Returning to the District Court record, we note that, in direct testimony, the president of Nova Health Systems, implicitly relying on this general fact, pointed out that it would be difficult for doctors regularly performing abortions at the El Paso clinic to obtain admitting privileges at nearby hospitals because “[d]uring the past 10 years, over 17,000 abortion procedures were performed at the El Paso clinic [and n]ot a single one of those patients had to be transferred to a hospital for emergency treatment, much less admitted to the hospital.” App. 730. In a word, doctors would be unable to maintain admitting privileges or obtain those privileges for the future, because the fact that abortions are so safe meant that providers were unlikely to have any patients to admit. Other amicus briefs filed here set forth without dispute other common prerequisites to obtaining admitting privileges that have nothing to do with ability to perform medical procedures. See Brief for Medical Staff Professionals as Amici Curiae 20–25 (listing, for example, requirements that an applicant has treated a high number of patients in the hospital setting in the past year, clinical data requirements, residency requirements, and other discretionary factors); see also Brief for American College of Obstetricians and Gynecologists et al. as Amici Curiae 16 (ACOG Brief) (“[S]ome academic hospitals will only allow medical staff membership for clinicians who also . . . accept faculty appointments”). Again, returning to the District Court record, we note that Dr. Lynn of the McAllen clinic, a veteran obstetrics and gynecology doctor who estimates that he has delivered over 15,000 babies in his 38 years in practice was unable to get admitting privileges at any of the seven hospitals within 30 miles of his clinic. App. 390–394. He was refused admitting privileges at a nearby hospital for reasons, as the hospital wrote, “not based on clinical competence considerations.” Id., at 393–394 (emphasis deleted). The admitting-privileges requirement does not serve any relevant credentialing function. In our view, the record contains sufficient evidence that the admitting-privileges requirement led to the closure of half of Texas’ clinics, or thereabouts. Those closures meant fewer doctors, longer waiting times, and increased crowding. Record evidence also supports the finding that after the admitting-privileges provision went into effect, the “number of women of reproductive age living in a county . . . more than 150 miles from a provider increased from approximately 86,000 to 400,000 . . . and the number of women living in a county more than 200 miles from a provider from approximately 10,000 to 290,000.” 46 F. Supp. 3d, at 681. We recognize that increased driving distances do not always constitute an “undue burden.” See Casey, 505 U. S., at 885–887 (joint opinion of O’Connor, Kennedy, and Souter, JJ.). But here, those increases are but one additional burden, which, when taken together with others that the closings brought about, and when viewed in light of the virtual absence of any health benefit, lead us to conclude that the record adequately supports the District Court’s “undue burden” conclusion. Cf. id., at 895 (opinion of the Court) (finding burden “undue” when requirement places “substantial obstacle to a woman’s choice” in “a large fraction of the cases in which” it “is relevant”). The dissent’s only argument why these clinic closures, as well as the ones discussed in Part V, infra, may not have imposed an undue burden is this: Although “H. B. 2 caused the closure of some clinics,” post, at 26 (emphasis added), other clinics may have closed for other reasons (so we should not “actually count” the burdens resulting from those closures against H. B. 2), post, at 30–31. But petitioners satisfied their burden to present evidence of causation by presenting direct testimony as well as plausible inferences to be drawn from the timing of the clinic closures. App. 182–183, 228–231. The District Court credited that evidence and concluded from it that H. B. 2 in factled to the clinic closures. 46 F. Supp. 3d, at 680–681. The dissent’s speculation that perhaps other evidence, not presented at trial or credited by the District Court, might have shown that some clinics closed for unrelated reasons does not provide sufficient ground to disturb the District Court’s factual finding on that issue. In the same breath, the dissent suggests that one benefit of H. B. 2’s requirements would be that they might “force unsafe facilities to shut down.” Post, at 26. To support that assertion, the dissent points to the Kermit Gosnell scandal. Gosnell, a physician in Pennsylvania, was convicted of first-degree murder and manslaughter. He “staffed his facility with unlicensed and indifferent workers, and then let them practice medicine unsupervised” and had “[d]irty facilities; unsanitary instruments; an absence of functioning monitoring and resuscitation equipment; the use of cheap, but dangerous, drugs; illegal procedures; and inadequate emergency access for when things inevitably went wrong.” Report of Grand Juryin No. 0009901–2008 (1st Jud. Dist. Pa., Jan. 14,2011), p. 24, online at http://www.phila.gov/districtattorney/pdfs/grandjurywomensmedical.pdf (as last visited June 24, 2016). Gosnell’s behavior was terribly wrong. But there is no reason to believe that an extra layerof regulation would have affected that behavior. Deter-mined wrongdoers, already ignoring existing stat-utes and safety measures, are unlikely to be convinced to adopt safe practices by a new overlay of regulations. Regardless, Gosnell’s deplorable crimes could escape detection only because his facility went uninspected for more than 15 years. Id., at 20. Pre-existing Texas law already contained numerous detailed regulations covering abortion facilities, including a requirement that facilities be inspected at least annually. See infra, at 28 (describing those regulations). The record contains nothing to suggest that H. B. 2 would be more effective than pre-existing Texas law at deterring wrongdoers like Gosnell from criminal behavior. V Undue Burden—Surgical-Center Requirement The second challenged provision of Texas’ new law sets forth the surgical-center requirement. Prior to enactment of the new requirement, Texas law required abortion facilities to meet a host of health and safety requirements. Under those pre-existing laws, facilities were subject to annual reporting and recordkeeping requirements, see Tex. Admin. Code, tit. 25, §§139.4, 139.5, 139.55, 139.58; a quality assurance program, see §139.8; personnel policies and staffing requirements, see §§139.43, 139.46; physical and environmental requirements, see §139.48; infection control standards, see §139.49; disclosure requirements, see §139.50; patient-rights standards, see §139.51; and medical- and clinical-services standards, see §139.53, including anesthesia standards, see §139.59. These requirements are policed by random and announced inspections, at least annually, see §§139.23, 139.31; Tex. Health & Safety Code Ann. §245.006(a) (West 2010), as well as administrative penalties, injunctions, civil penalties,and criminal penalties for certain violations, see Tex. Admin. Code, tit. 25, §139.33; Tex. Health & Safety Code Ann. §245.011 (criminal penalties for certain reporting violations). H. B. 2 added the requirement that an “abortion facility” meet the “minimum standards . . . for ambulatory surgical centers” under Texas law. §245.010(a) (West Cum. Supp. 2015). The surgical-center regulations include, among other things, detailed specifications relating to the size of the nursing staff, building dimensions, and other building requirements. The nursing staff must comprise at least “an adequate number of [registered nurses] on duty to meet the following minimum staff requirements: director of the department (or designee), and supervisory and staff personnel for each service area to assure the immediate availability of [a registered nurse] for emergency care or for any patient when needed,” Tex. Admin. Code, tit. 25, §135.15(a)(3) (2016), as well as “a second individual on duty on the premises who is trained and currently certified in basic cardiac life support until all patients have been discharged from the facility” for facilities that provide moderate sedation, such as most abortion facilities, §135.15(b)(2)(A). Facilities must include a full surgical suite with an operating room that has “a clear floor area of at least 240 square feet” in which “[t]he minimum clear dimension between built-in cabinets, counters, and shelves shall be 14 feet.” §135.52(d)(15)(A). There must be a preoperative patient holding room and a postoperative recovery suite. The former “shall be provided and arranged in a one-way traffic pattern so that patients entering from outside the surgical suite can change, gown, and move directly into the restricted corridor of the surgical suite,” §135.52(d)(10)(A), and the latter “shall be arranged to provide a one-way traffic pattern from the restricted surgical corridor to the postoperative recovery suite, and then to the extended observation rooms or discharge,” §135.52(d)(9)(A). Surgical centers must meet numerous other spatial requirements, see generally §135.52, including specific corridor widths, §135.52(e)(1)(B)(iii). Surgical centers must also have an advanced heating, ventilation, and air conditioning system, §135.52(g)(5), and must satisfy particular piping system and plumbing requirements, §135.52(h). Dozens of other sections list additional requirements that apply to surgical centers. See generally §§135.1–135.56. There is considerable evidence in the record supporting the District Court’s findings indicating that the statutory provision requiring all abortion facilities to meet all surgical-center standards does not benefit patients and is not necessary. The District Court found that “risks are not appreciably lowered for patients who undergo abortions at ambulatory surgical centers as compared to nonsurgical-center facilities.” 46 F. Supp. 3d, at 684. The court added that women “will not obtain better care or experience more frequent positive outcomes at an ambulatory surgical center as compared to a previously licensed facility.” Ibid. And these findings are well supported. The record makes clear that the surgical-center requirement provides no benefit when complications arise in the context of an abortion produced through medication. That is because, in such a case, complications would almost always arise only after the patient has left the facil-ity. See supra, at 23; App. 278. The record also contains evidence indicating that abortions taking place in an abortion facility are safe—indeed, safer than numerous procedures that take place outside hospitals and to which Texas does not apply its surgical-center requirements. See, e.g., id., at 223–224, 254, 275–279. The total number of deaths in Texas from abortions was five in the period from 2001 to 2012, or about one every two years (that is to say, one out of about 120,000 to 144,000 abortions). Id., at 272. Nationwide, childbirth is 14 times more likely than abortion to result in death, ibid., but Texas law allows a midwife to oversee childbirth in the patient’s own home. Colonoscopy, a procedure that typically takes place outside a hospital (or surgical center) setting, has a mortality rate 10 times higher than an abortion. Id., at 276–277; see ACOG Brief 15 (the mortality rate for liposuction, another outpatient procedure, is 28 times higher than the mortal-ity rate for abortion). Medical treatment after an incomplete miscarriage often involves a procedure identical to that involved in a nonmedical abortion, but it often takes place outside a hospital or surgical center. App. 254; see ACOG Brief 14 (same). And Texas partly or wholly grandfathers (or waives in whole or in part the surgical-center requirement for) about two-thirds of the facilities to which the surgical-center standards apply. But it neither grandfathers nor provides waivers for any of the facilities that perform abortions. 46 F. Supp. 3d, at 680–681; see App. 184. These facts indicate that the surgical-center provision imposes “a requirement that simply is not based on differences” between abortion and other surgical procedures “that are reasonably related to” preserving women’s health, the asserted “purpos[e] of the Act in which it is found.” Doe, 410 U. S., at 194 (quoting Morey v. Doud, 354 U. S. 457, 465 (1957) ; internal quotation marks omitted). Moreover, many surgical-center requirements are inappropriate as applied to surgical abortions. Requiring scrub facilities; maintaining a one-way traffic pattern through the facility; having ceiling, wall, and floor fin-ishes; separating soiled utility and sterilization rooms; and regulating air pressure, filtration, and humidity control can help reduce infection where doctors conduct procedures that penetrate the skin. App. 304. But abortions typically involve either the administration of medicines or procedures performed through the natural opening of the birth canal, which is itself not sterile. See id., at 302–303. Nor do provisions designed to safeguard heavily sedated patients (unable to help themselves) during fire emergencies, see Tex. Admin. Code, tit. 25, §135.41; App. 304, provide any help to abortion patients, as abortion facilities do not use general anesthesia or deep sedation, id., at 304–305. Further, since the few instances in which serious complications do arise following an abortion almost always require hospitalization, not treatment at a surgical center, id., at 255–256, surgical-center standards will not help in those instances either. The upshot is that this record evidence, along with the absence of any evidence to the contrary, provides ample support for the District Court’s conclusion that “[m]any of the building standards mandated by the act and its implementing rules have such a tangential relationship to patient safety in the context of abortion as to be nearly arbitrary.” 46 F. Supp. 3d, at 684. That conclusion, along with the supporting evidence, provides sufficient support for the more general conclusion that the surgical-center requirement “will not [provide] better care or . . . more frequent positive outcomes.” Ibid. The record evidence thus supports the ultimate legal conclusion that thesurgical-center requirement is not necessary. At the same time, the record provides adequate evidentiary support for the District Court’s conclusion that the surgical-center requirement places a substantial obstacle in the path of women seeking an abortion. The parties stipulated that the requirement would further reduce the number of abortion facilities available to seven or eight facilities, located in Houston, Austin, San Antonio, and Dallas/Fort Worth. See App. 182–183. In the District Court’s view, the proposition that these “seven or eight providers could meet the demand of the entire State stretches credulity.” 46 F. Supp. 3d, at 682. We take this statement as a finding that these few facilities could not “meet” that “demand.” The Court of Appeals held that this finding was “clearly erroneous.” 790 F. 3d, at 590. It wrote that the finding rested upon the “ ‘ipse dixit’ ” of one expert, Dr. Grossman, and that there was no evidence that the current surgical centers (i.e., the seven or eight) are operating at full capacity or could not increase capacity. Ibid. Unlike the Court of Appeals, however, we hold that the record provides adequate support for the District Court’s finding. For one thing, the record contains charts and oral testimony by Dr. Grossman, who said that, as a result of the surgical-center requirement, the number of abortions that the clinics would have to provide would rise from “ ‘14,000 abortions annually’ ” to “ ‘60,000 to 70,000’ ”—an increase by a factor of about five. Id., at 589–590. The District Court credited Dr. Grossman as an expert witness. See 46 F. Supp. 3d, at 678–679, n. 1; id., at 681, n. 4 (finding “indicia of reliability” in Dr. Grossman’s conclusions). The Federal Rules of Evidence state that an expert may testify in the “form of an opinion” as long as that opinion rests upon “sufficient facts or data” and “reliable principles and methods.” Rule 702. In this case Dr. Grossman’s opinion rested upon his participation, along with other university researchers, in research that tracked “the number of open facilities providing abortion care in the state by . . . requesting information from the Texas Department of State Health Services . . . [, t]hrough interviews with clinic staff[,] and review of publicly available information.” App. 227. The District Court acted within its legal authority in determining that Dr. Grossman’s testimony was admissible. See Fed. Rule Evid. 702; see also Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U. S. 579, 589 (1993) (“[U]nder the Rules the trial judge must ensure that any and all [expert] evidence admitted is not only relevant, but reliable”); 29 C. Wright & V. Gold, Federal Practice and Procedure: Evidence §6266, p. 302 (2016) (“Rule 702 impose[s] on the trial judge additional responsibility to determine whether that [expert] testimony is likely to promote accurate factfinding”). For another thing, common sense suggests that, more often than not, a physical facility that satisfies a certain physical demand will not be able to meet five times that demand without expanding or otherwise incurring significant costs. Suppose that we know only that a certain grocery store serves 200 customers per week, that a certain apartment building provides apartments for 200 families, that a certain train station welcomes 200 trains per day. While it is conceivable that the store, the apartment building, or the train station could just as easily provide for 1,000 customers, families, or trains at no significant additional cost, crowding, or delay, most of us would find this possibility highly improbable. The dissent takes issue with this general, intuitive point by arguing that many places operate below capacity and that in any event, facilities could simply hire additional providers. See post, at 32. We disagree that, according to common sense, medical facilities, well known for their wait times, operate below capacity as a general matter. And the fact that so many facilities were forced to close by the admitting-privileges requirement means that hiring more physi-cians would not be quite as simple as the dissent suggests. Courts are free to base their findings on commonsense inferences drawn from the evidence. And that is what the District Court did here. The dissent now seeks to discredit Dr. Grossman by pointing out that a preliminary prediction he made in his testimony in Abbott about the effect of the admitting-privileges requirement on capacity was not borne out after that provision went into effect. See post, at 31, n. 22. If every expert who overestimated or underestimated any figure could not be credited, courts would struggle to find expert assistance. Moreover, making a hypothesis—and then attempting to verify that hypothesis with further studies, as Dr. Grossman did—is not irresponsible. It is an essential element of the scientific method. The District Court’s decision to credit Dr. Grossman’s testimony was sound, particularly given that Texas provided no credible experts to rebut it. See 46 F. Supp. 3d, at 680, n. 3 (declining to credit Texas’ expert witnesses, in part because Vincent Rue, a nonphysician consultant for Texas, had exercised “considerable editorial and discretionary control over the contents of the experts’ reports”). Texas suggests that the seven or eight remaining clinics could expand sufficiently to provide abortions for the 60,000 to 72,000 Texas women who sought them each year. Because petitioners had satisfied their burden, the obligation was on Texas, if it could, to present evidence rebutting that issue to the District Court. Texas admitted that it presented no such evidence. Tr. of Oral Arg. 46. Instead, Texas argued before this Court that one new clinic now serves 9,000 women annually. Ibid. In addition to being outside the record, that example is not representative. The clinic to which Texas referred apparently cost $26 million to construct—a fact that even more clearly demonstrates that requiring seven or eight clinics to serve five times their usual number of patients does indeed represent an undue burden on abortion access. See Planned Parenthood Debuts New Building: Its $26 Million Center in Houston is Largest of Its Kind in U. S., Houston Chronicle, May 21, 2010, p. B1. Attempting to provide the evidence that Texas did not, the dissent points to an exhibit submitted in Abbott showing that three Texas surgical centers, two in Dallas as well as the $26-million facility in Houston, are each capable of serving an average of 7,000 patients per year. See post, at 33–35. That “average” is misleading. In addition to including the Houston clinic, which does not represent most facilities, it is underinclusive. It ignores the evidence as to the Whole Woman’s Health surgical-center facility in San Antonio, the capacity of which is described as “severely limited.” The exhibit does nothing to rebut the commonsense inference that the dramatic decline in the number of available facilities will cause a shortfall in capacity should H. B. 2 go into effect. And facilities that were still operating after the effective date of the admitting-privileges provision were not able to accommodate increased demand. See App. 238; Tr. of Oral Arg. 30–31; Brief for National Abortion Federation et al. as Amici Curiae 17–20 (citing clinics’ experiences since theadmitting-privileges requirement went into effect of 3-week wait times, staff burnout, and waiting rooms so full, patients had to sit on the floor or wait outside). More fundamentally, in the face of no threat to women’s health, Texas seeks to force women to travel long distances to get abortions in crammed-to-capacity superfacilities. Patients seeking these services are less likely to get the kind of individualized attention, serious conversation, and emotional support that doctors at less taxed facilities may have offered. Healthcare facilities and medical professionals are not fungible commodities. Surgical centers attempting to accommodate sudden, vastly increased demand, see 46 F. Supp. 3d, at 682, may find that quality of care declines. Another commonsense inference that the District Court made is that these effects would be harmful to, not supportive of, women’s health. See id., at 682–683. Finally, the District Court found that the costs that a currently licensed abortion facility would have to incur to meet the surgical-center requirements were considerable, ranging from $1 million per facility (for facilities with adequate space) to $3 million per facility (where additional land must be purchased). Id., at 682. This evidence supports the conclusion that more surgical centers will not soon fill the gap when licensed facilities are forced to close. We agree with the District Court that the surgical-center requirement, like the admitting-privileges requirement, provides few, if any, health benefits for women, poses a substantial obstacle to women seeking abortions, and constitutes an “undue burden” on their constitutional right to do so. VI We consider three additional arguments that Texas makes and deem none persuasive. First, Texas argues that facial invalidation of both challenged provisions is precluded by H. B. 2’s severability clause. See Brief for Respondents 50–52. The severability clause says that “every provision, section, subsection, sentence, clause, phrase, or word in this Act, and every application of the provision in this Act, are severable from each other.” H. B. 2, §10(b), App. to Pet. for Cert. 200a. It further provides that if “any application of any provision in this Act to any person, group of persons, or circumstances is found by a court to be invalid, the remaining applications of that provision to all other persons and circumstances shall be severed and may not be affected.” Ibid. That language, Texas argues, means that facial invalidation of parts of the statute is not an option; instead, it says, the severability clause mandates a more narrowly tailored judicial remedy. But the challenged provisions of H. B. 2 close most of the abortion facilities in Texas and place added stress on those facilities able to remain open. They vastly increase the obstacles confronting women seeking abortions in Texas without providing any benefit to women’s health capable of withstanding any meaningful scrutiny. The provisions are unconstitutional on their face: Including a severability provision in the law does not change that conclusion. Severability clauses, it is true, do express the enacting legislature’s preference for a narrow judicial remedy. As a general matter, we attempt to honor that preference. But our cases have never required us to proceed application by conceivable application when confronted with a facially unconstitutional statutory provision. “We have held that a severability clause is an aid merely; not an inexorable command.” Reno v. American Civil Liberties Union, 521 U. S. 844 –885, n. 49 (1997) (internal quotation marks omitted). Indeed, if a severability clause could impose such a requirement on courts, legislatures would easily be able to insulate unconstitutional statutes from most facial review. See ibid. (“It would certainly be dangerous if the legislature could set a net large enough to catch all possible offenders, and leave it to the courts to step inside and say who could be rightfully detained, and who should be set at large. This would, to some extent, substitute the judicial for the legislative department of the government” (internal quotation marks omitted)). A severability clause is not grounds for a court to “devise a judicial remedy that . . . entail[s] quintessentially legislative work.” Ayotte v. Planned Parenthood of Northern New Eng., 546 U. S. 320, 329 (2006) . Such an approach would inflict enormous costs on both courts and litigants, who would be required to proceed in this manner whenever a single application of a law might be valid. We reject Texas’ invitation to pave the way for legislatures to immunize their statutes from facial review. Texas similarly argues that instead of finding the entire surgical-center provision unconstitutional, we should invalidate (as applied to abortion clinics) only those spe-cific surgical-center regulations that unduly burden the pro-vision of abortions, while leaving in place other surgical-center regulations (for example, the reader could pickany of the various examples provided by the dissent, see post, at 42–43). See Brief for Respondents 52–53. As we have explained, Texas’ attempt to broadly draft a requirement to sever “applications” does not require us to proceed in piecemeal fashion when we have found the statutory provisions at issue facially unconstitutional. Nor is that approach to the regulations even required by H. B. 2 itself. The statute was meant to require abortion facilities to meet the integrated surgical-center standards—not some subset thereof. The severability clause refers to severing applications of words and phrases in the Act, such as the surgical-center requirement as a whole. See H. B. 2, §4, App. to Pet. for Cert. 194a. It does not say that courts should go through the individual components of the different, surgical-center statute, let alone the individual regulations governing surgical centers to see whether those requirements are severable from each other as applied to abortion facilities. Facilities subject to some subset of those regulations do not qualify as surgical centers. And the risk of harm caused by inconsistent application of only a fraction of interconnected regulations counsels against doing so. Second, Texas claims that the provisions at issue here do not impose a substantial obstacle because the women affected by those laws are not a “large fraction” of Texan women “of reproductive age,” which Texas reads Casey to have required. See Brief for Respondents 45, 48. But Casey used the language “large fraction” to refer to “a large fraction of cases in which [the provision at issue] is relevant,” a class narrower than “all women,” “pregnant women,” or even “the class of women seeking abortions identified by the State.” 505 U. S., at 894–895 (opinion of the Court) (emphasis added). Here, as in Casey, the relevant denominator is “those [women] for whom [the provision] is an actual rather than an irrelevant restriction.” Id., at 895. Third, Texas looks for support to Simopoulos v. Virginia, 462 U. S. 506 (1983) , a case in which this Court uphelda surgical-center requirement as applied to second-trimester abortions. This case, however, unlike Simopoulos, involves restrictions applicable to all abortions, not simply to those that take place during the second trimester. Most abortions in Texas occur in the first trimester, not the second. App. 236. More importantly, in Casey we discarded the trimester framework, and we now use “viability” as the relevant point at which a State may begin limiting women’s access to abortion for reasons unrelated to maternal health. 505 U. S., at 878 (plurality opinion). Because the second trimester includes time that is both previability and postviability, Simopoulos cannot provide clear guidance. Further, the Court in Simopoulos found that the petitioner in that case, unlike petitioners here, had waived any argument that the regulation did not significantly help protect women’s health. 462 U. S., at 517. * * * For these reasons the judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

579.US.2015_15-5040

Petitioner Williams was convicted of the 1984 murder of Amos Norwood and sentenced to death. During the trial, the then-district attorney of Philadelphia, Ronald Castille, approved the trial prosecutor’s request to seek the death penalty against Williams. Over the next 26 years, Williams’s conviction and sentence were upheld on direct appeal, state postconviction review, and federal habeas review. In 2012, Williams filed a successive petition pursuant to Pennsylvania’s Post Conviction Relief Act (PCRA), arguing that the prosecutor had obtained false testimony from his codefendant and suppressed material, exculpatory evidence in violation of Brady v. Maryland, 373 U. S. 83 . Finding that the trial prosecutor had committed Brady violations, the PCRA court stayed Williams’s execution and ordered a new sentencing hearing. The Commonwealth asked the Pennsylvania Supreme Court, whose chief justice was former District Attorney Castille, to vacate the stay. Williams filed a response, along with a motion asking Chief Justice Castille to recuse himself or, if he declined to do so, to refer the motion to the full court for decision. Without explanation, the chief justice denied Williams’s motion for recusal and the request for its referral. He then joined the State Supreme Court opinion vacating the PCRA court’s grant of penalty-phase relief and reinstating Williams’s death sentence. Two weeks later, Chief Justice Castille retired from the bench. Held: 1. Chief Justice Castille’s denial of the recusal motion and his subsequent judicial participation violated the Due Process Clause of the Fourteenth Amendment. Pp. 5–12. (a) The Court’s due process precedents do not set forth a specific test governing recusal when a judge had prior involvement in a case as a prosecutor; but the principles on which these precedents rest dictate the rule that must control in the circumstances here: Under the Due Process Clause there is an impermissible risk of actual bias when a judge earlier had significant, personal involvement as a prosecutor in a critical decision regarding the defendant’s case. The Court applies an objective standard that requires recusal when the likelihood of bias on the part of the judge “is too high to be constitutionally tolerable.” Caperton v. A. T. Massey Coal Co., 556 U. S. 868 . A constitutionally intolerable probability of bias exists when the same person serves as both accuser and adjudicator in a case. See In re Murchison, 349 U. S. 133 –137. No attorney is more integral to the accusatory process than a prosecutor who participates in a major adversary decision. As a result, a serious question arises as to whether a judge who has served as an advocate for the State in the very case the court is now asked to adjudicate would be influenced by an improper, if inadvertent, motive to validate and preserve the result obtained through the adversary process. In these circumstances, neither the involvement of multiple actors in the case nor the passage of time relieves the former prosecutor of the duty to withdraw in order to ensure the neutrality of the judicial process in determining the consequences his or her own earlier, critical decision may have set in motion. Pp. 5–8. (b) Because Chief Justice Castille’s authorization to seek the death penalty against Williams amounts to significant, personal involvement in a critical trial decision, his failure to recuse from Williams’s case presented an unconstitutional risk of bias. The decision to pursue the death penalty is a critical choice in the adversary process, and Chief Justice Castille had a significant role in this decision. Without his express authorization, the Commonwealth would not have been able to pursue a death sentence against Williams. Given the importance of this decision and the profound consequences it carries, a responsible prosecutor would deem it to be a most significant exercise of his or her official discretion. The fact that many jurisdictions, including Pennsylvania, have statutes and professional codes of conduct that already require recusal under the circumstances of this case suggests that today’s decision will not occasion a significant change in recusal practice. Pp. 9–12. 2. An unconstitutional failure to recuse constitutes structural error that is “not amenable” to harmless-error review, regardless of whether the judge’s vote was dispositive, Puckett v. United States, 556 U. S. 129 . Because an appellate panel’s deliberations are generally confidential, it is neither possible nor productive to inquire whether the jurist in question might have influenced the views of his or her colleagues during the decisionmaking process. Indeed, one purpose of judicial confidentiality is to ensure that jurists can reexamine old ideas and suggest new ones, while both seeking to persuade and being open to persuasion by their colleagues. It does not matter whether the disqualified judge’s vote was necessary to the disposition of the case. The fact that the interested judge’s vote was not dispositive may mean only that the judge was successful in persuading most members of the court to accept his or her position—an outcome that does not lessen the unfairness to the affected party. A multimember court must not have its guarantee of neutrality undermined, for the appearance of bias demeans the reputation and integrity not just of one jurist, but of the larger institution of which he or she is a part. Because Chief Justice Castille’s participation in Williams’s case was an error that affected the State Supreme Court’s whole adjudicatory framework below, Williams must be granted an opportunity to present his claims to a court unburdened by any “possible temptation . . . not to hold the balance nice, clear and true between the State and the accused,” Tumey v. Ohio, 273 U. S. 510 . Pp. 12–14. __ Pa. __, 105 A. 3d 1234, vacated and remanded. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Alito, J., joined. Thomas, J., filed a dissenting opinion.

In this case, the Supreme Court of Pennsylvania vacated the decision of a postconviction court, which had granted relief to a prisoner convicted of first-degree murder and sentenced to death. One of the justices on the State Supreme Court had been the district attorney who gave his official approval to seek the death penalty in the prisoner’s case. The justice in question denied the prisoner’s motion for recusal and participated in the decision to deny relief. The question presented is whether the justice’s denial of the recusal motion and his subsequent judicial participation violated the Due Process Clause of the Fourteenth Amendment. This Court’s precedents set forth an objective standard that requires recusal when the likelihood of bias on the part of the judge “ ‘is too high to be constitutionally tolerable.’ ” Caperton v. A. T. Massey Coal Co., 556 U. S. 868, 872 (2009) (quoting Withrow v. Larkin, 421 U. S. 35, 47 (1975) ). Applying this standard, the Court concludes that due process compelled the justice’s recusal. I Petitioner is Terrance Williams. In 1984, soon after Williams turned 18, he murdered 56-year-old Amos Norwood in Philadelphia. At trial, the Commonwealth presented evidence that Williams and a friend, Marc Draper, had been standing on a street corner when Norwood drove by. Williams and Draper requested a ride home from Norwood, who agreed. Draper then gave Norwood false directions that led him to drive toward a cemetery. Williams and Draper ordered Norwood out of the car and into the cemetery. There, the two men tied Norwood in his own clothes and beat him to death. Testifying for the Commonwealth, Draper suggested that robbery was the motive for the crime. Williams took the stand in his own defense, stating that he was not involved in the crime and did not know the victim. During the trial, the prosecutor requested permission from her supervisors in the district attorney’s office to seek the death penalty against Williams. To support the request, she prepared a memorandum setting forth the details of the crime, information supporting two statutory aggravating factors, and facts in mitigation. After reviewing the memorandum, the then-district attorney of Philadelphia, Ronald Castille, wrote this note at the bottom of the document: “Approved to proceed on the death penalty.” App. 426a. During the penalty phase of the trial, the prosecutor argued that Williams deserved a death sentence because he killed Norwood “ ‘for no other reason but that a kind man offered him a ride home.’ ” Brief for Petitioner 7. The jurors found two aggravating circumstances: that the murder was committed during the course of a robbery and that Williams had a significant history of violent felony convictions. That criminal history included a previous conviction for a murder he had committed at age 17. The jury found no mitigating circumstances and sentenced Williams to death. Over a period of 26 years, Williams’s conviction and sentence were upheld on direct appeal, state postconviction review, and federal habeas review. In 2012, Williams filed a successive petition pursuant to Pennsylvania’s Post Conviction Relief Act (PCRA), 42 Pa. Cons. Stat. §9541 et seq. (2007). The petition was based on new information from Draper, who until then had refused to speak with Williams’s attorneys. Draper told Williams’s counsel that he had informed the Commonwealth before trial that Williams had been in a sexual relationship with Norwood and that the relationship was the real motive for Norwood’s murder. According to Draper, the Commonwealth had instructed him to give false testimony that Williams killed Norwood to rob him. Draper also admitted he had received an undisclosed benefit in exchange for his testimony: the trial prosecutor had promised to write a letter to the state parole board on his behalf. At trial, the prosecutor had elicited testimony from Draper indicating that his only agreement with the prosecution was to plead guilty in exchange for truthful testimony. No mention was made of the additional promise to write the parole board. The Philadelphia Court of Common Pleas, identified in the proceedings below as the PCRA court, held an evidentiary hearing on Williams’s claims. Williams alleged in his petition that the prosecutor had procured false testimony from Draper and suppressed evidence regarding Norwood’s sexual relationship with Williams. At the hearing, both Draper and the trial prosecutor testified regarding these allegations. The PCRA court ordered the district attorney’s office to produce the previously undisclosed files of the prosecutor and police. These documents included the trial prosecutor’s sentencing memorandum, bearing then-District Attorney Castille’s authorization to pursue the death penalty. Based on the Commonwealth’s files and the evidentiary hearing, the PCRA court found that the trial prosecutor had suppressed material, exculpatory evidence in violation of Brady v. Maryland, 373 U. S. 83 (1963) , and engaged in “prosecutorial gamesmanship.” App. 168a. The court stayed Williams’s execution and ordered a new sentencing hearing. Seeking to vacate the stay of execution, the Commonwealth submitted an emergency application to the Pennsylvania Supreme Court. By this time, almost three decades had passed since Williams’s prosecution. Castille had been elected to a seat on the State Supreme Court and was serving as its chief justice. Williams filed a response to the Commonwealth’s application. The disclosure of the trial prosecutor’s sentencing memorandum in the PCRA proceedings had alerted Williams to Chief Justice Castille’s involvement in the decision to seek a death sentence in his case. For this reason, Williams also filed a motion asking Chief Justice Castille to recuse himself or, if he declined to do so, to refer the recusal motion to the full court for decision. The Commonwealth opposed Williams’s recusal motion. Without explanation, Chief Justice Castille denied the motion for recusal and the request for its referral. Two days later, the Pennsylvania Supreme Court denied the application to vacate the stay and ordered full briefing on the issues raised in the appeal. The State Supreme Court then vacated the PCRA court’s order granting penalty-phase relief and reinstated Williams’s death sentence. Chief Justice Castille and Justices Baer and Stevens joined the majority opinion written by Justice Eakin. Justices Saylor and Todd concurred in the result without issuing a separate opinion. See ___ Pa. ___, ___, 105 A. 3d 1234, 1245 (2014). Chief Justice Castille authored a concurrence. He lamented that the PCRA court had “lost sight of its role as a neutral judicial officer” and had stayed Williams’s execution “for no valid reason.” Id., at ___, 105 A. 3d, at 1245. “[B]efore condemning officers of the court,” the chief justice stated, “the tribunal should be aware of the substantive status of Brady law,” which he believed the PCRA court had misapplied. Id., at ___, 105 A. 3d, at 1246. In addition, Chief Justice Castille denounced what he perceived as the “obstructionist anti-death penalty agenda” of Williams’s attorneys from the Federal Community Defender Office. Ibid. PCRA courts “throughout Pennsylvania need to be vigilant and circumspect when it comes to the activities of this particular advocacy group,” he wrote, lest Defender Office lawyers turn postconviction proceedings “into a circus where [they] are the ringmasters, with their parrots and puppets as a sideshow.” Id., at ___, 105 A. 3d, at 1247. Two weeks after the Pennsylvania Supreme Court decided Williams’s case, Chief Justice Castille retired from the bench. This Court granted Williams’s petition for certiorari. 576 U. S. ___ (2015). II A Williams contends that Chief Justice Castille’s decision as district attorney to seek a death sentence against him barred the chief justice from later adjudicating Williams’s petition to overturn that sentence. Chief Justice Castille, Williams argues, violated the Due Process Clause of the Fourteenth Amendment by acting as both accuser and judge in his case. The Court’s due process precedents do not set forth a specific test governing recusal when, as here, a judge had prior involvement in a case as a prosecutor. For the reasons explained below, however, the principles on which these precedents rest dictate the rule that must control in the circumstances here. The Court now holds that under the Due Process Clause there is an impermissible risk of actual bias when a judge earlier had significant, personal involvement as a prosecutor in a critical decision regarding the defendant’s case. Due process guarantees “an absence of actual bias” on the part of a judge. In re Murchison, 349 U. S. 133, 136 (1955) . Bias is easy to attribute to others and difficult to discern in oneself. To establish an enforceable and work-able framework, the Court’s precedents apply an objective standard that, in the usual case, avoids having to determine whether actual bias is present. The Court asks not whether a judge harbors an actual, subjective bias, but instead whether, as an objective matter, “the average judge in his position is ‘likely’ to be neutral, or whether there is an unconstitutional ‘potential for bias.’ ” Caperton, 556 U. S., at 881. Of particular relevance to the instant case, the Court has determined that an unconstitutional potential for bias exists when the same person serves as both accuser and adjudicator in a case. See Murchison, 349 U. S., at 136–137. This objective risk of bias is reflected in the due process maxim that “no man can be a judge in his own case and no man is permitted to try cases where he has an interest in the outcome.” Id., at 136. The due process guarantee that “no man can be a judge in his own case” would have little substance if it did not disqualify a former prosecutor from sitting in judgment of a prosecution in which he or she had made a critical decision. This conclusion follows from the Court’s analysis in In re Murchison. That case involved a “one-man judge-grand jury” proceeding, conducted pursuant to state law, in which the judge called witnesses to testify about suspected crimes. Id., at 134. During the course of the examinations, the judge became convinced that two witnesses were obstructing the proceeding. He charged one witness with perjury and then, a few weeks later, tried and convicted him in open court. The judge charged the other witness with contempt and, a few days later, tried and convicted him as well. This Court overturned the convictions on the ground that the judge’s dual position as accuser and decisionmaker in the contempt trials violated due process: “Having been a part of [the accusatory] process a judge cannot be, in the very nature of things, wholly disinterested in the conviction or acquittal of those accused.” Id., at 137. No attorney is more integral to the accusatory process than a prosecutor who participates in a major adversary decision. When a judge has served as an advocate for the State in the very case the court is now asked to adjudicate, a serious question arises as to whether the judge, even with the most diligent effort, could set aside any personal interest in the outcome. There is, furthermore, a risk that the judge “would be so psychologically wedded” to his or her previous position as a prosecutor that the judge “would consciously or unconsciously avoid the appearance of having erred or changed position.” Withrow, 421 U. S., at 57. In addition, the judge’s “own personal knowledge and impression” of the case, acquired through his or her role in the prosecution, may carry far more weight with the judge than the parties’ arguments to the court. Murchison, supra, at 138; see also Caperton, supra, at 881. Pennsylvania argues that Murchison does not lead to the rule that due process requires disqualification of a judge who, in an earlier role as a prosecutor, had significant involvement in making a critical decision in the case. The facts of Murchison, it should be acknowledged, differ in many respects from a case like this one. In Murchison, over the course of several weeks, a single official (the so-called judge-grand jury) conducted an investigation into suspected crimes; made the decision to charge witnesses for obstruction of that investigation; heard evidence on the charges he had lodged; issued judgments of conviction; and imposed sentence. See 349 U. S., at 135 (petitioners objected to “trial before the judge who was at the same time the complainant, indicter and prosecutor”). By contrast, a judge who had an earlier involvement in a prosecution might have been just one of several prosecutors working on the case at each stage of the proceedings; the prosecutor’s immediate role might have been limited to a particular aspect of the prosecution; and decades might have passed before the former prosecutor, now a judge, is called upon to adjudicate a claim in the case. These factual differences notwithstanding, the constitutional principles explained in Murchison are fully applicable where a judge had a direct, personal role in the defendant’s prosecution. The involvement of other actors and the passage of time are consequences of a complex criminal justice system, in which a single case may be litigated through multiple proceedings taking place over a period of years. This context only heightens the need for objective rules preventing the operation of bias that otherwise might be obscured. Within a large, impersonal system, an individual prosecutor might still have an influence that, while not so visible as the one-man grand jury in Murchison, is nevertheless significant. A prosecutor may bear responsibility for any number of critical decisions, including what charges to bring, whether to extend a plea bargain, and which witnesses to call. Even if decades intervene before the former prosecutor revisits the matter as a jurist, the case may implicate the effects and continuing force of his or her original decision. In these circumstances, there remains a serious risk that a judge would be influenced by an improper, if inadvertent, motive to validate and preserve the result obtained through the adversary process. The involvement of multiple actors and the passage of time do not relieve the former prosecutor of the duty to withdraw in order to ensure the neutral-ity of the judicial process in determining the consequences that his or her own earlier, critical decision may have set in motion. B This leads to the question whether Chief Justice Castille’s authorization to seek the death penalty against Williams amounts to significant, personal involvement in a critical trial decision. The Court now concludes that it was a significant, personal involvement; and, as a result, Chief Justice Castille’s failure to recuse from Williams’s case presented an unconstitutional risk of bias. As an initial matter, there can be no doubt that the decision to pursue the death penalty is a critical choice in the adversary process. Indeed, after a defendant is charged with a death-eligible crime, whether to ask a jury to end the defendant’s life is one of the most serious discretionary decisions a prosecutor can be called upon to make. Nor is there any doubt that Chief Justice Castille had a significant role in this decision. Without his express authorization, the Commonwealth would not have been able to pursue a death sentence against Williams. The importance of this decision and the profound consequences it carries make it evident that a responsible prosecutor would deem it to be a most significant exercise of his or her official discretion and professional judgment. Pennsylvania nonetheless contends that Chief Justice Castille in fact did not have significant involvement in the decision to seek a death sentence against Williams. The chief justice, the Commonwealth points out, was the head of a large district attorney’s office in a city that saw many capital murder trials. Tr. of Oral Arg. 36. According to Pennsylvania, his approval of the trial prosecutor’s request to pursue capital punishment in Williams’s case amounted to a brief administrative act limited to “the time it takes to read a one-and-a-half-page memo.” Ibid. In this Court’s view, that characterization cannot be credited. The Court will not assume that then-District Attorney Castille treated so major a decision as a perfunctory task requiring little time, judgment, or reflection on his part. Chief Justice Castille’s own comments while running for judicial office refute the Commonwealth’s claim that he played a mere ministerial role in capital sentencing decisions. During the chief justice’s election campaign, multiple news outlets reported his statement that he “sent 45 people to death rows” as district attorney. Seelye, Castille Keeps His Cool in Court Run, Philadelphia Inquirer, Apr. 30, 1993, p. B1; see also, e.g., Brennan, State Voters Must Choose Next Supreme Court Member, Legal Intelligencer, Oct. 28, 1993, pp. 1, 12. Chief Justice Castille’s willingness to take personal responsibility for the death sentences obtained during his tenure as district attorney indicate that, in his own view, he played a meaningful role in those sentencing decisions and considered his involvement to be an important duty of his office. Although not necessary to the disposition of this case, the PCRA court’s ruling underscores the risk of permitting a former prosecutor to be a judge in what had been his or her own case. The PCRA court determined that the trial prosecutor—Chief Justice Castille’s former subordinate in the district attorney’s office—had engaged in multiple, intentional Brady violations during Williams’s prosecution. App. 131–145, 150–154. While there is no indication that Chief Justice Castille was aware of the alleged prosecutorial misconduct, it would be difficult for a judge in his position not to view the PCRA court’s findings as a criticism of his former office and, to some extent, of his own leadership and supervision as district attorney. The potential conflict of interest posed by the PCRA court’s findings illustrates the utility of statutes and professional codes of conduct that “provide more protection than due process requires.” Caperton, 556 U. S., at 890. It is important to note that due process “demarks only the outer boundaries of judicial disqualifications.” Aetna Life Ins. Co. v. Lavoie, 475 U. S. 813 , 828 (1986). Most questions of recusal are addressed by more stringent and detailed ethical rules, which in many jurisdictions already require disqualification under the circumstances of this case. See Brief for American Bar Association as Amicus Curiae 5, 11–14; see also ABA Model Code of Judicial Conduct Rules 2.11(A)(1), (A)(6)(b) (2011) (no judge may participate “in any proceeding in which the judge’s impartiality might reasonably be questioned,” including where the judge “served in governmental employment, and in such capacity participated personally and substantially as a lawyer or public official concerning the proceeding”); ABA Center for Professional Responsibility Policy Implementation Comm., Comparison of ABA Model Judicial Code and State Variations (Dec. 14, 2015), available at http://www.americanbar.org/content/dam/aba/administrative/professional_responsibility/2_11.authcheckdam.pdf (as last visited June 7, 2016) (28 States have adopted language similar to ABA Model Judicial Code Rule 2.11); 28 U. S. C. §455(b)(3) (recusal required where judge “has served in governmental employment and in such capacity participated as counsel, adviser or material witness concerning the proceeding”). At the time Williams filed his recusal motion with the Pennsylvania Supreme Court, for example, Pennsylvania’s Code of Judicial Conduct disqualified judges from any proceeding in which “they served as a lawyer in the matter in controversy, or a lawyer with whom they previously practiced law served during such association as a lawyer concerning the matter. . . .” Pa. Code of Judicial Conduct, Canon 3C (1974, as amended). The fact that most jurisdictions have these rules in place suggests that today’s decision will not occasion a significant change in recusal practice. Chief Justice Castille’s significant, personal involvement in a critical decision in Williams’s case gave rise to an unacceptable risk of actual bias. This risk so endangered the appearance of neutrality that his participation in the case “must be forbidden if the guarantee of due process is to be adequately implemented.” Withrow, 421 U. S., at 47. III Having determined that Chief Justice Castille’s participation violated due process, the Court must resolve whether Williams is entitled to relief. In past cases, the Court has not had to decide the question whether a due process violation arising from a jurist’s failure to recuse amounts to harmless error if the jurist is on a multimember court and the jurist’s vote was not decisive. See Lavoie, supra, at 827–828 (addressing “the question whether a decision of a multimember tribunal must be vacated because of the participation of one member who had an interest in the outcome of the case,” where that member’s vote was outcome determinative). For the reasons discussed below, the Court holds that an unconstitutional failure to recuse constitutes structural error even if the judge in question did not cast a deciding vote. The Court has little trouble concluding that a due process violation arising from the participation of an inter-ested judge is a defect “not amenable” to harmless-error review, regardless of whether the judge’s vote was dispositive. Puckett v. United States, 556 U. S. 129, 141 (2009) (emphasis deleted). The deliberations of an appellate panel, as a general rule, are confidential. As a result, it is neither possible nor productive to inquire whether the jurist in question might have influenced the views of his or her colleagues during the decisionmaking process. Indeed, one purpose of judicial confidentiality is to assure jurists that they can reexamine old ideas and suggest new ones, while both seeking to persuade and being open to persuasion by their colleagues. As Justice Brennan wrote in his Lavoie concurrence, “The description of an opinion as being ‘for the court’ connotes more than merely that the opinion has been joined by a majority of the participating judges. It reflects the fact that these judges have exchanged ideas and arguments in deciding the case. It reflects the collective process of deliberation which shapes the court’s perceptions of which issues must be addressed and, more importantly, how they must be addressed. And, while the influence of any single participant in this process can never be measured with precision, experience teaches us that each member’s involvement plays a part in shaping the court’s ultimate disposition.” 475 U. S., at 831. These considerations illustrate, moreover, that it does not matter whether the disqualified judge’s vote was necessary to the disposition of the case. The fact that the interested judge’s vote was not dispositive may mean only that the judge was successful in persuading most members of the court to accept his or her position. That outcome does not lessen the unfairness to the affected party. See id., at 831–832 (Blackmun, J., concurring in judgment). A multimember court must not have its guarantee of neutrality undermined, for the appearance of bias demeans the reputation and integrity not just of one jurist, but of the larger institution of which he or she is a part. An insistence on the appearance of neutrality is not some artificial attempt to mask imperfection in the judicial process, but rather an essential means of ensuring the reality of a fair adjudication. Both the appearance and reality of impartial justice are necessary to the public legitimacy of judicial pronouncements and thus to the rule of law itself. When the objective risk of actual bias on the part of a judge rises to an unconstitutional level, the failure to recuse cannot be deemed harmless. The Commonwealth points out that ordering a rehearing before the Pennsylvania Supreme Court may not provide complete relief to Williams because judges who were exposed to a disqualified judge may still be influenced by their colleague’s views when they rehear the case. Brief for Respondent 51, 62. An inability to guarantee complete relief for a constitutional violation, however, does not justify withholding a remedy altogether. Allowing an appellate panel to reconsider a case without the participation of the interested member will permit judges to probe lines of analysis or engage in discussions they may have felt constrained to avoid in their first deliberations. Chief Justice Castille’s participation in Williams’s case was an error that affected the State Supreme Court’s whole adjudicatory framework below. Williams must be granted an opportunity to present his claims to a court unburdened by any “possible temptation . . . not to hold the balance nice, clear and true between the State and the accused.” Tumey v. Ohio, 273 U. S. 510, 532 (1927) . * * * Where a judge has had an earlier significant, personal involvement as a prosecutor in a critical decision in the defendant’s case, the risk of actual bias in the judicial proceeding rises to an unconstitutional level. Due process entitles Terrance Williams to “a proceeding in which he may present his case with assurance” that no member of the court is “predisposed to find against him.” Marshall v. Jerrico, Inc., 446 U. S. 238, 242 (1980) . The judgment of the Supreme Court of Pennsylvania is vacated, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.

578.US.2015_14-1504

Appellee voters from Virginia’s Congressional District 3 filed suit challenging the Commonwealth’s 2013 congressional redistricting plan on the ground that the legislature’s redrawing of their district was an unconstitutional racial gerrymander. Appellant Members of Congress from Virginia, including, as relevant here, Representatives Randy Forbes, Robert Wittman, and David Brat, intervened to help defend the Commonwealth’s plan. The District Court struck down the plan, and the intervenors appealed to this Court, which vacated the judgment below and remanded the case in light of one of the Court’s recent decisions. Again, the District Court held that the redistricting plan was unconstitutional, and again, the intervenors appealed. This time, the Court directed the parties to address whether appellants lack standing, since none reside in or represent Congressional District 3. Held: Appellants lack standing to pursue this appeal. Pp. 3–6. (a) A party invoking a federal court’s jurisdiction can establish Article III standing only by showing that he has suffered an “injury in fact,” that the injury is “fairly traceable” to the challenged conduct, and that the injury is likely to be “redressed” by a favorable decision. Lujan v. Defenders of Wildlife, 504 U.S. 555, 560–561. The need to satisfy these requirements persists throughout the life of the suit. Arizonans for Official English v. Arizona, 520 U.S. 43, 67. Pp. 3–4. (b) In light of the District Court’s decision striking down the redistricting plan, Representative Forbes, the Republican incumbent in District 4, decided to run in District 2. Originally, Representative Forbes argued that he would abandon his campaign in District 2 and run in District 4 if this Court ruled in his favor. Now, however, he has informed the Court that he will continue to seek election in District 2 regardless of this appeal’s outcome. Given this change, this Court does not see how any injury that Forbes might have suffered is “likely to be redressed by a favorable judicial decision.” Hollingsworth v. Perry, 570 U. S. ___, ___. Regardless of whether Forbes had standing at the time he first intervened, he does not have standing now. Representatives Wittman and Brat, the incumbents in Congressional Districts 1 and 7, respectively, have not identified any record evidence to support their allegation that the redistricting plan has harmed their prospects of reelection. The allegation of an injury, without more, is not sufficient to satisfy Article III. See Lujan, supra, at 561. Given the complete lack of evidence of any injury in this case, the Court need not decide when, or whether, evidence of the kind of injury the Representatives allege would prove sufficient for Article III purposes. Pp. 4–6. Appeal dismissed. Breyer, J., delivered the opinion for a unanimous Court.

Ten Members of Congress from Virginia, intervenors in the District Court below, have appealed a judgment from a three-judge panel striking down a congressional redistricting plan applicable to the November 2016 election. We conclude that the intervenors now lack standing to pursue the appeal. And we consequently order the appeal dismissed. I This lawsuit began in October 2013, after the then-Governor of Virginia signed into law a new congressional redistricting plan (which we shall call the “Enacted Plan”) designed to reflect the results of the 2010 census. Three voters from Congressional District 3 brought this lawsuit against the Commonwealth. They challenged the Enacted Plan on the ground that its redrawing of their district’s lines was an unconstitutional racial gerrymander. The Members of Congress now before us intervened to help defend the Enacted Plan. After a bench trial, a divided three-judge District Court agreed with the voters. It concluded that the Commonwealth had used race as the predominant basis for modifying the boundaries of District 3. Page v. Virginia State Bd. of Elections, 58 F. Supp. 3d 533, 550 (ED Va. 2014). And it found that the Commonwealth’s use of race, when scrutinized strictly, was not narrowly tailored to serve a compelling governmental interest. Id., at 553. The Commonwealth of Virginia did not appeal. Instead, the intervenor Members of Congress appealed the District Court’s judgment to this Court. See 28 U. S. C. §1253 (granting the right to directly appeal certain three-judge district court orders to the Supreme Court). Having just decided a racial gerrymandering case, Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___ (2015), we vacated the District Court’s judgment and remanded for reconsideration in light of that recent decision. Cantor v. Personhuballah, 575 U. S. ___ (2015). On remand the District Court again decided that District 3, as modified by the Enacted Plan, was an unconstitutional racial gerrymander. Page v. Virginia State Bd. of Elections, 2015 WL 3604029, *19 (ED Va., June 5, 2015). The court’s order set forth a deadline of September 1, 2015, for the Virginia Legislature to adopt a new redistricting plan. Again, the Commonwealth of Virginia decided not to appeal. And again, the intervenor Members of Congress appealed to this Court. On September 28, 2015, we asked the parties to file supplemental briefs addressing whether the intervenors had standing to appeal the District Court’s decision. 576 U. S. ___. As relevant here, the intervenors argued in their supplemental brief that they had standing because the District Court’s order, if allowed to stand, would necessarily result in a redrawing of their districts that would harm some of the intervenors’ reelection prospects. On November 13, 2015, we issued an order explaining that the Court was “postpon[ing]” “consideration of the question of jurisdiction” until “the hearing of the case on the merits.” In addition, our order instructed the parties to dedicate a portion of their briefs and their oral argument time to the issue of standing—specifically, “[w]hether [the intervenors] lack standing because none reside in or represent the only congressional district whose constitutionality is at issue in this case.” 577 U. S. ___. In the meantime, the Virginia Legislature failed to meet the September 1 deadline imposed by the District Court. The District Court thus appointed a Special Master to develop a new districting plan. The Special Master did so, and on January 7, 2016, the District Court approved that plan (which we shall call the “Remedial Plan”). The intervenor Members of Congress asked this Court to stay implementation of the Remedial Plan pending resolution of their direct appeal to this Court. We declined to do so. 577 U. S. ___ (2016). On March 21, we heard oral argument. That argument focused both on (1) the merits of intervenors’ claims denying any racial gerrymander and (2) the question of standing. In respect to standing, the Court focused on whether the District Court’s approval of the Remedial Plan on January 7 supported, or undermined, the intervenors’ standing argument that, in the absence of the original Enacted Plan, they would suffer harm. Tr. of Oral Arg. 9–23. II As our request for supplemental briefing, our order postponing consideration of jurisdiction, and our questions at oral argument suggested, we cannot decide the merits of this case unless the intervenor Members of Congress challenging the District Court’s racial-gerrymandering decision have standing. We conclude that the intervenors now lack standing. We must therefore dismiss the appeal for lack of jurisdiction. Article III of the Constitution grants the federal courts the power to decide legal questions only in the presence of an actual “Cas[e]” or “Controvers[y].” This restriction requires a party invoking a federal court’s jurisdiction to demonstrate standing. Arizonans for Official English v. Arizona, 520 U. S. 43, 64 (1997) . A party has standing only if he shows that he has suffered an “injury in fact,” that the injury is “fairly traceable” to the conduct being challenged, and that the injury will likely be “redressed” by a favorable decision. Lujan v. Defenders of Wildlife, 504 U. S. 555 –561 (1992) (internal quotation marks and ellipsis omitted). The need to satisfy these three requirements persists throughout the life of the lawsuit. Arizonans for Official English, 520 U. S., at 67. The relevant parties here are the intervenor Members of Congress. Since the Commonwealth of Virginia has not pursued an appeal, only the intervenors currently attack the District Court’s decision striking down the Enacted Plan. And an “intervenor cannot step into the shoes of the original party” (here, the Commonwealth) “unless the intervenor independently ‘fulfills the requirements of Article III.’ ” Id., at 65 (quoting Diamond v. Charles, 476 U. S. 54, 68 (1986) ). Although 10 current and former Members of Congress are technically intervenors, only 3 of the 10 now claim before this Court that they have standing. Those three Members are Representative Randy Forbes, Representative Robert Wittman, and Representative David Brat. Representative Forbes, the Republican incumbent in Congressional District 4, told us in his brief that, unless the Enacted Plan is upheld, District 4 will be “completely transform[ed] from a 48% Democratic district into a safe 60% Democratic district.” Brief for Appellants 58. According to Forbes, the threat of that kind of transformation compelled him to run in a different district, namely, Congressional District 2. At oral argument, Forbes’ counsel told the Court that, if the Enacted Plan were reinstated, Representative Forbes would abandon his election effort in Congressional District 2 and run in his old district, namely, Congressional District 4. Tr. of Oral Arg. 10. Soon after oral argument, however, the Court received a letter from counsel stating that Representative Forbes would “continue to seek election in District 2 regardless of whether the Enacted Plan is reinstated.” Letter from Counsel for Appellants to Scott S. Harris, Clerk of Court (Mar. 25, 2016), p. 2. Given this letter, we do not see how any injury that Forbes might have suffered “is likely to be redressed by a favorable judicial decision.” Hollingsworth v. Perry, 570 U. S. ___, ___–___ (2013) (slip op., at 5–6). Consequently, we need not decide whether, at the time he first intervened, Representative Forbes possessed standing. Regardless, he does not possess standing now. See Arizonans for Official English, supra, at 65; Lewis v. Continental Bank Corp., 494 U. S. 472 –478 (1990). Representative Wittman and Representative Brat are Republicans representing Congressional District 1 and Congressional District 7, respectively. In their opening brief they argue that they have standing to challenge the District Court’s order because, unless the Enacted Plan is reinstated, “a portion of the[ir] ‘base electorate’ ” will necessarily be replaced with “unfavorable Democratic voters,” thereby reducing the likelihood of the Representatives’ reelection. Brief for Appellants 58; see also Application for Stay of Remedial Plan Pending Resolution of Direct Appeal of Liability Judgment 25. Even assuming, without deciding, that this kind of injury is legally cognizable, Representatives Wittman and Brat have not identified record evidence establishing their alleged harm. We have made clear that the “party invoking federal jurisdiction bears the burden of establishing” that he has suffered an injury by submitting “affidavit[s] or other evidence.” Lujan, 504 U. S., at 561. When challenged by a court (or by an opposing party) concerned about standing, the party invoking the court’s jurisdiction cannot simply allege a nonobvious harm, without more. Ibid. Here, there is no “more.” Representatives Wittman and Brat claim that unless the Enacted Plan is reinstated, their districts will be flooded with Democratic voters and their chances of reelection will accordingly be reduced. But we have examined the briefs, looking for any evidence that an alternative to the Enacted Plan (including the Remedial Plan) will reduce the relevant intervenors’ chances of reelection, and have found none. The briefs focus on Congressional District 3 and Congressional District 4, districts with which Representatives Wittman and Brat are not associated. We need go no further. Given the lack of evidence that any of the three Representatives has standing, we need not decide when, or whether, evidence of the kind of injury they allege would prove sufficient for purposes of Article III’s requirements. In light of the letter we have received about Representative Forbes, and the absence of any evidence in the briefs supporting any harm to the other two Representatives, we conclude that none of the intervenors has standing to bring an appeal in this case. We consequently lack jurisdiction and therefore dismiss this appeal. It is so ordered.

581.US.2016_16-74

The Employee Retirement Income Security Act of 1974 (ERISA) generally obligates private employers offering pension plans to adhere to an array of rules designed to ensure plan solvency and protect plan participants. “[C]hurch plan[s],” however, are exempt from those regulations. 29 U. S. C. §1003(b)(2). From the beginning, ERISA has defined a “church plan” as “a plan established and maintained . . . for its employees . . . by a church.” §1002(33)(A). Congress then amended the statute to expand that definition, adding the provision whose effect is at issue here: “A plan established and maintained for its employees . . . by a church . . . includes a plan maintained by an organization . . . the principal purpose . . . of which is the administration or funding of [such] plan . . . for the employees of a church . . . , if such organization is controlled by or associated with a church.” §1002(33)(C)(i). (This opinion refers to the organizations described in that provision as “principal-purpose organizations.”) Petitioners, who identify themselves as three church-affiliated nonprofits that run hospitals and other healthcare facilities (collectively, hospitals), offer their employees defined-benefit pension plans. Those plans were established by the hospitals themselves, and are managed by internal employee-benefits committees. Respondents, current and former hospital employees, filed class actions alleging that the hospitals’ pension plans do not fall within ERISA’s church-plan exemption because they were not established by a church. The District Courts, agreeing with the employees, held that a plan must be established by a church to qualify as a church plan. The Courts of Appeals affirmed. Held: A plan maintained by a principal-purpose organization qualifies as a “church plan,” regardless of who established it. Pp. 5–15. (a) The term “church plan” initially “mean[t]” only “a plan established and maintained . . . by a church.” But subparagraph (C)(i) provides that the original definitional phrase will now “include” another—“a plan maintained by [a principal-purpose] organization.” That use of the word “include” is not literal, but tells readers that a different type of plan should receive the same treatment (i.e., an exemption) as the type described in the old definition. In other words, because Congress deemed the category of plans “established and maintained by a church” to “include” plans “maintained by” principal-purpose organizations, those plans—and all those plans—are exempt from ERISA’s requirements. Had Congress wanted, as the employees contend, to alter only the maintenance requirement, it could have provided in subparagraph (C)(i) that “a plan maintained by a church includes a plan maintained by” a principal-purpose organization—removing “established and” from the first part of the sentence. But Congress did not adopt that ready alternative. Instead, it added language whose most natural reading is to enable a plan “maintained” by a principal-purpose organization to substitute for a plan both “established” and “maintained” by a church. And as a corollary to that point, the employees’ construction runs aground on the so-called surplusage canon—the presumption that each word Congress uses is there for a reason. The employees read subparagraph (C)(i) as if it were missing the two words “established and.” This Court, however, “give[s] effect, if possible, to every clause and word of a statute.” Williams v. Taylor, 529 U. S. 362 . Pp. 5–12. (b) Both parties’ accounts of Congress’s purpose in enacting subparagraph (C)(i) tend to confirm this Court’s reading that plans maintained by principal-purpose organizations are eligible for the church-plan exemption, whatever their origins. According to the hospitals, Congress wanted to ensure that churches and church-affiliated organizations received comparable treatment under ERISA. If that is so, this Court’s construction of the text fits Congress’s objective to a T, as a church-establishment requirement would necessarily disfavor plans created by church affiliates. The employees, by contrast, claim that subparagraph (C)(i)’s main goal was to bring within the church-plan exemption plans managed by local pension boards—organizations often used by congregational denominations—so as to ensure parity between congregational and hierarchical churches. But that account cuts against, not in favor of, their position. Keeping the church-establishment requirement would have prevented some plans run by pension boards—the very entities the employees say Congress most wanted to benefit—from qualifying as “church plans” under ERISA. Pp. 12–14. No. 16–74, 817 F. 3d 517; No. 16–86, 810 F. 3d 175; and No. 16–258, 830 F. 3d 900, reversed. Kagan, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the cases. Sotomayor, J., filed a concurring opinion.Notes 1 Together with No. 16–86, Saint Peter’s Healthcare System et al. v. Kaplan, on certiorari to the United States Court of Appeals for the Third Circuit, and No. 16–258, Dignity Health et al. v. Rollins, on certiorari to the United States Court of Appeals for the Ninth Circuit.

The Employee Retirement Income Security Act of 1974 (ERISA) exempts “church plan[s]” from its otherwise-comprehensive regulation of employee benefit plans. 88Stat. 840, as amended, 29 U. S. C. §1003(b)(2). Under the statute, certain plans for the employees of churches or church-affiliated nonprofits count as “church plans” even though not actually administered by a church. See §1002(33)(C)(i). The question presented here is whether a church must have originally established such a plan for it to so qualify. ERISA, we hold, does not impose that requirement. I Petitioners identify themselves as three church-affiliated nonprofits that run hospitals and other healthcare facilities (collectively, hospitals).[1] They offer defined-benefit pension plans to their employees. Those plans were established by the hospitals themselves—not by a church—and are managed by internal employee-benefits committees. ERISA generally obligates private employers offering pension plans to adhere to an array of rules designed to ensure plan solvency and protect plan participants. See generally New York State Conference of Blue Cross & Blue Shield Plans v. Travelers Ins. Co., 514 U. S. 645, 651 (1995) (cataloguing ERISA’s “reporting and disclosure mandates,” “participation and vesting requirements,” and “funding standards”). But in enacting the statute, Congress made an important exception. “[C]hurch plan[s]” have never had to comply with ERISA’s requirements. §1003(b)(2). The statutory definition of “church plan” came in two distinct phases. From the beginning, ERISA provided that “[t]he term ‘church plan’ means a plan established and maintained . . . for its employees . . . by a church or by a convention or association of churches.” §1002(33)(A). Then, in 1980, Congress amended the statute to expand that definition by deeming additional plans to fall within it. The amendment specified that for purposes of the church-plan definition, an “employee of a church” would include an employee of a church-affiliated organization (like the hospitals here). §1002(33)(C)(ii)(II). And it added the provision whose effect is at issue in these cases: “A plan established and maintained for its employees . . . by a church or by a convention or association of churches includes a plan maintained by an organization . . . the principal purpose or function of which is the administration or funding of a plan or program for the provision of retirement benefits or welfare benefits, or both, for the employees of a church or a convention or association of churches, if such organi-zation is controlled by or associated with a church or a convention or association of churches.” §1002(33)(C)(i). That is a mouthful, for lawyers and non-lawyers alike; to digest it more easily, note that everything after the word “organization” in the third line is just a (long-winded) description of a particular kind of church-associated en-tity—which this opinion will call a “principal-purpose organization.” The main job of such an entity, as the statute explains, is to fund or manage a benefit plan for the employees of churches or (per the 1980 amendment’s other part) of church affiliates. The three federal agencies responsible for administering ERISA have long read those provisions, when taken together, to exempt plans like the hospitals’ from the statute’s mandates. (The relevant agencies are the Internal Revenue Service, Department of Labor, and Pension Benefit Guaranty Corporation.) The original definitional provision—§1002(33)(A), or paragraph (A) for short—defines a “church plan” as one “established and maintained . . . by a church”—not by a church-affiliated nonprofit. But according to the agencies, the later (block-quoted) provision—§1002(33)(C)(i), or just subparagraph (C)(i)—expands that definition to include any plan maintained by a principal-purpose organization, regardless of whether a church initially established the plan. And, the agencies believe, the internal benefits committee of a church-affiliated nonprofit counts as such an organization. See, e.g., IRS General Counsel Memorandum No. 39007 (Nov. 2, 1982), App. 636–637. That interpretation has appeared in hundreds of private letter rulings and opinion letters issued since 1982, including several provided to the hospitals here. See App. 57–69, 379–386, 668–715. The three cases before us are part of a recent wave of litigation challenging the agencies’ view. Respondents, current and former employees of the hospitals, filed class actions alleging that their employers’ pension plans do not fall within ERISA’s church-plan exemption (and thus must satisfy the statute’s requirements). That is so, the employees claim, because those plans were not established by a church—and ERISA, even as amended, demands that all “church plans” have such an origin. According to the employees, the addition of subparagraph (C)(i) allowed principal-purpose organizations to maintain such plans in lieu of churches; but that provision kept as-is paragraph (A)’s insistence that churches themselves establish “church plans.” See id., at 265–268, 435–437, 783–785. The District Courts handling the cases agreed with the employees’ position, and therefore held that the hospitals’ plans must comply with ERISA.[2] The Courts of Appeals for the Third, Seventh, and Ninth Circuits affirmed those decisions. The Third Circuit ruled first, concluding that ERISA’s “plain text” requires that a pension plan be established by a church to qualify for the church-plan exemption. Kaplan v. Saint Peter’s Healthcare System, 810 F. 3d 175, 177 (2015). In the court’s view, paragraph (A) set out “two requirements” for the exemption—“establishment and maintenance”—and “only the latter is expanded by the use of ‘includes’ ” in subparagraph (C)(i). Id., at 181. The Seventh and Ninth Circuits relied on similar reasoning to decide in the employees’ favor. See Stapleton v. Advocate Health Care Network, 817 F. 3d 517, 523 (CA7 2016); Rollins v. Dignity Health, 830 F. 3d 900, 906 (CA9 2016). In light of the importance of the issue, this Court granted certiorari. 579 U. S. ___ (2016). II The dispute in these cases about what counts as a “church plan” hinges on the combined meaning of paragraph (A) and subparagraph (C)(i). Interpretive purists may refer back as needed to the provisions as quoted above. See supra, at 3. But for those who prefer their statutes in (comparatively) user-friendly form, those provisions go as follows: Under paragraph (A), a “ ‘church plan’ means a plan established and maintained . . . by a church.” Under subparagraph (C)(i), “[a] plan established and maintained . . . by a church . . . includes a plan maintained by [a principal-purpose] organization.”[3] The parties agree that under those provisions, a “church plan” need not be maintained by a church; it may instead be maintained by a principal-purpose organization. But the parties differ as to whether a plan maintained by that kind of organization must still have been established by a church to qualify for the church-plan exemption. The hospitals say no: The effect of subparagraph (C)(i) was to bring within the church-plan definition all pension plans maintained by a principal-purpose organization, regardless of who first established them. The employees say yes: Subparagraph (C)(i) altered only the requirement that a pension plan be maintained by a church, while leaving intact the church-establishment condition. We conclude that the hospitals have the better of the argument. Start, as we always do, with the statutory language—here, a new definitional phrase piggy-backing on the one already existing. The term “church plan,” as just stated, initially “mean[t]” only “a plan established and maintained . . . by a church.” But subparagraph (C)(i) provides that the original definitional phrase will now “include” another—“a plan maintained by [a principal-purpose] organization.” That use of the word “include” is not literal—any more than when Congress says something like “a State ‘includes’ Puerto Rico and the District of Columbia.” See, e.g., 29 U. S. C. §1002(10).[4] Rather, it tells readers that a different type of plan should receive the same treatment (i.e., an exemption) as the type described in the old definition. And those newly favored plans, once again, are simply those “maintained by a principal-purpose organization”—irrespective of their origins. In effect, Congress provided that the new phrase can stand in for the old one as follows: “The term ‘church plan’ means a plan established and maintained by a church [a plan maintained by a principal-purpose organization].” The church-establishment condition thus drops out of the picture. Consider the same point in the form of a simple logic problem, with paragraph (A) and subparagraph (C)(i) as its first two steps: Premise 1: A plan established and maintained by a church is an exempt church plan. Premise 2: A plan established and maintained by a church includes a plan maintained by a principal-purpose organization. Deduction: A plan maintained by a principal-purpose organization is an exempt church plan. Or, as one court put the point without any of the ERISA terminology: “[I]f A is exempt, and A includes C, then C is exempt.” Overall v. Ascension, 23 F. Supp. 3d 816, 828 (ED Mich. 2014). Just so. Because Congress deemed the category of plans “established and maintained by a church” to “include” plans “maintained by” principal-purpose organizations, those plans—and all those plans—are exempt from ERISA’s requirements. Had Congress wanted, as the employees contend, to alter only the maintenance requirement, it had an easy way to do so—differing by only two words from the language it chose, but with an altogether different meaning. Suppose Congress had provided that “a plan maintained by a church includes a plan maintained by” a principal-purpose organization, leaving out the words “established and” from the first part of the sentence. That amendment would have accomplished exactly what the employees argue Congress intended: The language, that is, would have enabled a principal-purpose organization to take on the maintenance of a “church plan,” but left untouched the requirement that a church establish the plan in the first place. But Congress did not adopt that ready alternative. Instead, it added language whose most natural reading is to enable a plan “maintained” by a principal-purpose organization to substitute for a plan both “established” and “maintained” by a church. That drafting decision indicates that Congress did not in fact want what the employees claim. See, e.g., Lozano v. Montoya Alvarez, 572 U. S. 1 , ___–___ (2014) (slip op., at 13–14) (When legislators did not adopt “obvious alternative” language, “the natural implication is that they did not intend” the alternative). A corollary to this point is that the employees’ construction runs aground on the so-called surplusage canon—the presumption that each word Congress uses is there for a reason. See generally A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 174–179 (2012). As just explained, the employees urge us to read subparagraph (C)(i) as if it were missing the two words “established and.” The employees themselves do not contest that point: They offer no account of what function that language would serve on their proposed interpretation. See Brief for Respondents 34–35. In essence, the employees ask us to treat those words as stray marks on a page—notations that Congress regrettably made but did not really intend. Our practice, however, is to “give effect, if possible, to every clause and word of a statute.” Williams v. Taylor, 529 U. S. 362, 404 (2000) (internal quotation marks omitted). And here, that means construing the words “established and” in subparagraph (C)(i) as removing, for plans run by principal-purpose organizations, paragraph (A)’s church-establishment condition. The employees’ primary argument to the contrary takes the form of a supposed interpretive principle: “[I]f a definition or rule has two criteria, and a further provision expressly modifies only one of them, that provision is understood to affect only the criterion it expands or modifies.” Brief for Respondents 22. Applied here, the employees explain, that principle requires us to read subparagraph (C)(i) as “modify[ing] only the criterion” in paragraph (A) that “it expressly expands (‘maintained’), while leaving the other criterion (‘established’) unchanged.” Id., at 14. The employees cite no precedent or other authority to back up their proposed rule of construction, but they offer a thought-provoking hypothetical to demonstrate its good sense. Id., at 22. Imagine, they say, that a statute provides free insurance to a “person who is disabled and a veteran,” and an amendment then states that “a person who is disabled and a veteran includes a person who served in the National Guard.” Ibid. (quoting 810 F. 3d, at 181). Would a non-disabled member of the National Guard be entitled to the insurance benefit? Surely not, the employees answer: All of us would understand the “includes” provision to expand (or clarify) only the meaning of “veteran”—leaving unchanged the requirement of a disability. And the same goes here, the employees claim. But one good example does not a general rule make. Consider a variant of the employees’ hypothetical: A statute offers free insurance to a “person who enlisted and served in the active Armed Forces,” with a later amendment providing that “a person who enlisted and served in the active Armed Forces includes a person who served in the National Guard.” Would a person who served in the National Guard be ineligible for benefits unless she had also enlisted in the active Armed Forces—say, the regular Army or Navy? Of course not.[5] Two hypotheticals with similar grammatical constructions, two different results. In the employees’ example, the mind rebels against reading the statute literally, in line with the logical and canonical principles described above. In the variant, by contrast, the statute’s literal meaning and its most natural meaning cohere: Satisfaction of the amendment’s single eligibility criterion—service in the National Guard—is indeed enough. What might account for that divergence? And what does such an explanation suggest for ERISA? Two features of the employees’ hypothetical, when taken in combination, make it effective. First, the criteria there—veteran-status and disability—are relatively distinct from one another. (Compare enlistment and service, which address similar matters and tend to travel in tandem, the one preceding the other.) The more independent the specified variables, the more likely that they were designed to have standalone relevance. Second and yet more crucial, the employees’ example trades on our background understanding that a given interpretation is simply implausible—that it could not possibly have been what Congress wanted. Congress, we feel sure, would not have intended all National Guardsmen to get a benefit that is otherwise reserved for disabled veterans. (Compare that to our sense of whether Congress would have meant to hinge benefits to Guardsmen on their enlistment in a different service.) That sense of inconceivability does most of the work in the employees’ example, urging readers to discard usual rules of interpreting text because they will lead to a “must be wrong” outcome. But subparagraph (C)(i) possesses neither of those characteristics. For starters, the criteria at issue—establishment and maintenance—are not unrelated. The former serves as a necessary precondition of the latter, and both describe an aspect of an entity’s involvement with a benefit plan. Indeed, for various purposes, ERISA treats the terms “establish” and “maintain” interchange-ably. See, e.g., §1002(16)(B) (defining the “sponsor” of a plan as the organization that “establishe[s] or maintain[s]” the plan). So an amendment altering the one requirement could naturally alter the other too. What’s more, nothing we know about the way ERISA is designed to operate makes that an utterly untenable result. Whereas the disability condition is central to the statutory scheme in the employees’ hypothetical, the church-establishment condition, taken on its own, has limited functional significance. Establishment of a plan, after all, is a one-time, historical event; it is the entity maintaining the plan that has the primary ongoing responsibility (and potential liability) to plan participants. See Brief for United States as Amicus Curiae 31; Rose v. Long Island R. R. Pension Plan, 828 F. 2d 910, 920 (CA2 1987), cert. denied, 485 U. S. 936 (1988) (“[T]he status of the entity which currently maintains a particular pension plan bears more relation to Congress’ goals in enacting ERISA and its various exemptions[ ] than does the status of the entity which established the plan”). So removing the establishment condition for plans run by principal-purpose organizations has none of the contextual implausibility—the “Congress could not possibly have meant that” quality—on which the employees’ example principally rides. To the contrary, everything we can tell from extra-statutory sources about Congress’s purpose in enacting subparagraph (C)(i) supports our reading of its text. We say “everything we can tell” because in fact we cannot tell all that much. The legislative materials in these cases consist almost wholly of excerpts from committee hearings and scattered floor statements by individual lawmakers—the sort of stuff we have called “among the least illuminating forms of legislative history.” NLRB v. SW General, Inc., 580 U. S. ___, ___ (2017) (slip op., at 16). And even those lowly sources speak at best indirectly to the precise question here: None, that is, comments in so many words on whether subparagraph (C)(i) altered paragraph (A)’s church-establishment condition. Still, both the hospitals and the employees have constructed narratives from those bits and pieces about Congress’s goals in amending paragraph (A). And our review of their accounts—the employees’ nearly as much as the hospitals’—tends to confirm our conviction that plans maintained by principal-purpose organizations are eligible for ERISA’s “church plan” exemption, whatever their origins. According to the hospitals, Congress wanted to eliminate any distinction between churches and church-affiliated organizations under ERISA. See Brief for Petition-ers 18, 33–35. The impetus behind the 1980 amend-ment, they claim, was an IRS decision holding that pension plans established by orders of Catholic Sisters (to benefit their hospitals’ employees) did not qualify as “church plans” because the orders were not “carrying out [the Church’s] religious functions.” IRS General Counsel Memorandum No. 37266, 1977 WL 46200, *5 (Sept. 22, 1977). Many religious groups protested that ruling, criticizing the IRS for “attempting to define what is and what is not [a] ‘church’ and how the mission of the church is to be carried out.” 125 Cong. Rec. 10054 (1979) (letter to Sen. Talmadge from the Lutheran Church–Missouri Synod); see id., at 10054–10058 (similar letters). And that anger, the hospitals maintain, was what prompted ERISA’s amendment: Congress, they say, designed the new provision to ensure that, however categorized, all groups associated with church activities would receive comparable treatment. See Brief for Petitioners 35. If that is so, our construction of the text fits Congress’s objective to a T. A church-establishment requirement necessarily puts the IRS in the business of deciding just what a church is and is not—for example (as in the IRS’s ruling about the Sisters), whether a particular Catholic religious order should count as one. And that requirement, by definition, disfavors plans created by church affiliates, as compared to those established by (whatever the IRS has decided are) churches. It thus makes key to the “church plan” exemption the very line that, on the hospitals’ account, Congress intended to erase. The employees tell a different story about the origins of subparagraph (C)(i)—focusing on the pension boards that congregational denominations often used. See Brief for Respondents 14, 38–42; see also Brief for United States as Amicus Curiae 19–22. In line with their non-hierarchical nature, those denominations typically relied on separately incorporated local boards—rather than entities integrated into a national church structure—to administer benefits for their ministers and lay workers. According to the employees, subparagraph (C)(i)’s main goal was to bring those local pension boards within the church-plan exemption, so as to ensure that congregational and hierarchical churches would receive the same treatment. In support of their view, the employees cite several floor statements in which the amendment’s sponsors addressed that objective. See Brief for Respondents 38. Senator Talmadge, for example, stated that under the amendment, a “plan or program funded or administered through a pension board . . . will be considered a church plan.” 124 Cong. Rec. 16523; see also 124 Cong. Rec. 12107 (remarks of Rep. Conable). But that account of subparagraph (C)(i)’s primary purpose cuts against, not in favor of, the employees’ position. See Brief for United States as Amicus Curiae 21 (accepting the employees’ narrative, but arguing that it buttresses the opposite conclusion). That is because, as hearing testimony disclosed, plans run by church-affiliated pension boards came in different varieties: Some were created by church congregations, but others were established by the boards themselves. See, e.g., Hearings on S. 1090 et al. before the Subcommittee on Private Pension Plans and Employee Fringe Benefits of the Senate Committee on Finance, 96th Cong., 1st Sess., 400–401, 415–417 (1979). And still others were sufficiently old that their provenance could have become the subject of dispute. See id., at 411; 125 Cong. Rec. 10052 (remarks of Sen. Talmadge) (“The average age of a church plan is at least 40 years”). So keeping the church-establishment requirement would have prevented some plans run by pension boards—the very entities the employees say Congress most wanted to benefit—from qualifying as “church plans” under ERISA. No argument the employees have offered here supports that goal-defying (much less that text-defying) statutory construction. III ERISA provides (1) that a “church plan” means a “plan established and maintained . . . by a church” and (2) that a “plan established and maintained . . . by a church” is to “include[ ] a plan maintained by” a principal-purpose organization. Under the best reading of the statute, a plan maintained by a principal-purpose organization therefore qualifies as a “church plan,” regardless of who established it. We accordingly reverse the judgments of the Courts of Appeals. It is so ordered. Justice Gorsuch took no part in the consideration or decision of these cases.Notes 1 The parties disputed the hospitals’ church ties in the courts below, see n. 2, infra, but we assume for purposes of this decision that the facts are as the hospitals describe them. On those facts: Advocate Health Care Network operates 12 hospitals and some 250 other healthcare facilities in Illinois, and is associated with the Evangelical Lutheran Church in America and the United Church of Christ. Saint Peter’s Healthcare System runs a teaching hospital and several other medical facilities in New Jersey, and is both owned and controlled by a Roman Catholic diocese there. And Dignity Health runs an extensive network of community hospitals throughout the country, and maintains ties to the Catholic religious orders that initially sponsored some of its facilities. 2 The employees alternatively argued in the District Courts that the hospitals’ pension plans are not “church plans” because the hospitals do not have the needed association with a church and because, even if they do, their internal benefits committees do not count as principal-purpose organizations. See App. 267–269, 437–438, 785–786. Those issues are not before us, and nothing we say in this opinion expresses a view of how they should be resolved. 3 Again, we use the term “principal-purpose organization” as shorthand for the entity described in subparagraph (C)(i): a church-associated organization whose chief purpose or function is to fund or administer a benefits plan for the employees of either a church or a church-affiliated nonprofit. See supra, at 3. And again, the scope of that term—and whether it comprehends the hospitals’ internal benefits committees—is not at issue here. See n. 2, supra. 4 Or any more than when Congress, in the same 1980 amendment to ERISA, provided that an “employee of a church” was to “include[ ]” an employee of a church-affiliated organization. §1002(33)(C)(ii); see supra, at 3. 5 You might ask yourself, on reading this hypothetical statute, why Congress would not have made the removal of both original conditions clearer still by stating that the original provision “includes a person who enlisted and served in the National Guard.” We won’t go down the rabbit hole of further expounding on a fictional statute, but we can answer a parallel question for subparagraph (C)(i). Suppose Congress had stated that “[a] plan established and maintained . . . by a church . . . includes a plan established and maintained by [a principal-purpose] organization.” That language would have left out of the “church plan” definition pension plans originally established by churches, but subsequently maintained by principal-purpose organizations. And everyone agrees—the employees no less than the hospitals—that Congress wanted to treat those plans as “church plans.” (The dispute is only as to plans that principal-purpose organizations both establish and maintain.) See supra, at 6; Brief for Petitioners 25–26; Brief for Respondents 14, 35; Brief for United States as Amicus Curiae 24. So Congress could not have taken such a drafting tack to eliminate the necessity of church establishment.

580.US.2016_15-8544

Petitioner Beckles was convicted of possession of a firearm by a convicted felon, 18 U. S. C. §922(g)(1). His presentence investigation report concluded that he was eligible for a sentencing enhancement as a “career offender” under United States Sentencing Guideline §4B1.1(a) because his offense qualified as a “crime of violence” under §4B1.2(a)’s residual clause. The District Court sentenced petitioner as a career offender, and the Eleventh Circuit affirmed. Petitioner then filed a postconviction motion to vacate his sentence, arguing that his offense was not a “crime of violence.” The District Court denied the motion, and the Eleventh Circuit affirmed. Petitioner next filed a petition for a writ of certiorari from this Court. While his petition was pending, this Court held that the identically worded residual clause in the Armed Career Criminal Act of 1984 (ACCA), §924(e)(2)(b), was unconstitutionally vague, Johnson v. United States, 576 U. S. ___. The Court vacated and remanded petitioner’s case in light of Johnson. On remand, the Eleventh Circuit affirmed again, distinguishing the ACCA’s unconstitutionally vague residual clause from the residual clause in the Sentencing Guidelines. Held: The Federal Sentencing Guidelines, including §4B1.2(a)’s residual clause, are not subject to vagueness challenges under the Due Process Clause. Pp. 4–13. (a) The Due Process Clause prohibits the Government from “taking away someone’s life, liberty, or property under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement.” Johnson, supra, at ___–___. Under the void-for-vagueness doctrine, laws that fix the permissible sentences for criminal offenses must specify the range of available sentences with “sufficient clarity.” United States v. Batchelder, 442 U. S. 114 . In Johnson, this Court held that the ACCA’s residual clause fixed—in an impermissibly vague way—a higher range of sentences for certain defendants. But the advisory Guidelines do not fix the permissible range of sentences. They merely guide the exercise of a court’s discretion in choosing an appropriate sentence within the statutory range. Pp. 4–10. (1) The limited scope of the void-for-vagueness doctrine in this context is rooted in the history of federal sentencing. Congress has long permitted district courts “wide discretion to decide whether the offender should be incarcerated and for how long.” Mistretta v. United States, 488 U. S. 361 . Yet this Court has “never doubted the authority of a judge to exercise broad discretion in imposing a sentence within a statutory range,” United States v. Booker, 543 U. S. 220 , nor suggested that a defendant can successfully challenge as vague a sentencing statute conferring discretion to select an appropriate sentence from within a statutory range, even when that discretion is unfettered, see Batchelder, supra, at 123, 126. Pp. 6–7. (2) The Sentencing Reform Act of 1984 departed from this regime by establishing several factors to guide district courts in exercising their sentencing discretion. It also created the United States Sentencing Commission and charged it with establishing the Federal Sentencing Guidelines. Because the Guidelines have been rendered “effectively advisory” by this Court, Booker, supra, at 245, they guide district courts in exercising their discretion, but do not constrain that discretion. Accordingly, they are not amenable to vagueness challenges: If a system of unfettered discretion is not unconstitutionally vague, then it is difficult to see how the present system of guided discretion could be. Neither do they implicate the twin concerns underlying vagueness doctrine—providing notice and preventing arbitrary enforcement. The applicable statutory range, which establishes the permissible bounds of the court’s sentencing discretion, provides all the notice that is required. Similarly, the Guidelines do not invite arbitrary enforcement within the meaning of this Court’s case law, because they do not permit the sentencing court to prohibit behavior or to prescribe the sentencing ranges available. Rather, they advise sentencing courts how to exercise their discretion within the bounds established by Congress. Pp. 7–10. (b) The holding in this case does not render the advisory Guidelines immune from constitutional scrutiny, see, e.g., Peugh v. United States, 569 U. S. ___, or render “sentencing procedure[s]” entirely “immune from scrutiny under the due process clause,” Williams v. New York, 337 U. S. 241 , n. 18. This Court holds only that the Sentencing Guidelines are not subject to a challenge under the void-for-vagueness doctrine. Pp. 10–11. (c) Nor does this holding cast doubt on the validity of the other factors that sentencing courts must consider in exercising their sentencing discretion. See §§3553(a)(1)–(3), (5)–(7). A contrary holding, however, would cast serious doubt on those other factors because many of them appear at least as unclear as §4B1.2(a)’s residual clause. This Court rejects the Government’s argument that the individualized sentencing required by those other factors is distinguishable from that required by the Guidelines. It is far from obvious that §4B1.2(a)’s residual clause implicates the twin concerns of vagueness more than the other factors do, and neither the Guidelines nor the other factors implicate those concerns more than the absence of any guidance at all, which the Government concedes is constitutional. Pp. 11–13. 616 Fed. Appx. 415, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, and Alito, JJ., joined. Kennedy, J., filed a concurring opinion. Ginsburg, J., and Sotomayor, J., filed opinions concurring in the judgment. Kagan, J., took no part in the consideration or decision of the case.

At the time of petitioner’s sentencing, the advisory Sentencing Guidelines included a residual clause defining a “crime of violence” as an offense that “involves conduct that presents a serious potential risk of physical injury to another.” United States Sentencing Commission, Guidelines Manual §4B1.2(a)(2) (Nov. 2006) (USSG). This Court held in Johnson v. United States, 576 U. S. ___ (2015), that the identically worded residual clause in the Armed Career Criminal Act of 1984 (ACCA), 18 U. S. C. §924(e)(2)(B), was unconstitutionally vague. Petitioner contends that the Guidelines’ residual clause is also void for vagueness. Because we hold that the advisory Guidelines are not subject to vagueness challenges under the Due Process Clause, we reject petitioner’s argument. I Petitioner Travis Beckles was convicted in 2007 of possession of a firearm by a convicted felon, §922(g)(1). According to the presentence investigation report, the firearm was a sawed-off shotgun, and petitioner was therefore eligible for a sentencing enhancement as a “career of-fender” under the Sentencing Guidelines. The 2006 version of the Guidelines, which were in effect when petitioner was sentenced,[1] provided that “[a] defendant is a career offender if “(1) the defendant was at least eighteen years old at the time the defendant committed the instant offense of conviction; (2) the instant offense of conviction is a felony that is either a crime of violence or a controlled substance offense; and (3) the defendant has at least two prior felony convictions of either a crime of violence or a controlled substance offense.” USSG §4B1.1(a). The Guidelines defined “crime of violence” as “any offense under federal or state law, punishable by imprisonment for a term exceeding one year that— “(1) has as an element the use, attempted use, or threatened use of physical force against the person of another, or “(2) is burglary of a dwelling, arson, or extortion, involves use of explosives, or otherwise involves conduct that presents a serious potential risk of physical injury to another.” §4B1.2(a) (emphasis added). The clause beginning with “or otherwise” in this definition is known as the residual clause. The commentary to the career-offender Guideline provided that possession of a sawed-off shotgun was a crime of violence. See §4B1.2, comment., n. 1 (“Unlawfully possessing a firearm described in 26 U. S. C. §5845(a) (e.g., a sawed-off shotgun . . . ) is a ‘crime of violence’ ”); §5845(a) (“The term ‘firearm’ means (1) a shotgun having a barrel or barrels of less than 18 inches in length”). The District Court agreed that petitioner qualified as a career offender under the Guidelines. Petitioner was over 18 years of age at the time of his offense, and his criminal history included multiple prior felony convictions for controlled substance offenses. Furthermore, in the District Court’s view, petitioner’s §922(g)(1) conviction qualified as a “crime of violence.” Because he qualified as a career offender, petitioner’s Guidelines range was 360 months to life imprisonment. The District Court sentenced petitioner to 360 months. The Court of Appeals affirmed petitioner’s conviction and sentence, and this Court denied certiorari. United States v. Beckles, 565 F. 3d 832, 846 (CA11), cert denied, 558 U. S. 906 (2009) . In September 2010, petitioner filed a motion to vacate his sentence under 28 U. S. C. §2255, arguing that his conviction for unlawful possession of a firearm was not a “crime of violence,” and therefore that he did not qualify as a career offender under the Guidelines. The District Court denied the motion, and the Court of Appeals affirmed. Petitioner then filed a second petition for certiorari in this Court. While his petition was pending, the Court decided Johnson, holding that “imposing an increased sentence under the residual clause of the [ACCA]”—which contained the same language as the Guidelines’ residual clause—“violate[d] the Constitution’s guarantee of due process” because the clause was unconstitutionally vague. 576 U. S., at ___ (slip op., at 15). We subsequently granted his petition, vacated the judgment of the Court of Ap-peals, and remanded for further consideration in light of Johnson. Beckles v. United States, 576 U. S. ___ (2015). On remand, petitioner argued that his enhanced sentence was based on §4B1.2(a)’s residual clause, which he contended was unconstitutionally vague under Johnson. The Court of Appeals again affirmed. It noted that petitioner “was sentenced as a career offender based not on the ACCA’s residual clause, but based on express language in the Sentencing Guidelines classifying [his] offense as a ‘crime of violence.’ ” 616 Fed. Appx. 415, 416 (2015) ( per curiam). “Johnson,” the Court of Appeals reasoned, “says and decided nothing about career-offender enhancements under the Sentencing Guidelines or about the Guidelines commentary underlying [petitioner]’s sta-tus as a career-offender.” Ibid. The Court of Appeals denied rehearing en banc. Petitioner filed another petition for certiorari in this Court, again contending that §4B1.2(a)’s residual clause is void for vagueness. To resolve a conflict among the Courts of Appeals on the question whether Johnson’s vagueness holding applies to the residual clause in §4B1.2(a) of the Guidelines,[2] we granted certiorari. 579 U. S. ___ (2016). Because the United States, as respondent, agrees with petitioner that the Guidelines are subject to vagueness challenges, the Court appointed Adam K. Mortara as amicus curiae to argue the contrary position. 579 U. S. ___ (2016). He has ably discharged his responsibilities. II This Court has held that the Due Process Clause prohibits the Government from “taking away someone’s life, liberty, or property under a criminal law so vague that it fails to give ordinary people fair notice of the conduct it punishes, or so standardless that it invites arbitrary enforcement.” Johnson, 576 U. S., at ___–___ (slip op., at 3–4) (citing Kolender v. Lawson, 461 U. S. 352 –358 (1983)). Applying this standard, the Court has invalidated two kinds of criminal laws as “void for vagueness”: laws that define criminal offenses and laws that fix the permissible sentences for criminal offenses. For the former, the Court has explained that “the void-for-vagueness doctrine requires that a penal statute define the criminal offense with sufficient definiteness that ordinary people can understand what conduct is prohibited and in a manner that does not encourage arbitrary and discriminatory enforcement.” Id., at 357. For the latter, the Court has explained that “statutes fixing sentences,” Johnson, supra, at ___ (slip op., at 4) (citing United States v. Batchelder, 442 U. S. 114, 123 (1979) ), must specify the range of available sentences with “sufficient clarity,” id., at 123; see also United States v. Evans, 333 U. S. 483 (1948) ; cf. Giaccio v. Pennsylvania, 382 U. S. 399 (1966) . In Johnson, we applied the vagueness rule to a statute fixing permissible sentences. The ACCA’s residual clause, where applicable, required sentencing courts to increase a defendant’s prison term from a statutory maximum of 10 years to a minimum of 15 years. That requirement thus fixed—in an impermissibly vague way—a higher range of sentences for certain defendants. See Alleyne v. United States, 570 U. S. ___, ___ (2013) (describing the legally prescribed range of available sentences as the penalty fixed to a crime). Unlike the ACCA, however, the advisory Guidelines do not fix the permissible range of sentences. To the contrary, they merely guide the exercise of a court’s discretion in choosing an appropriate sentence within the statutory range. Accordingly, the Guidelines are not subject to a vagueness challenge under the Due Process Clause. The residual clause in §4B1.2(a)(2) therefore is not void for vagueness. A The limited scope of the void-for-vagueness doctrine in this context is rooted in the history of federal sentencing. Instead of enacting specific sentences for particular fed-eral crimes, Congress historically permitted district courts “wide discretion to decide whether the offender should be incarcerated and for how long.” Mistretta v. United States, 488 U. S. 361, 363 (1989) . For most crimes, Congress set forth a range of sentences, and sentencing courts had “almost unfettered discretion” to select the actual length of a defendant’s sentence “within the customarily wide range” Congress had enacted. Id., at 364; see also, e.g., Apprendi v. New Jersey, 530 U. S. 466 –482 (2000); Williams v. New York, 337 U. S. 241 –248 (1949). That discretion allowed district courts to craft individualized sentences, taking into account the facts of the crime and the history of the defendant. As a result, “[s]erious disparities in sentences . . . were common.” Mistretta, supra, at 365. Yet in the long history of discretionary sentencing, this Court has “never doubted the authority of a judge to exercise broad discretion in imposing a sentence within a statutory range.” United States v. Booker, 543 U. S. 220, 233 (2005) ; see also, e.g., Apprendi, supra, at 481 (“[N]othing in this history suggests that it is impermissible for judges to exercise discretion . . . in imposing a judgment within the range prescribed by statute”); Giaccio, supra, at 405, n. 8 (“[W]e intend to cast no doubt whatever on the constitutionality of the settled practice of many States to leave to juries finding defendants guilty of a crime the power to fix punishment within legally prescribed limits”). More specifically, our cases have never suggested that a defendant can successfully challenge as vague a sentencing statute conferring discretion to select an appropriate sentence from within a statutory range, even when that discretion is unfettered. In fact, our reasoning in Batchelder suggests the opposite. This Court considered in that case the constitutionality of two overlapping criminal provisions that authorized different maximum penalties for the same conduct. 442 U. S., at 115–116. The Court held that the sentencing provisions were not void for vagueness because they specified the “penalties available” and defined the “punishment authorized” upon conviction for each crime. Id., at 123. “Although the statutes create[d] uncertainty as to which crime may be charged and therefore what penalties may be imposed, they d[id] so to no greater extent than would a single statute authorizing various alternative punishments.” Ibid. (emphasis added). By specifying “the range of penalties that prosecutors and judges may seek and impose,” Congress had “fulfilled its duty.” Id., at 126 (citing Evans, supra, at 483; emphasis added). Indeed, no party to this case suggests that a system of purely discretionary sentencing could be subject to a vagueness challenge. B The Sentencing Reform Act of 1984 departed from this regime by establishing several factors to guide district courts in exercising their traditional sentencing discretion. 18 U. S. C. §3553. Congress in the same Act created the United States Sentencing Commission and charged it with establishing guidelines to be used for sentencing. Mis-tretta, supra, at 367. The result of the Commission’s work is the Federal Sentencing Guidelines, which are one of the sentencing factors that the Act requires courts to consider. §3553(a)(4). The Guidelines were initially binding on district courts, Booker, 543 U. S., at 233, but this Court in Booker rendered them “effectively advisory,” id., at 245. Although the Guidelines remain “the starting point and the initial benchmark” for sentencing, a sentencing court may no longer rely exclusively on the Guidelines range; rather, the court “must make an individualized assessment based on the facts presented” and the other statutory factors. Gall v. United States, 552 U. S. 38, 49, 50 (2007) . The Guidelines thus continue to guide district courts in exercising their discretion by serving as “the framework for sentencing,” Peugh v. United States, 569 U. S. ___, ___ (2013) (slip op., at 11), but they “do not constrain th[at] discretion,” id., at ___ (Thomas, J., dissenting) (slip op., at 2). Because they merely guide the district courts’ discretion, the Guidelines are not amenable to a vagueness challenge. As discussed above, the system of purely discretionary sentencing that predated the Guidelines was constitutionally permissible. If a system of unfettered discretion is not unconstitutionally vague, then it is difficult to see how the present system of guided discretion could be. The advisory Guidelines also do not implicate the twin concerns underlying vagueness doctrine—providing notice and preventing arbitrary enforcement. As to notice, even perfectly clear Guidelines could not provide notice to a person who seeks to regulate his conduct so as to avoid particular penalties within the statutory range. See, e.g., Grayned v. City of Rockford, 408 U. S. 104, 108 (1972) . That is because even if a person behaves so as to avoid an enhanced sentence under the career-offender guideline, the sentencing court retains discretion to impose the enhanced sentence. See, e.g., Pepper v. United States, 562 U. S. 476, 501 (2011) (“[O]ur post-Booker decisions make clear that a district court may in appropriate cases impose a non-Guidelines sentence based on a disagreement with the Commission’s views”). As we held in Irizarry v. United States, 555 U. S. 708 (2008) , “[t]he due process concerns that . . . require notice in a world of mandatory Guidelines no longer” apply. Id., at 714; see id., at 713 (“Any expectation subject to due process protection . . . that a criminal defendant would receive a sentence within the presumptively applicable Guidelines range did not survive our decision in [Booker], which invalidated the mandatory features of the Guidelines”). All of the notice required is provided by the applicable statutory range, which establishes the permissible bounds of the court’s sentencing discretion. The advisory Guidelines also do not implicate the vagueness doctrine’s concern with arbitrary enforcement. Laws that “regulate persons or entities,” we have explained, must be sufficiently clear “that those enforcing the law do not act in an arbitrary or discriminatory way.” FCC v. Fox Television Stations, Inc., 567 U. S. 239, 253 (2012) ; see also Grayned, supra, at 108–109 (“A vague law impermissibly delegates basic policy matters” to judges “for resolution on an ad hoc and subjective basis”). An unconstitutionally vague law invites arbitrary enforcement in this sense if it “leaves judges and jurors free to decide, without any legally fixed standards, what is prohibited and what is not in each particular case,” Giaccio, 382 U. S., at 402–403, or permits them to prescribe the sentences or sentencing range available, cf. Alleyne, 570 U. S., at ___ (slip op., at 11) (“[T]he legally prescribed range is the penalty affixed to the crime”). The Guidelines, however, do not regulate the public by prohibiting any conduct or by “establishing minimum and maximum penalties for [any] crime.” Mistretta, 488 U. S., at 396 (Sentencing Guidelines “do not bind or regulate the primary conduct of the public”). Rather, the Guidelines advise sentencing courts how to exercise their discretion within the bounds established by Congress. In this case, for example, the District Court did not “enforce” the career-offender Guideline against petitioner. It enforced 18 U. S. C. §922(g)(1)’s prohibition on possession of a firearm by a felon—which prohibited petitioner’s conduct—and §924(e)(1)’s mandate of a sentence of 15 years to life imprisonment—which fixed the permissible range of petitioner’s sentence. The court relied on the career-offender Guideline merely for advice in exercising its discretion to choose a sentence within those statutory limits. Justice Sotomayor’s concurrence suggests that judges interpreting a vague sentencing Guideline might rely on “statistical analysis,” “gut instinct,” or the judge’s “own feelings” to decide whether a defendant’s conviction is a crime of violence. Post, at 6 (opinion concurring in judgment) (internal quotation marks omitted). A judge granted unfettered discretion could use those same approaches in determining a defendant’s sentence. Indeed, the concurrence notes that federal judges before the Guidelines considered their own “view[s] of proper sentencing policy,” among other considerations. Post, at 11. Yet we have never suggested that unfettered discretion can be void for vagueness. Accordingly, we hold that the advisory Sentencing Guidelines are not subject to a vagueness challenge under the Due Process Clause and that §4B1.2(a)’s residual clause is not void for vagueness. III Our holding today does not render the advisory Guidelines immune from constitutional scrutiny. This Court held in Peugh, for example, that a “retrospective increase in the Guidelines range applicable to a defendant” violates the Ex Post Facto Clause. 569 U. S., at ___ (slip op., at 13). But the void-for-vagueness and ex post facto inquiries are “analytically distinct.” See id., at ___ (slip op., at 19) (distinguishing an ex post facto inquiry from a Sixth Amendment inquiry). Our ex post facto cases “have focused on whether a change in law creates a ‘significant risk’ of a higher sentence.” Ibid. A retroactive change in the Guidelines creates such a risk because “sentencing decisions are anchored by the Guidelines,” which establish “the framework for sentencing.” Id., at ___, ___ (slip op., at 10, 11). In contrast, the void-for-vagueness doctrine requires a different inquiry. The question is whether a law regulating private conduct by fixing permissible sentences provides notice and avoids arbitrary enforcement by clearly specifying the range of penalties available. The Government’s rebuttal that both doctrines are concerned with “ ‘fundamental justice,’ ” Reply Brief for United States 7, ignores the contours of our precedents. The Court has also recognized “in the Eighth Amendment context” that a district court’s reliance on a vague sentencing factor in a capital case, even indirectly, “can taint the sentence.” Brief for United States 43 (citing Espinosa v. Florida, 505 U. S. 1079, 1082 (1992) ( per curiam); emphasis added). But our approach to vagueness under the Due Process Clause is not interchangeable with “the rationale of our cases construing and applying the Eighth Amendment.” Maynard v. Cartwright, 486 U. S. 356, 361 (1988) . Our decision in Espinosa is thus inapposite, as it did not involve advisory Sentencing Guidelines or the Due Process Clause. Finally, our holding today also does not render “sentencing procedure[s]” entirely “immune from scrutiny under the due process clause.” Williams, 337 U. S., at 252, n. 18; see, e.g., Townsend v. Burke, 334 U. S. 736, 741 (1948) (holding that due process is violated when a court relies on “extensively and materially false” evidence to impose a sentence on an uncounseled defendant). We hold only that the advisory Sentencing Guidelines, including §4B1.2(a)’s residual clause, are not subject to a challenge under the void-for-vagueness doctrine. IV In addition to directing sentencing courts to consider the Guidelines, see §3553(a)(4)(A), Congress has directed them to consider a number of other factors in exercising their sentencing discretion, see §§3553(a)(1)–(3), (5)–(7). The Government concedes that “American judges have long made th[e] sorts of judgments” called for by the §3553(a) factors “in indeterminate-sentencing schemes, and this Court has never understood such discretionary determinations to raise vagueness concerns.” Brief for United States 42. Because the §3553 factors—like the Guidelines—do not mandate any specific sentences, but rather guide the exercise of a district court’s discretion within the applicable statutory range, our holding today casts no doubt on their validity. Holding that the Guidelines are subject to vagueness challenges under the Due Process Clause, however, would cast serious doubt on their validity. Many of these other factors appear at least as unclear as §4B1.2(a)’s residual clause. For example, courts must assess “the need for the sentence imposed” to achieve certain goals—such as to “reflect the seriousness of the offense,” “promote respect for the law,” “provide just punishment for the offense,” “afford adequate deterrence to criminal conduct,” and “provide the defendant with needed educational or vocational training . . . in the most effective manner.” §3553(a)(2). If petitioner were correct that §4B1.2(a)’s residual clause were subject to a vagueness challenge, we would be hard pressed to find these factors sufficiently definite to provide adequate notice and prevent arbitrary enforcement. The Government tries to have it both ways, arguing that the individualized sentencing required by the other §3553(a) factors is different in kind from that required by the Guidelines. “An inscrutably vague advisory guideline,” it contends, “injects arbitrariness into the sentencing process that is not found in the exercise of unguided discretion in a traditional sentencing system.” Reply Brief for United States 10–11. But it is far from obvious that the residual clause implicates the twin concerns of vagueness any more than the statutory command that sentencing courts impose a sentence tailored, for example, “to promote respect for the law.” §3553(a)(2)(A). And neither the Guidelines nor the other §3553 factors implicate those concerns more than the absence of any guidance at all, which the Government concedes is constitutional. The Government also suggests that the Guidelines are not like the other §3553(a) factors “because they require a court to decide whether the facts of the case satisfy a legal standard in order to derive a specific numerical range.” Id., at 22. But that does not distinguish the other sentencing factors, which require courts to do the same thing. Section 3553(a) states that district courts “shall impose a sentence sufficient, but not greater than necessary, to comply with the purposes set forth in [§3553(a)(2)].” In fact, the Guidelines generally offer more concrete advice in imposing a particular sentence and make it easier to review whether a court has abused its substantial discretion. There is no sound reason to conclude that the Guidelines—but not §3553(a)’s other sentencing factors—are amenable to vagueness review. * * * Because the advisory Sentencing Guidelines are not subject to a due process vagueness challenge, §4B1.2(a)’s residual clause is not void for vagueness. The judgment of the Court of Appeals, accordingly, is affirmed. It is so ordered. Justice Kagan took no part in the consideration or decision of this case.Notes 1 With one exception not relevant here, 18 U. S. C. §3553(a)(4)(A) instructs sentencing courts to consider the Guidelines ranges that “are in effect on the date the defendant is sentenced.” Accordingly, references in this opinion to the Guidelines are to the 2006 version. 2 Compare United States v. Matchett, 802 F. 3d 1185, 1193–1196 (CA11 2015) (holding that the Guidelines are not subject to due process vagueness challenges), with, e.g., United States v. Townsend, 638 Fed. Appx. 172, 178, n. 14 (CA3 2015) (declining to follow Matchett); United States v. Pawlak, 822 F. 3d 902, 905–911 (CA6 2016) (holding that the Guidelines are subject to due process vagueness challenges); United States v. Hurlburt, 835 F. 3d 715, 721–725 (CA7 2016) (en banc) (same); United States v. Madrid, 805 F. 3d 1204, 1210–1211 (CA10 2015) (same).

580.US.2016_15-680

After the 2010 census, the Virginia State Legislature drew new lines for 12 state legislative districts, with a goal of ensuring that each district would have a black voting-age population (BVAP) of at least 55%. Certain voters filed suit, claiming that the new districts violated the Fourteenth Amendment’s Equal Protection Clause. State legislative officials (State) intervened to defend the plan. A three-judge District Court rejected the challenges. As to 11 of the districts, the court concluded that the voters had not shown, as this Court’s precedent requires, “that race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district,” Miller v. Johnson, 515 U. S. 900 . In so doing, the court held that race predominates only where there is an “ ‘actual conflict between traditional redistricting criteria and race.’ ” 141 F. Supp. 3d 505, 524. It thus confined the predominance analysis to the portions of the new lines that appeared to deviate from traditional criteria. As to the remaining district, District 75, the court found that race did predominate, but that the lines were constitutional because the legislature’s use of race was narrowly tailored to a compelling state interest. In particular, the court found the legislature had good reasons to believe that a 55% racial target was necessary in District 75 to avoid diminishing the ability of black voters to elect their preferred candidates, which at the time would have violated §5 of the Voting Rights Act of 1965, see Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___. Held: 1. The District Court employed an incorrect legal standard in determining that race did not predominate in 11 of the 12 districts. Pp. 6–13. (a) The Equal Protection Clause prohibits a State, without sufficient justification, from “separat[ing] its citizens into different voting districts on the basis of race.” Miller, 515 U. S., at 911. Courts must “exercise extraordinary caution in adjudicating claims” of racial gerrymandering, id., at 916, since a legislature is always “aware of race when it draws district lines, just as it is aware of . . . other demographic factors,” Shaw v. Reno, 509 U. S. 630 (Shaw I). A plaintiff alleging racial gerrymandering thus bears the burden “to show, either through circumstantial evidence of a district’s shape and demographics or more direct evidence going to legislative purpose, that race was the predominant factor motivating the legislature’s [districting] decision,” which requires proving “that the legislature subordinated traditional race-neutral districting principles . . . to racial considerations.” Miller, supra, at 916. Here, the District Court misapplied controlling law in two principal ways. Pp. 6–7. (b) First, the District Court misunderstood relevant precedents when it required the challengers to establish, as a prerequisite to showing racial predominance, an actual conflict between the enacted plan and traditional redistricting principles. This Court has made clear that parties may show predominance “either through circumstantial evidence of a district’s shape and demographics or more direct evidence going to legislative purpose,” Miller, supra, at 916, and that race may predominate even when a plan respects traditional principles, Shaw v. Hunt, 517 U. S. 899 (Shaw II). The State’s theory in this case is irreconcilable with Miller and Shaw II. The State insists, e.g., that the harm from racial gerrymandering lies not in racial line-drawing per se but in grouping voters of the same race together when they otherwise lack shared interests. But “the constitutional violation” in racial gerrymandering cases stems from the “racial purpose of state action, not its stark manifestation.” Miller, supra, at 913. The State also contends that race does not have a prohibited effect on a district’s lines if the legislature could have drawn the same lines in accordance with traditional criteria. The proper inquiry, however, concerns the actual considerations that provided the essential basis for the lines drawn, not post hoc justifications that the legislature could have used but did not. A legislature could construct a plethora of potential maps that look consistent with traditional, race-neutral principles, but if race is the overriding reason for choosing one map over others, race still may predominate. A conflict or inconsistency may be persuasive circumstantial evidence tending to show racial predomination, but no rule requires challengers to present this kind of evidence in every case. As a practical matter, this kind of evidence may be necessary in many or even most cases. But there may be cases where challengers can establish racial predominance without evidence of an actual conflict. Pp. 7–11. (c) The District Court also erred in considering the legislature’s racial motive only to the extent that the challengers identified deviations from traditional redistricting criteria attributable to race and not to some other factor. Racial gerrymandering claims proceed “district-by-district,” Alabama, supra, at ___, and courts should not divorce any portion of a district’s lines—whatever their relationship to traditional principles—from the rest of the district. Courts may consider evidence pertaining to an area that is larger or smaller than the district at issue. But the ultimate object of the inquiry is the legislature’s predominant motive for the district’s design as a whole, and any explanation for a particular portion of the lines must take account of the districtwide context. A holistic analysis is necessary to give the proper weight to districtwide evidence, such as stark splits in the racial composition of populations moved into and out of a district, or the use of a racial target. Pp. 11–12. (d) The District Court is best positioned to determine on remand the extent to which, under the proper standard, race directed the shape of these 11 districts, and if race did predominate, whether strict scrutiny is satisfied. Pp. 12–13. 2. The District Court’s judgment regarding District 75 is consistent with the basic narrow tailoring analysis explained in Alabama. Where a challenger succeeds in establishing racial predominance, the burden shifts to the State to “demonstrate that its districting legislation is narrowly tailored to achieve a compelling interest.” Miller, supra, at 920. Here, it is assumed that the State’s interest in complying with the Voting Rights Act was a compelling interest. When a State justifies the predominant use of race in redistricting on the basis of the need to comply with the Voting Rights Act, “the narrow tailoring requirement insists only that the legislature have a ‘strong basis in evidence’ in support of the (race-based) choice that it has made.” Alabama, 575 U. S., at ___–___. The State must show not that its action was actually necessary to avoid a statutory violation, but only that the legislature had “ ‘good reasons to believe’ ” its use of race was needed in order to satisfy the Voting Rights Act. Ibid. There was no error in the District Court’s conclusion that the legislature had sufficient grounds to determine that the race-based calculus it employed in District 75 was necessary to avoid violating §5. Under the facts found by that court, the legislature performed the kind of functional analysis of District 75 necessary under §5, and the result reflected the good-faith efforts of legislators to achieve an informed bipartisan consensus. In contesting the sufficiency of that evidence and the evidence justifying the 55% BVAP floor, the challengers ask too much from state officials charged with the sensitive duty of reapportioning legislative districts. As to the claim that the BVAP floor is akin to the “mechanically numerical view” of §5 rejected in Alabama, supra, at ___, the record here supports the State’s conclusion that this was an instance where a 55% BVAP was necessary for black voters to have a functional working majority. Pp. 13–16. 141 F. Supp. 3d 505, affirmed in part, vacated in part, and remanded. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in part and concurring in the judgment. Thomas, J., filed an opinion concurring in the judgment in part and dissenting in part.

This case addresses whether the Virginia state legislature’s consideration of race in drawing new lines for 12 state legislative districts violated the Equal Protection Clause of the Fourteenth Amendment. After the 2010 census, some redistricting was required to ensure proper numerical apportionment for the Virginia House of Delegates. It is undisputed that the boundary lines for the 12 districts at issue were drawn with a goal of ensuring that each district would have a black voting-age population (BVAP) of at least 55%.Certain voters challenged the new districts as unconstitutional racial gerrymanders. The United States District Court for the Eastern District of Virginia, constituted as a three-judge district court, rejected the challenges as to each of the 12 districts. As to 11 of the districts, the District Court concluded that the voters had not shown, as this Court’s precedent requires, “that race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller v. Johnson,515 U. S. 900 (1995). The District Court held that race predominates only where there is an “ ‘actual conflict between traditional redistricting criteria and race,’ ” 141 F. Supp. 3d 505, 524 (ED Va. 2015), so it confined the predominance analysis to the portions of the new lines that appeared to deviate from traditional criteria, and found no violation. As to the remaining district, District 75, the District Court found that race did predominate. It concluded, however, that the lines were constitutional because the legislature’s use of race was narrowly tailored to a compelling state interest. In particular, the District Court determined that the legislature had “good reasons to believe” that a 55% racial target was necessary in District 75 to avoid diminishing the ability of black voters to elect their preferred candidates, which at the time would have violated §5 of the Voting Rights Act of 1965. Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___ (2015) (slip op., at 22) (internal quotation marks omitted and emphasis deleted).On appeal to this Court, the challengers contend that the District Court employed an incorrect legal standard for racial predominance and that the legislature lacked good reasons for its use of race in District 75. This Court now affirms as to District 75 and vacates and remands as to the remaining 11 districts.IAfter the 2010 census, the Virginia General Assembly set out to redraw the legislative districts for the State Senate and House of Delegates in time for the 2011 elections. In February 2011, the House Committee on Privileges and Elections adopted a resolution establishing criteria to guide the redistricting process. Among those criteria were traditional redistricting factors such as compactness, contiguity of territory, and respect for communities of interest. But above those traditional objectives, the committee gave priority to two other goals. First, in accordance with the principle of one person, one vote, the committee resolved that “[t]he population of each district shall be as nearly equal to the population of every other district as practicable,” with any deviations falling “within plus-or-minus one percent.” 141 F. Supp. 3d, at 518. Second, the committee resolved that the new map must comply with the “protections against . . . unwarranted retrogression” contained in §5 of the Voting Rights Act. Ibid. At the time, §5 required covered jurisdictions, including Virginia, to preclear any change to a voting standard, practice, or procedure by showing federal authorities that the change would not have the purpose or effect of “diminishing the ability of [members of a minority group] to elect their preferred candidates of choice.” §5,120Stat.580–581,52 U. S. C. §10304(b). After the redistricting process here was completed, this Court held that the coverage formula in §4(b) of the Voting Rights Act no longer may be used to require preclearance under §5. See Shelby County v. Holder, 570 U. S. ___, ___ (2013) (slip op., at 24).The committee’s criteria presented potential problems for 12 House districts. Under §5 as Congress amended it in 2005, “[a] plan leads to impermissible retrogression when, compared to the plan currently in effect (typically called a ‘benchmark plan’), the new plan diminishes the number of districts in which minority groups can ‘elect their preferred candidates of choice’ (often called ‘ability-to-elect’ districts).” Harris v. Arizona Independent Redistricting Comm’n, 578 U. S. ___, ___–___ (2016) (slip op., at 5–6) (quoting52 U. S. C. §10304(b)). The parties agree that the 12 districts at issue here, where minorities had constituted a majority of the voting-age population for many past elections, qualified as “ability-to-elect” districts. Most of the districts were underpopulated, however, so any new plan required moving significant numbers of new voters into these districts in order to comply with the principle of one person, one vote. Under the benchmark plan, the districts had BVAPs ranging from 62.7% down to 46.3%. Three districts had BVAPs below 55%.Seeking to maintain minority voters’ ability to elect their preferred candidates in these districts while complying with the one-person, one-vote criterion, legislators concluded that each of the 12 districts “needed to contain a BVAP of at least 55%.” 141 F. Supp. 3d, at 519. At trial, the parties disputed whether the 55% figure “was an aspiration or a target or a rule.” Ibid. But they did not dispute “the most important question—whether [the 55%] figure was used in drawing the Challenged Districts.” Ibid. The parties agreed, and the District Court found, “that the 55% BVAP figure was used in structuring the districts.” Ibid. In the enacted plan all 12 districts contained a BVAP greater than 55%.Who first suggested the 55% BVAP criterion and how the legislators agreed upon it was less clear from the evidence. See id., at 521 (describing the “[t]estimony on this question” as “a muddle”). In the end, the District Court found that the 55% criterion emerged from discussions among certain members of the House Black Caucus and the leader of the redistricting effort in the House, Delegate Chris Jones, “based largely on concerns pertaining to the re-election of Delegate Tyler in [District] 75.” Id., at 522. The 55% figure “was then applied across the board to all twelve” districts. Ibid.In April 2011, the General Assembly passed Delegate Jones’ plan with broad support from both parties and members of the Black Caucus. One of only two dissenting members of the Black Caucus was Delegate Tyler of District 75, who objected solely on the ground that the 55.4% BVAP in her district was too low. In June 2011, the U. S. Department of Justice precleared the plan.Three years later, before this suit was filed, a separate District Court struck down Virginia’s third federal congressional district (not at issue here), based in part on the legislature’s use of a 55% BVAP threshold. See Page v. Virginia State Bd. of Elections, 58 F. Supp. 3d 533, 553 (ED Va. 2014), vacated and remanded sub nom. Cantor v. Personhuballah, 575 U. S. ___ (2015), judgt. entered sub nom. Page v. Virginia State Bd. of Elections, 2015 WL 3604029 (June 5, 2015), appeal dism’d sub nom. Wittman v. Personhuballah, 578 U. S. ___ (2016). After that decision, 12 voters registered in the 12 districts here at issue filed this action challenging the district lines under the Equal Protection Clause. Because the claims “challeng[ed] the constitutionality of . . . the apportionment of [a] statewide legislative body,” the case was heard by a three-judge District Court.28 U. S. C. §2284(a). The Virginia House of Delegates and its Speaker, William Howell (together referred to hereinafter as the State), intervened and assumed responsibility for defending the plan, both before the District Court and now before this Court.After a 4-day bench trial, a divided District Court ruled for the State. With respect to each challenged district, the court first assessed whether “racial considerations predominated over—or ‘subordinated’—traditional redistricting criteria.” 141 F. Supp. 3d, at 523. An essential premise of the majority opinion was that race does not predominate unless there is an “actual conflict between traditional redistricting criteria and race that leads to the subordination of the former.” Id., at 524. To implement that standard, moreover, the court limited its inquiry into racial motive to those portions of the district lines that appeared to deviate from traditional criteria. The court thus “examine[d] those aspects of the [district] that appear[ed] to constitute ‘deviations’ from neutral criteria” to ascertain whether the deviations were attributable to race or to other considerations, “such as protection of incumbents.” Id., at 533–534. Only if the court found a deviation attributable to race did it proceed to “determine whether racial considerations qualitatively subordinated all other non-racial districting criteria.” Ibid. Under that analysis, the court found that race did not predominate in 11 of the 12 districts.When it turned to District 75, the District Court found that race did predominate. The court reasoned that “[a]chieving a 55% BVAP floor required ‘drastic maneuvering’ that is reflected on the face of the district.” Id., at 557. Applying strict scrutiny, the court held that compliance with §5 was a compelling state interest and that the legislature’s consideration of race in District 75 was narrowly tailored. As to narrow tailoring, the court explained that the State had “a strong basis in evidence” to believe that its actions were “reasonably necessary” to avoid retrogression. Id., at 548. In particular, the court found that Delegate Jones had considered “precisely the kinds of evidence that legislators are encouraged to use” in achieving compliance with §5, including turnout rates, the district’s large disenfranchised prison population, and voting patterns in the contested 2005 primary and general elections. Id., at 558.Judge Keenan dissented as to all 12 districts. She concluded that the majority applied an incorrect understanding of racial predominance and that Delegate Jones’ analysis of District 75 was too “general and conclusory.” Id., at 578. This appeal followed, and probable jurisdiction was noted. 578 U. S. ___ (2016); see28 U. S. C. §1253.IIAgainst the factual and procedural background set out above, it is now appropriate to consider the controlling legal principles in this case. The Equal Protection Clause prohibits a State, without sufficient justification, from “separat[ing] its citizens into different voting districts on the basis of race.” Miller, 515 U. S., at 911. The harms that flow from racial sorting “include being personally subjected to a racial classification as well as being represented by a legislator who believes his primary obligation is to represent only the members of a particular racial group.” Alabama, 575 U. S., at ___ (slip op., at 6) (alterations, citation, and internal quotation marks omitted). At the same time, courts must “exercise extraordinary caution in adjudicating claims that a State has drawn district lines on the basis of race.” Miller, 515 U. S., at 916. “Electoral districting is a most difficult subject for legislatures,” requiring a delicate balancing of competing considerations. Id., at 915. And “redistricting differs from other kinds of state decisionmaking in that the legislature always is aware of race when it draws district lines, just as it is aware of . . . a variety of other demographic factors.” Shaw v. Reno,509 U. S. 630,646 (1993) (Shaw I).In light of these considerations, this Court has held that a plaintiff alleging racial gerrymandering bears the burden “to show, either through circumstantial evidence of a district’s shape and demographics or more direct evidence going to legislative purpose, that race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller, 515 U. S. at 916. To satisfy this burden, the plaintiff “must prove that the legislature subordinated traditional race-neutral districting principles . . . to racial considerations.” Ibid. The challengers contend that, in finding that race did not predominate in 11 of the 12 districts, the District Court misapplied controlling law in two principal ways. This Court considers them in turn.AThe challengers first argue that the District Court misunderstood the relevant precedents when it required the challengers to establish, as a prerequisite to showing racial predominance, an actual conflict between the enacted plan and traditional redistricting principles. The Court agrees with the challengers on this point.A threshold requirement that the enacted plan must conflict with traditional principles might have been reconcilable with this Court’s case law at an earlier time. In Shaw I, the Court recognized a claim of racial gerrymandering for the first time. See 509 U. S., at 652. Certain language in Shaw I can be read to support requiring a challenger who alleges racial gerrymandering to show an actual conflict with traditional principles. The opinion stated, for example, that strict scrutiny applies to “redistricting legislation that is so bizarre on its face that it is unexplainable on grounds other than race.” Id., at 644 (internal quotation marks omitted). The opinion also stated that “reapportionment is one area in which appearances do matter.” Id., at 647.The Court’s opinion in Miller, however, clarified the racial predominance inquiry. In particular, it rejected the argument that, “regardless of the legislature’s purposes, a plaintiff must demonstrate that a district’s shape is so bizarre that it is unexplainable other than on the basis of race.” 515 U. S., at 910–911. The Court held to the contrary in language central to the instant case: “Shape is relevant not because bizarreness is a necessary element of the constitutional wrong or a threshold requirement of proof, but because it may be persuasive circumstantial evidence that race for its own sake, and not other districting principles, was the legislature’s dominant and controlling rationale.” Id., at 913. Parties therefore “may rely on evidence other than bizarreness to establish race-based districting,” and may show predominance “either through circumstantial evidence of a district’s shape and demographics or more direct evidence going to legislative purpose.” Id., at 913, 916.The Court addressed racial gerrymandering and traditional redistricting factors again in Shaw v. Hunt,517 U. S. 899 (1996) (Shaw II). The Court there rejected the view of one of the dissents that “strict scrutiny does not apply where a State ‘respects’ or ‘complies with traditional districting principles.’ ” Id., at 906 (quoting id., at 931–932 (Stevens, J., dissenting); alteration omitted). Race may predominate even when a reapportionment plan respects traditional principles, the Court explained, if “[r]ace was the criterion that, in the State’s view, could not be compromised,” and race-neutral considerations “came into play only after the race-based decision had been made.” Id., at 907.The State’s theory in this case is irreconcilable with Miller and Shaw II. The State insists, for example, that the harm from racial gerrymandering lies not in racial line-drawing per se but in grouping voters of the same race together when they otherwise lack shared interests. But “the constitutional violation” in racial gerrymandering cases stems from the “racial purpose of state action, not its stark manifestation.” Miller, supra, at 913. The Equal Protection Clause does not prohibit misshapen districts. It prohibits unjustified racial classifications.The State contends further that race does not have a prohibited effect on a district’s lines if the legislature could have drawn the same lines in accordance with traditional criteria. That argument parallels the District Court’s reasoning that a reapportionment plan is not an express racial classification unless a racial purpose is apparent from the face of the plan based on the irregular nature of the lines themselves. See 141 F. Supp. 3d, at 524–526. This is incorrect. The racial predominance inquiry concerns the actual considerations that provided the essential basis for the lines drawn, not post hoc justifications the legislature in theory could have used but in reality did not.Traditional redistricting principles, moreover, are numerous and malleable. The District Court here identified no fewer than 11 race-neutral redistricting factors a legislature could consider, some of which are “surprisingly ethereal” and “admi[t] of degrees.” Id., at 535, 537. By deploying those factors in various combinations and permutations, a State could construct a plethora of potential maps that look consistent with traditional, race-neutral principles. But if race for its own sake is the overriding reason for choosing one map over others, race still may predominate.For these reasons, a conflict or inconsistency between the enacted plan and traditional redistricting criteria is not a threshold requirement or a mandatory precondition in order for a challenger to establish a claim of racial gerrymandering. Of course, a conflict or inconsistency may be persuasive circumstantial evidence tending to show racial predomination, but there is no rule requiring challengers to present this kind of evidence in every case.As a practical matter, in many cases, perhaps most cases, challengers will be unable to prove an unconstitutional racial gerrymander without evidence that the enacted plan conflicts with traditional redistricting criteria. In general, legislatures that engage in impermissible race-based redistricting will find it necessary to depart from traditional principles in order to do so. And, in the absence of a conflict with traditional principles, it may be difficult for challengers to find other evidence sufficient to show that race was the overriding factor causing neutral considerations to be cast aside. In fact, this Court to date has not affirmed a predominance finding, or remanded a case for a determination of predominance, without evidence that some district lines deviated from traditional principles. See Alabama, 575 U. S., at ___ (slip op., at 17); Hunt v. Cromartie,526 U. S. 541,547 (1999); Bush v. Vera,517 U. S. 952,962,966,974 (1996) (plurality opinion); Shaw II, supra, at 905–906; Miller, supra, at 917; Shaw I, supra, at 635–636. Yet the law responds to proper evidence and valid inferences in ever-changing circumstances, as it learns more about ways in which its commands are circumvented. So there may be cases where challengers will be able to establish racial predominance in the absence of an actual conflict by presenting direct evidence of the legislative purpose and intent or other compelling circumstantial evidence.BThe challengers submit that the District Court erred further when it considered the legislature’s racial motive only to the extent that the challengers identified deviations from traditional redistricting criteria that were attributable to race and not to some other factor. In the challengers’ view, this approach foreclosed a holistic analysis of each district and led the District Court to give insufficient weight to the 55% BVAP target and other relevant evidence that race predominated. Again, this Court agrees.As explained, showing a deviation from, or conflict with, traditional redistricting principles is not a necessary prerequisite to establishing racial predominance. Supra, at 10. But even where a challenger alleges a conflict, or succeeds in showing one, the court should not confine its analysis to the conflicting portions of the lines. That is because the basic unit of analysis for racial gerrymandering claims in general, and for the racial predominance inquiry in particular, is the district. Racial gerrymandering claims proceed “district-by-district.” Alabama, 575 U. S., at ___ (slip op., at 6). “We have consistently described a claim of racial gerrymandering as a claim that race was improperly used in the drawing of the boundaries of one or more specific electoral districts.” Ibid. And Miller’s basic predominance test scrutinizes the legislature’s motivation for placing “a significant number of voters within or without a particular district.” 515 U. S., at 916. Courts evaluating racial predominance therefore should not divorce any portion of the lines—whatever their relationship to traditional principles—from the rest of the district.This is not to suggest that courts evaluating racial gerrymandering claims may not consider evidence pertaining to an area that is larger or smaller than the district at issue. The Court has recognized that “[v]oters, of course, can present statewide evidence in order to prove racial gerrymandering in a particular district.” Alabama, supra, at ___ (slip op., at 7) (emphasis deleted). Districts share borders, after all, and a legislature may pursue a common redistricting policy toward multiple districts. Likewise, a legislature’s race-based decisionmaking may be evident in a notable way in a particular part of a district. It follows that a court may consider evidence regarding certain portions of a district’s lines, including portions that conflict with traditional redistricting principles.The ultimate object of the inquiry, however, is the legislature’s predominant motive for the design of the district as a whole. A court faced with a racial gerrymandering claim therefore must consider all of the lines of the district at issue; any explanation for a particular portion of the lines, moreover, must take account of the districtwide context. Concentrating on particular portions in isolation may obscure the significance of relevant districtwide evidence, such as stark splits in the racial composition of populations moved into and out of disparate parts of the district, or the use of an express racial target. A holistic analysis is necessary to give that kind of evidence its proper weight.CThe challengers ask this Court not only to correct the District Court’s racial predominance standard but also to apply that standard and conclude that race in fact did predominate in the 11 districts where the District Court held that it did not. For its part, the State asks the Court to hold that, even if race did predominate in these districts, the State’s predominant use of race was narrowly tailored to the compelling interest in complying with §5.The Court declines these requests. “[O]urs is a court of final review and not first view.” Department of Transportation v. Association of American Railroads, 575 U. S. ___, ___ (2015) (slip op., at 12) (internal quotation marks omitted). The District Court is best positioned to determine in the first instance the extent to which, under the proper standard, race directed the shape of these 11 districts. And if race did predominate, it is proper for the District Court to determine in the first instance whether strict scrutiny is satisfied. These matters are left for the District Court on remand.IIIThe Court now turns to the arguments regarding District 75. Where a challenger succeeds in establishing racial predominance, the burden shifts to the State to “demonstrate that its districting legislation is narrowly tailored to achieve a compelling interest.” Miller, supra, at 920. The District Court here determined that the State’s predominant use of race in District 75 was narrowly tailored to achieve compliance with §5. The challengers contest the finding of narrow tailoring, but they do not dispute that compliance with §5 was a compelling interest at the relevant time. As in previous cases, therefore, the Court assumes, without deciding, that the State’s interest in complying with the Voting Rights Act was compelling. E.g., Alabama, supra, at ___–___ (slip op., at 19–23); Shaw II, 517 U. S., at 915.Turning to narrow tailoring, the Court explained the contours of that requirement in Alabama. When a State justifies the predominant use of race in redistricting on the basis of the need to comply with the Voting Rights Act, “the narrow tailoring requirement insists only that the legislature have a strong basis in evidence in support of the (race-based) choice that it has made.” 575 U. S., at ___ (slip op., at 22) (internal quotation marks omitted). That standard does not require the State to show that its action was “actually . . . necessary” to avoid a statutory violation, so that, but for its use of race, the State would have lost in court. Ibid. (internal quotation marks omitted). Rather, the requisite strong basis in evidence exists when the legislature has “good reasons to believe” it must use race in order to satisfy the Voting Rights Act, “even if a court does not find that the actions were necessary for statutory compliance.” Ibid. (internal quotation marks omitted).The Court now finds no error in the District Court’s conclusion that the State had sufficient grounds to determine that the race-based calculus it employed in District 75 was necessary to avoid violating §5. As explained, §5 at the time barred Virginia from adopting any districting change that would “have the effect of diminishing the ability of [members of a minority group] to elect their preferred candidates of choice.”52 U. S. C. §10304(b). Determining what minority population percentage will satisfy that standard is a difficult task requiring, in the view of the Department of Justice, a “functional analysis of the electoral behavior within the particular . . . election district.” Guidance Concerning Redistricting Under Section 5 of the Voting Rights Act, 76 Fed. Reg. 7471 (2011).Under the facts found by the District Court, the legislature performed that kind of functional analysis of District 75 when deciding upon the 55% BVAP target. Redrawing this district presented a difficult task, and the result reflected the good-faith efforts of Delegate Jones and his colleagues to achieve an informed bipartisan consensus. Delegate Jones met with Delegate Tyler “probably half a dozen times to configure her district” in order to avoid retrogression. 141 F. Supp. 3d, at 558 (internal quotation marks omitted). He discussed the district with incumbents from other majority-minority districts. He also considered turnout rates, the results of the recent contested primary and general elections in 2005, and the dis-trict’s large population of disenfranchised black prisoners. The challengers, moreover, do not dispute that District 75 was an ability-to-elect district, or that white and black voters in the area tend to vote as blocs. See id., at 557–559. In light of Delegate Jones’ careful assessment of local conditions and structures, the State had a strong basis in evidence to believe a 55% BVAP floor was required to avoid retrogression.The challengers’ responses ask too much from state officials charged with the sensitive duty of reapportioning legislative districts. First, the challengers contest the sufficiency of the evidence showing that Delegate Jones in fact performed a functional analysis, in part because that analysis was not memorialized in writing. But the District Court’s factual findings are reviewed only for clear error. See Easley v. Cromartie,532 U. S. 234,242 (2001). The findings regarding how the legislature arrived at the 55% BVAP target are well supported, and “we do not . . . require States engaged in redistricting to compile a comprehensive administrative record.” Vera, 517 U. S., at 966 (internal quotation marks omitted).The challengers argue further that the drafters of the plan had insufficient evidence to justify a 55% BVAP floor. The 2005 elections were idiosyncratic, the challengers contend; moreover, demographic information about the prison in the district is absent from the record, and Delegate Tyler’s perspective was influenced by a personal interest in reelection. That may have been so, and for those reasons, it is possible that, if the State had drawn District 75 with a BVAP below 55% and had sought judicial preclearance, a court would have found no §5 violation. But that is not the question here. “The law cannot insist that a state legislature, when redistricting, determine precisely what percent minority population §5 demands.” Alabama, 575 U. S., at ___ (slip op., at 22). The question is whether the State had “good reasons” to believe a 55% BVAP floor was necessary to avoid liability under §5. Ibid. (internal quotation marks omitted). The State did have good reasons under these circumstances. Holding otherwise would afford state legislatures too little breathing room, leaving them “trapped between the competing hazards of liability” under the Voting Rights Act and the Equal Protection Clause. Vera, supra, at 977 (internal quotation marks omitted).As a final point, the challengers liken the 55% BVAP floor here to the “mechanically numerical view” of §5 this Court rejected in Alabama. 575 U. S., at ___ (slip op., at 21). But Alabama did not condemn the use of BVAP targets to comply with §5 in every instance. Rather, this Court corrected the “misperception” that §5 required a State to “maintai[n] the same population percentages in majority-minority districts as in the prior plan.” Id., at ___–___ (slip op., at 19–20). “[I]t would seem highly unlikely,” the Court explained, that reducing a district’s BVAP “from, say, 70% to 65% would have a significant impact on the black voters’ ability to elect their preferred candidate.” Id., at ___ (slip op., at 21). Yet reducing the BVAP below 55% well might have that effect in some cases. The record here supports the legislature’s conclusion that this was one instance where a 55% BVAP was necessary for black voters to have a functional working majority.IVThe Court’s holding in this case is controlled by precedent. The Court reaffirms the basic racial predominance analysis explained in Miller and Shaw II, and the basic narrow tailoring analysis explained in Alabama. The District Court’s judgment as to District 75 is consistent with these principles. Applying these principles to the remaining 11 districts is entrusted to the District Court in the first instance.The judgment of the District Court is affirmed in part and vacated in part. The case is remanded for further proceedings consistent with this opinion.It is so ordered.

581.US.2016_16-405

The Federal Employers’ Liability Act (FELA), 45 U. S. C. §51 et seq., makes railroads liable in money damages to their employees for on-the-job injuries. Respondent Robert Nelson, a North Dakota resident, brought a FELA suit against petitioner BNSF Railway Company (BNSF) in a Montana state court, alleging that he had sustained injuries while working for BNSF. Respondent Kelli Tyrrell, appointed in South Dakota as the administrator of her husband Brent Tyrrell’s estate, also sued BNSF under FELA in a Montana state court, alleging that Brent had developed a fatal cancer from his exposure to carcinogenic chemicals while working for BNSF. Neither worker was injured in Montana. Neither incorporated nor headquartered there, BNSF maintains less than 5% of its work force and about 6% of its total track mileage in the State. Contending that it is not “at home” in Montana, as required for the exercise of general personal jurisdiction under Daimler AG v. Bauman, 571 U. S. ___, ___, BNSF moved to dismiss both suits. Its motion was granted in Nelson’s case and denied in Tyrrell’s. After consolidating the two cases, the Montana Supreme Court held that Montana courts could exercise general personal jurisdiction over BNSF because the railroad both “d[id] business” in the State within the meaning of 45 U. S. C. §56 and was “found within” the State within the compass of Mont. Rule Civ. Proc. 4(b)(1). The due process limits articulated in Daimler, the court added, did not control because Daimler did not involve a FELA claim or a railroad defendant. Held: 1. Section 56 does not address personal jurisdiction over railroads. Pp. 4–9. (a) Section 56’s first relevant sentence provides that “an action may be brought in a district court of the United States,” in, among other places, the district “in which the defendant shall be doing business at the time of commencing such action.” This Court has comprehended that sentence as a venue prescription, not as one governing personal jurisdiction. Baltimore & Ohio R. Co. v. Kepner, 314 U. S. 44 . Congress generally uses the expression, where suit “may be brought,” to indicate the federal districts in which venue is proper, see, e.g., 28 U. S. C. §1391(b), while it typically provides for the exercise of personal jurisdiction by authorizing service of process, see, e.g., 15 U. S. C. §22. Nelson and Tyrrell contend that the 1888 Judiciary Act provision that prompted §56’s enactment concerned both personal jurisdiction and venue, but this Court has long read that Judiciary Act provision to concern venue only, see, e.g., Green v. Chicago, B. & Q. R. Co., 205 U. S. 530 –533. Pp. 5–7. (b) The second relevant sentence of §56—that “[t]he jurisdiction of the courts of the United States under this chapter shall be concurrent with that of the courts of the several States”—refers to concurrent subject-matter jurisdiction of state and federal courts over FELA actions. See Second Employers’ Liability Cases, 223 U. S. 1 –56. Congress added this clarification after the Connecticut Supreme Court held that Congress intended to confine FELA litigation to federal courts, and that state courts had no obligation to entertain FELA claims. Pp. 7–8. (c) None of the cases featured by the Montana Supreme Court in reaching its contrary conclusion resolved a question of personal jurisdiction. Pope v. Atlantic Coast Line R. Co., 345 U. S. 379 ; Miles v. Illinois Central R. Co., 315 U. S. 698 ; Kepner, 314 U. S. 44 ; and Denver & Rio Grande Western R. Co. v. Terte, 284 U. S. 284 , distinguished. Moreover, all these cases, save Pope, were decided before this Court’s transformative decision on personal jurisdiction in International Shoe Co. v. Washington, 326 U. S. 310 . Pp. 8–9. 2. The Montana courts’ exercise of personal jurisdiction under Montana law does not comport with the Fourteenth Amendment’s Due Process Clause. Only the propriety of general personal jurisdiction is at issue here because neither Nelson nor Tyrrell alleges injury from work in or related to Montana. A state court may exercise general jurisdiction over out-of-state corporations when their “affiliations with the State are so ‘continuous and systematic’ as to render them essentially at home in the forum State.” Daimler, 571 U. S., at ___. The “paradigm” forums in which a corporate defendant is “at home” are the corporation’s place of incorporation and its principal place of business, e.g., id., at ___, but in an “exceptional case,” a corporate defendant’s operations in another forum “may be so substantial and of such a nature as to render the corporation at home in that State,” id., at ___, n. 19. Daimler involved no FELA claim or railroad defendant, but the due process constraint described there applies to all state-court assertions of general jurisdiction over nonresident defendants; that constraint does not vary with the type of claim asserted or business enterprise sued. Here, BNSF is not incorporated or headquartered in Montana and its activity there is not “so substantial and of such a nature as to render the corporation at home in that State.” Ibid. Pp. 9–12. 383 Mont. 417, 373 P. 3d 1, reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Breyer, Alito, Kagan, and Gorsuch, JJ., joined. Sotomayor, J., filed an opinion concurring in part and dissenting in part.

The two cases we decide today arise under the Federal Employers’ Liability Act (FELA), 35Stat. 65, as amended, 45 U. S. C. §51 et seq., which makes railroads liable in money damages to their employees for on-the-job injuries. Both suits were pursued in Montana state courts although the injured workers did not reside in Montana, nor were they injured there. The defendant railroad, BNSF Railway Company (BNSF), although “doing business” in Montana when the litigation commenced, was not incorporated in Montana, nor did it maintain its principal place of business in that State. To justify the exercise of personal jurisdiction over BNSF, the Montana Supreme Court relied on §56, which provides in relevant part: “Under this chapter an action may be brought in a district court of the United States, in the district of the residence of the defendant, or in which the cause of action arose, or in which the defendant shall be doing business at the time of commencing such action. The jurisdiction of the courts of the United States under this chapter shall be concurrent with that of the courts of the several States.” We hold that §56 does not address personal jurisdiction over railroads. Its first relevant sentence is a venue prescription governing proper locations for FELA suits filed in federal court. The provision’s second relevant sentence, using the term “concurrent” jurisdiction, refers to subject-matter jurisdiction, not personal jurisdiction. It simply clarifies that the federal courts do not have exclusive subject-matter jurisdiction over FELA suits; state courts can hear them, too. Montana’s Supreme Court, in the alternative, relied on state law, under which personal jurisdiction could be asserted over “persons found within . . . Montana.” Mont. Rule Civ. Proc. 4(b)(1) (2015). BNSF fit that bill, the court stated, because it has over 2,000 miles of railroad track and employs more than 2,000 workers in Montana. Our precedent, however, explains that the Fourteenth Amendment’s Due Process Clause does not permit a State to hale an out-of-state corporation before its courts when the corporation is not “at home” in the State and theepisode-in-suit occurred elsewhere. Daimler AG v. Bauman, 571 U. S. ___, ___ (2014) (slip op., at 8) (internal quotation marks omitted). We therefore reverse the judgment of the Montana Supreme Court. I In March 2011, respondent Robert Nelson, a North Dakota resident, brought a FELA suit against BNSF in a Montana state court to recover damages for knee injuries Nelson allegedly sustained while working for BNSF as a fuel-truck driver. 383 Mont. 417, 419, 373 P. 3d 1, 3 (2016). In May 2014, respondent Kelli Tyrrell, appointed in South Dakota as the administrator of her husband Brent Tyrrell’s estate, similarly sued BNSF under FELA in a Montana state court. Id., at 419–420, 373 P. 3d, at 3. Brent Tyrrell, his widow alleged, had developed a fatal kidney cancer from his exposure to carcinogenic chemicals while working for BNSF. Id., at 420, 373 P. 3d, at 3. Neither plaintiff alleged injuries arising from or related to work performed in Montana; indeed, neither Nelson nor Brent Tyrrell appears ever to have worked for BNSF in Montana. Id., at 419–420, 373 P. 3d, at 3. BNSF is incorporated in Delaware and has its principal place of business in Texas. Id., at 419, 373 P. 3d, at 3. It operates railroad lines in 28 States. No. DV 14–699 (13th Jud. Dist., Yellowstone Cty., Mont., Oct. 7, 2014), App. to Pet. for Cert. 63a. BNSF has 2,061 miles of railroad track in Montana (about 6% of its total track mileage of 32,500), employs some 2,100 workers there (less than 5% of its total work force of 43,000), generates less than 10% of its total revenue in the State, and maintains only one of its 24 automotive facilities in Montana (4%). Ibid. Contending that it is not “at home” in Montana, as required for the exercise of general personal jurisdiction under Daimler AG v. Bauman, 571 U. S. ___, ___ (2014) (slip op., at 8) (internal quotation marks omitted), BNSF moved to dismiss both suits for lack of personal jurisdiction. Its motion was granted in Nelson’s case and denied in Tyrrell’s. 383 Mont., at 419, 373 P. 3d, at 2. After consolidating the two cases, the Montana Supreme Court held that Montana courts could exercise general personal jurisdiction over BNSF. Id., at 429, 373 P. 3d, at 9. Section 56, the court determined, authorizes state courts to exercise personal jurisdiction over railroads “doing business” in the State. Id., at 426, 373 P. 3d, at 7 (internal quotation marks omitted). In addition, the court observed, Montana law provides for the exercise of general jurisdiction over “[a]ll persons found within” the State. Id., at 427, 373 P. 3d, at 8 (quoting Mont. Rule Civ. Proc. 4(b)(1) (2015)). In view of the railroad’s many employees and miles of track in Montana, the court concluded, BNSF is both “doing business” and “found within” the State, such that both FELA and Montana law authorized the exercise of personal jurisdiction. 383 Mont., at 426, 428, 373 P. 3d, at 7–8 (internal quotation marks omitted). The due process limits articulated in Daimler, the court added, did not control, because Daimler did not involve a FELA claim or a railroad defendant. 383 Mont., at 424, 373 P. 3d, at 6. Justice McKinnon dissented. Section 56, she wrote, is a federal-court venue prescription, and also confers subject-matter jurisdiction on state courts in FELA cases, concurrent with federal courts. Id., at 435–437, 373 P. 3d, at 13. But §56, she maintained, does not touch or concern personal jurisdiction. Ibid. Furthermore, she concluded, Daimler controls, rendering the Montana courts’ exercise of personal jurisdiction impermissible because BNSF is not “at home” in Montana. 383 Mont., at 433–434, 373 P. 3d, at 11–12. We granted certiorari, 580 U. S. ___ (2017), to resolve whether §56 authorizes state courts to exercise personal jurisdiction over railroads doing business in their States but not incorporated or headquartered there, and whether the Montana courts’ exercise of personal jurisdiction in these cases comports with due process. II Nelson and Tyrrell contend that §56’s first relevant sentence confers personal jurisdiction on federal courts, and that the section’s second relevant sentence extends that grant of jurisdiction to state courts. Neither contention is tenable. Section 56’s first relevant sentence concerns venue; its next sentence speaks to subject-matter jurisdiction.[1] A The first sentence of §56 states that “an action may be brought in a district court of the United States,” in, among other places, the district “in which the defendant shall be doing business at the time of commencing such action.” In Baltimore & Ohio R. Co. v. Kepner, 314 U. S. 44 (1941) , we comprehended this clause as “establish[ing] venue” for a federal-court action. Id., at 52. Congress, we explained, designed §56 to expand venue beyond the limits of the 1888 Judiciary Act’s general venue provision, which allowed suit only “in districts of which the defendant was an inhabitant.” Id., at 49; see Act of Aug. 13, 1888, §1, 25Stat. 434. Nowhere in Kepner or in any other decision did we intimate that §56 might affect personal jurisdiction. Congress generally uses the expression, where suit “may be brought,” to indicate the federal districts in which venue is proper. See, e.g., 28 U. S. C. §1391(b) (general venue statute specifying where “[a] civil action may be brought”); J. Oakley, ALI, Fed. Judicial Code Rev. Project 253–290 (2004) (listing special venue statutes, many with similar language). See also Kepner, 314 U. S., at 56 (Frankfurter, J., dissenting) (“The phrasing of [§56] follows the familiar pattern generally employed by Congress in framing venue provisions.”). In contrast, Congress’ typical mode of providing for the exercise of personal jurisdiction has been to authorize service of process. See, e.g., 15 U. S. C. §22 (Clayton Act provision stating that “all process in [cases against a corporation arising under federal antitrust laws] may be served in the district of which [the defendant] is an inhabitant, or wherever [the defendant] may be found”); §53(a) (under Federal Trade Commission Act, “process may be served on any person, partnership, or corporation wherever it may be found”). See also Omni Capital Int’l, Ltd. v. Rudolf Wolff & Co., 484 U. S. 97 –107 (1987) (discussing statutes that authorize (or fail to authorize) nationwide service of process). But cf. Schlanger v. Seamans, 401 U. S. 487 , n. 4 (1971) (though “Congress has provided for nationwide service of process” in 28 U. S. C. §1391(e) (1964 ed., Supp. V), that statute was meant to expand venue, not personal jurisdiction). Congress uses this terminology because, absent consent, a basis for service of a summons on the defendant is prerequisite to the exercise of personal jurisdiction. See Omni Capital, 484 U. S., at 104. Nelson and Tyrrell, however, argue that §56 relates to personal jurisdiction. In their view, the 1888 Judiciary Act provision that prompted §56’s enactment, 25Stat. 434, concerned both personal jurisdiction and venue. According to House and Senate Reports, they contend, two cases had brought to Congress’ attention the problem with the prior provision—namely, that in federal-question cases it authorized suit only in the district of the defendant’s residence. Brief for Respondents 16–18. See H. R. Rep. No. 513, 61st Cong., 2d Sess., 6 (1910) (citing Macon Grocery Co. v. Atlantic Coast Line R. Co., 215 U. S. 501 (1910) ; Cound v. Atchison, T. & S. F. R. Co., 173 F. 527 (WD Tex. 1909)); S. Rep. No. 432, 61st Cong., 2d Sess., 4 (1910) (same). In both cases, the courts had dismissed FELA suits for “want of jurisdiction.” Macon Grocery, 215 U. S., at 510; Cound, 173 F., at 534. To avert such jurisdictional dismissals, they urge, Congress enacted §56. Legislative history “throws little light” here. Kepner, 314 U. S., at 50.[2] Driving today’s decision, we have long read the 1888 Judiciary Act provision to concern venue only. See Green v. Chicago, B. & Q. R. Co., 205 U. S. 530 –533 (1907) (analyzing personal jurisdiction separately, after concluding that venue was proper under 1888 Judiciary Act provision). See also Lee v. Chesapeake & Ohio R. Co., 260 U. S. 653, 655 (1923) (noting that materially identical successor to 1888 Judiciary Act provision, Act of Mar. 3, 1911, §51, 36Stat. 1101, “relates to the venue of suits”). Indeed, reading the 1888 Judiciary Act provision to authorize the exercise of personal jurisdiction would have yielded an anomalous result: In diversity cases, the provision allowed for suit “in the district of the residence of either the plaintiff or the defendant.” 25Stat. 434. Interpreting that clause to provide for jurisdiction would have allowed a plaintiff to hale a defendant into court in the plaintiff’s home district, even if the district was one with which the defendant had no affiliation, and the episode-in-suit, no connection. B The second §56 sentence in point provides that “[t]he jurisdiction of the courts of the United States under this chapter shall be concurrent with that of the courts of the several States.” Nelson and Tyrrell argue that this sentence extends to state courts the first sentence’s alleged conferral of personal jurisdiction on federal courts. But, as just discussed, the first sentence concerns federal-court venue and confers no personal jurisdiction on any court. We have understood §56’s second sentence to provide for the concurrent subject-matter jurisdiction of state and federal courts over actions under FELA. See Second Employers’ Liability Cases, 223 U. S. 1 –56 (1912). As Nelson and Tyrrell acknowledge, Congress added the provision to confirm concurrent subject-matter jurisdiction after the Connecticut Supreme Court held that Congress intended to confine FELA litigation to federal courts, and that state courts had no obligation to entertain FELA claims. See Brief for Respondents 23 (citing Hoxie v. New York, N. H. & H. R. Co., 82 Conn. 352, 73 A. 754 (1909)). As Justice McKinnon recognized in her dissent from the Montana Supreme Court’s decision in Nelson’s and Tyrrell’s cases, “[t]he phrase ‘concurrent jurisdiction’ is a well-known term of art long employed by Congress and courts to refer to subject-matter jurisdiction, not personal jurisdiction.” 383 Mont., at 436, 373 P. 3d, at 13. See, e.g., Mims v. Arrow Financial Services, LLC, 565 U. S. 368, 372 (2012) (“federal and state courts have concurrent jurisdiction over private suits arising under the [Telephone Consumer Protection Act of 1991, 47 U. S. C. §227]”); Claflin v. Houseman, 93 U. S. 130 –134 (1876) (State courts retain “concurrent jurisdiction” over “suits in which a bankrupt” party is involved, notwithstanding exclusive federal jurisdiction over bankruptcy matters). C Pointing to a quartet of cases, the Montana Supreme Court observed that this Court “consistently has interpreted [§]56 to allow state courts to hear cases brought under FELA even where the only basis for jurisdiction is the railroad doing business in the forum [S]tate.” 383 Mont., at 421–423, 425–426, 373 P. 3d, at 4–7 (citing Pope v. Atlantic Coast Line R. Co., 345 U. S. 379 (1953) ; Miles v. Illinois Central R. Co., 315 U. S. 698 (1942) ; Kepner, 314 U. S. 44 ; Denver & Rio Grande Western R. Co. v. Terte, 284 U. S. 284 (1932) ). None of the decisions featured by the Montana Supreme Court resolved a question of personal jurisdiction. Terte held that a FELA plaintiff, injured in Colorado, could bring suit in Missouri state court against a railroad incorporated elsewhere. 284 U. S., at 286–287. The dispute, however, was over the Dormant Commerce Clause, not personal jurisdiction; the railroad defendants argued that the suit would unduly burden interstate commerce, and the decision rested on two Commerce Clause decisions, Michigan Central R. Co. v. Mix, 278 U. S. 492 (1929) , and Hoffman v. Missouri ex rel. Foraker, 274 U. S. 21 (1927) , not on an interpretation of §56. See Terte, 284 U. S., at 285, 287. In Kepner and Miles, this Court held that a state court may not, based on inconvenience to a railroad defendant, enjoin its residents from bringing a FELA suit in another State’s federal (Kepner) or state (Miles) courts. Kepner, 314 U. S., at 54; Miles, 315 U. S., at 699–700, 704. Pope held that 28 U. S. C. §1404(a)’s provision for transfer from one federal court to another did not bear on the question decided in Miles: A state court still could not enjoin a FELA action brought in another State’s courts. 345 U. S., at 383–384. Moreover, all these cases, save Pope, were decided before this Court’s transformative decision on personal jurisdiction in International Shoe Co. v. Washington, 326 U. S. 310 (1945) . See Daimler, 571 U. S., at ___, n. 18 (slip op., at 20, n. 18) (cautioning against reliance on cases “decided in the era dominated by” the “territorial thinking” of Pennoyer v. Neff, 95 U. S. 714 (1878) ). III Because FELA does not authorize state courts to exercise personal jurisdiction over a railroad solely on the ground that the railroad does some business in their States, the Montana courts’ assertion of personal jurisdiction over BNSF here must rest on Mont. Rule Civ. Proc. 4(b)(1), the State’s provision for the exercise of personal jurisdiction over “persons found” in Montana. See supra, at 2–3. BNSF does not contest that it is “found within” Montana as the State’s courts comprehend that rule. We therefore inquire whether the Montana courts’ exercise of personal jurisdiction under Montana law comports with the Due Process Clause of the Fourteenth Amendment. In International Shoe, this Court explained that a state court may exercise personal jurisdiction over an out-of-state defendant who has “certain minimum contacts with [the State] such that the maintenance of the suit does not offend ‘traditional notions of fair play and substantial justice.’ ” 326 U. S., at 316. Elaborating on this guide, we have distinguished between specific or case-linked jurisdiction and general or all-purpose jurisdiction. See, e.g., Daimler, 571 U. S., at ___ (slip op., at 8); Goodyear Dunlop Tires Operations, S. A. v. Brown, 564 U. S. 915, 919 (2011) ; Helicopteros Nacionales de Colombia, S. A. v. Hall, 466 U. S. 408 , nn. 8, 9 (1984). Because neither Nelson nor Tyrrell alleges any injury from work in or related to Montana, only the propriety of general jurisdiction is at issue here. Goodyear and Daimler clarified that “[a] court may assert general jurisdiction over foreign (sister-state or foreign-country) corporations to hear any and all claims against them when their affiliations with the State are so ‘continuous and systematic’ as to render them essentially at home in the forum State.” Daimler, 571 U. S., at ___ (slip op., at 8) (quoting Goodyear, 564 U. S., at 919). The “paradigm” forums in which a corporate defendant is “at home,” we explained, are the corporation’s place of incorporation and its principal place of business. Daimler, 571 U. S., at ___ (slip op., at 18–19); Goodyear, 564 U. S., at 924. The exercise of general jurisdiction is not limited to these forums; in an “exceptional case,” a corporate defendant’s operations in another forum “may be so substantial and of such a nature as to render the corporation at home in that State.” Daimler, 571 U. S., at ___, n. 19 (slip op., at 20, n. 19). We suggested that Perkins v. Benguet Consol. Mining Co., 342 U. S. 437 (1952) , exemplified such a case. Daimler, 571 U. S., at ___, n. 19 (slip op., at 20, n. 19). In Perkins, war had forced the defendant corporation’s owner to temporarily relocate the enterprise from the Philippines to Ohio. 342 U. S., at 447–448. Because Ohio then became “the center of the corporation’s wartime activities,” Daimler, 571 U. S., at ___, n. 8 (slip op., at 12, n. 8), suit was proper there, Perkins, 342 U. S., at 448. The Montana Supreme Court distinguished Daimler on the ground that we did not there confront “a FELA claim or a railroad defendant.” 383 Mont., at 424, 373 P. 3d, at 6. The Fourteenth Amendment due process constraint described in Daimler, however, applies to all state-court assertions of general jurisdiction over nonresident defendants; the constraint does not vary with the type of claim asserted or business enterprise sued.[3] BNSF, we repeat, is not incorporated in Montana and does not maintain its principal place of business there. Nor is BNSF so heavily engaged in activity in Montana “as to render [it] essentially at home” in that State. See Daimler, 571 U. S., at ___ (slip op., at 8) (internal quotation marks omitted). As earlier noted, BNSF has over 2,000 miles of railroad track and more than 2,000 employees in Montana. But, as we observed in Daimler, “the general jurisdiction inquiry does not focus solely on the magnitude of the defendant’s in-state contacts.” Id., at ___, n. 20 (slip op., at 21, n. 20) (internal quotation marks and alterations omitted). Rather, the inquiry “calls for an appraisal of a corporation’s activities in their entirety”; “[a] corporation that operates in many places can scarcely be deemed at home in all of them.” Ibid. In short, the business BNSF does in Montana is sufficient to subject the railroad to specific personal jurisdiction in that State on claims related to the business it does in Montana. But in-state business, we clarified in Daimler and Goodyear, does not suffice to permit the assertion of general jurisdiction over claims like Nelson’s and Tyrrell’s that are unrelated to any activity occurring in Montana.[4] IV Nelson and Tyrrell present a further argument—that BNSF has consented to personal jurisdiction in Montana. See Brief for Respondents 50–51. The Montana Supreme Court did not address this contention, see 383 Mont., at 429, n. 3, 373 P. 3d, at 9, n. 3, so we do not reach it. See Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005) (“[W]e are a court of review, not of first view.”). * * * For the reasons stated, the judgment of the Montana Supreme Court is reversed, and the cases are remanded for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 Section 56’s first sentence, which provides a time bar for FELA claims, is not relevant to the issue at hand. For ease of reference, we hereinafter refer to the first relevant sentence, describing where suit “may be brought,” as the provision’s “first” sentence, and the sentence that immediately follows, referring to “concurrent” jurisdiction, as the “second.” 2 We note, moreover, that Nelson and Tyrrell overlooked the Senate Report’s explicit reference to the first sentence of §56 as a venue provision, with no mention of personal jurisdiction. S. Rep. No. 432, 61st Cong., 2d Sess., 3 (1910). 3 The Montana Supreme Court also erred in asserting that “Congress drafted the FELA to make a railroad ‘at home’ for jurisdictional purposes wherever it is ‘doing business.’ ” 383 Mont. 417, 425, 373 P. 3d 1, 6 (2016). As discussed, supra, at 5–7, in §56’s first sentence, Congress dealt with venue only, not personal jurisdiction. 4 Justice Sotomayor, dissenting in part, renews a debate comprehensively aired in Daimler AG v. Bauman, 571 U. S. ___ (2014). There, as again here, Justice Sotomayor treats the assertion of jurisdiction by the State of Washington courts in International Shoe Co. v. Washington, 326 U. S. 310 (1945) , as an exercise of general, dispute-blind, jurisdiction, post, at 3, thereby overlooking the fundamental difference between International Shoe and these cases. In International Shoe, the defendant corporation’s in-state activities had “not only been continuous and systematic, but also g[a]ve rise to the liabilities sued on.” 326 U. S., at 317. The state courts there asserted jurisdiction not over claims that had nothing to do with the State; instead, they exercised adjudicatory authority to hold the defendant corporation accountable for activity pursued within the State of Washington. Daimler, 571 U. S., at ___, ___, n. 10 (slip op., at 7, 14, n. 10). This Court, therefore, had no occasion in International Shoe to “engage in a comparison between International Shoe’s contacts within the State of Washington and the other States in which it operated.” Post, at 3. In marked contrast to International Shoe, Nelson’s and Tyrrell’s claims have no relationship to anything that occurred or had its principal impact in Montana.

580.US.2016_15-537

The issue-preclusion component of the Double Jeopardy Clause bars a second contest of an issue of fact or law raised and necessarily resolved by a prior judgment. Ashe v. Swenson, 397 U. S. 436 . The burden is on the defendant to demonstrate that the issue he seeks to shield from reconsideration was actually decided by a prior jury’s verdict of acquittal. Schiro v. Farley, 510 U. S. 222 . When the same jury returns irreconcilably inconsistent verdicts on the issue in question, a defendant cannot meet that burden. The acquittal, therefore, gains no preclusive effect regarding the count of conviction. United States v. Powell, 469 U. S. 57 –69. Issue preclusion does, however, attend a jury’s verdict of acquittal if the same jury in the same proceeding fails to reach a verdict on a different count turning on the same issue of ultimate fact. Yeager v. United States, 557 U. S. 110 –122. In this case, a jury convicted petitioners Juan Bravo-Fernandez (Bravo) and Hector Martínez-Maldonado (Martínez) of bribery in violation of 18 U. S. C. §666. Simultaneously, the jury acquitted them of conspiring to violate §666 and traveling in interstate commerce to violate §666. Because the only contested issue at trial was whether Bravo and Martínez had violated §666 (the other elements of the acquitted charges—agreement and travel—were undisputed), the jury’s verdicts were irreconcilably inconsistent. Unlike the guilty verdicts in Powell, however, petitioners’ convictions were later vacated on appeal because of error in the judge’s instructions unrelated to the verdicts’ inconsistency. In the First Circuit’s view, §666 proscribes only quid pro quo bribery, yet the charge had permitted the jury to find petitioners guilty on a gratuity theory. On remand, Bravo and Martínez moved for judgments of acquittal on the standalone §666 charges. They argued that the issue-preclusion component of the Double Jeopardy Clause barred the Government from retrying them on those charges because the jury necessarily determined that they were not guilty of violating §666 when it acquitted them of the related conspiracy and Travel Act offenses. The District Court denied the motions, and the First Circuit affirmed, holding that the eventual invalidation of petitioners’ §666 convictions did not undermine Powell’s instruction that issue preclusion does not apply when the same jury returns logically inconsistent verdicts. Held: The issue-preclusion component of the Double Jeopardy Clause does not bar the Government from retrying defendants, like petitioners, after a jury has returned irreconcilably inconsistent verdicts of conviction and acquittal and the convictions are later vacated for legal error unrelated to the inconsistency. Pp. 12–19. (a) Because petitioners’ trial yielded incompatible jury verdicts, petitioners cannot establish that the jury necessarily resolved in their favor the question whether they violated §666. In view of the Government’s inability to obtain review of the acquittals, Powell, 469 U. S., at 68, the inconsistent jury findings weigh heavily against according those acquittals issue-preclusive effect. The subsequent vacatur of petitioners’ bribery convictions does not alter this analysis. The critical inquiry is whether the jury actually decided that petitioners did not violate §666. Ashe instructs courts to approach that task with “realism and rationality,” 397 U. S., at 444, in particular, to examine the trial record “with an eye to all the circumstances of the proceedings,” ibid. The jury’s verdicts convicting petitioners of violating §666 remain relevant to this practical inquiry, even if the convictions are later vacated on appeal for unrelated trial error. Petitioners could not be retried if the Court of Appeals had vacated their §666 bribery convictions because of insufficient evidence, see Burks v. United States, 437 U. S. 1 –11, or if the trial error could resolve the apparent inconsistency in the jury’s verdicts. But the evidence here was sufficient to convict petitioners on the quid pro quo bribery theory the First Circuit approved. And the instructional error cannot account for the jury’s inconsistent determinations, for the error applied equally to every §666-related count. Pp. 12–16. (b) Petitioners argue that vacated judgments should be excluded from the Ashe inquiry because vacated convictions, like the hung counts in Yeager, are legal nullities that “have never been accorded respect as a matter of law or history.” Yeager, 557 U. S., at 124. That argument misapprehends the Ashe inquiry. Bravo and Martínez bear the burden of showing that the issue whether they violated §666 has been “determined by a valid and final judgment of acquittal.” 557 U. S., at 119 (internal quotation marks omitted). To judge whether they carried that burden, a court must realistically examine the record to identify the ground for the §666-based acquittals. Ashe, 397 U. S., at 444. A conviction that contradicts those acquittals is plainly relevant to that determination, no less so simply because it is later overturned on appeal for unrelated legal error. See Powell, 469 U. S., at 65. Petitioners further contend that, under Yeager, the §666 convictions are meaningless because the jury was allowed to convict on the basis of conduct not criminal in the First Circuit—payment of a gratuity. But Yeager did not rest on a court’s inability to detect the basis for a decision the jury in fact rendered. Rather, when a jury hangs, there is no decision, hence no inconsistency. 557 U. S., at 124–125. By contrast, a verdict of guilt is a jury decision, even if subsequently vacated, and therefore can evince jury inconsistency. That is the case here. Petitioners gained a second trial on the standalone bribery charges, but they are not entitled to more. Issue preclusion is not a doctrine they can commandeer when inconsistent verdicts shroud in mystery what the jury necessarily decided. Pp. 16–19. 790 F. 3d 41, affirmed. Ginsburg, J., delivered the opinion for a unanimous Court. Thomas, J., filed a concurring opinion.

This case concerns the issue-preclusion component of the Double Jeopardy Clause.[1] In criminal prosecutions, as in civil litigation, the issue-preclusion principle means that “when an issue of ultimate fact has once been determined by a valid and final judgment, that issue cannot again be litigated between the same parties in any future lawsuit.” Ashe v. Swenson, 397 U. S. 436, 443 (1970) . Does issue preclusion apply when a jury returns inconsistent verdicts, convicting on one count and acquitting on another count, where both counts turn on the very same issue of ultimate fact? In such a case, this Court has held, both verdicts stand. The Government is barred by the Double Jeopardy Clause from challenging the acquittal, see Green v. United States, 355 U. S. 184, 188 (1957) , but because the verdicts are rationally irreconcilable, the acquittal gains no preclusive effect, United States v. Powell, 469 U. S. 57, 68 (1984) . Does issue preclusion attend a jury’s acquittal verdict if the same jury in the same proceeding fails to reach a verdict on a different count turning on the same critical issue? This Court has answered yes, in those circumstances, the acquittal has preclusive force. Yeager v. United States, 557 U. S. 110 –122 (2009). As “there is no way to decipher what a hung count represents,” we reasoned, a jury’s failure to decide “has no place in the issue-preclusion analysis.” Ibid.; see id., at 125 (“[T]he fact that a jury hangs is evidence of nothing—other than, of course, that it has failed to decide anything.”). In the case before us, the jury returned irreconcilably inconsistent verdicts of conviction and acquittal. Without more, Powell would control. There could be no retrial of charges that yielded acquittals but, in view of the inconsistent verdicts, the acquittals would have no issue-preclusive effect on charges that yielded convictions. In this case, however, unlike Powell, the guilty verdicts were vacated on appeal because of error in the judge’s instructions unrelated to the verdicts’ inconsistency. Petitioners urge that, just as a jury’s failure to decide has no place in issue-preclusion analysis, so vacated guilty verdicts should not figure in that analysis. We hold otherwise. One cannot know from the jury’s report why it returned no verdict. “A host of reasons” could account for a jury’s failure to decide—“sharp dis-agreement, confusion about the issues, exhaustion after a long trial, to name but a few.” Yeager, 557 U. S., at 121. But actual inconsistency in a jury’s verdicts is a reality; vacatur of a conviction for unrelated legal error does not reconcile the jury’s inconsistent returns. We therefore bracket this case with Powell, not Yeager, and affirm the judgment of the Court of Appeals, which held that issue preclusion does not apply when verdict inconsistency renders unanswerable “what the jury necessarily decided.” 790 F. 3d 41, 47 (CA1 2015). I A The doctrine of claim preclusion instructs that a final judgment on the merits “foreclos[es] successive litigation of the very same claim.” New Hampshire v. Maine, 532 U. S. 742, 748 (2001) ; see Restatement (Second) of Judgments §19, p. 161 (1980) (hereinafter Restatement). So instructing, the doctrine serves to “avoid multiple suits on identical entitlements or obligations between the same parties.” 18 C. Wright, A. Miller, & E. Cooper, Federal Practice and Procedure §4402, p. 9 (2d ed. 2002) (herein-after Wright & Miller). Long operative in civil litigation, Restatement, at 2, claim preclusion is also essential to the Constitution’s prohibition against successive criminal prosecutions. No person, the Double Jeopardy Clause states, shall be “subject for the same offense to be twice put in jeopardy of life or limb.” Amdt. 5. The Clause “protects against a second prosecution for the same offense after conviction”; as well, “[i]t protects against a second prosecution for the same offense after acquittal.” North Carolina v. Pearce, 395 U. S. 711, 717 (1969) . “[A] verdict of acquittal [in our justice system] is final,” the last word on a criminal charge, and therefore operates as “a bar to a subsequent prosecution for the same offense.” Green v. United States, 355 U. S. 184, 188 (1957) . The allied doctrine of issue preclusion ordinarily bars relitigation of an issue of fact or law raised and necessarily resolved by a prior judgment. See Restatement §§17, 27, at 148, 250; Wright & Miller §4416, at 386. It applies in both civil and criminal proceedings, with an important distinction. In civil litigation, where issue preclusion and its ramifications first developed, the availability of appellate review is a key factor. Restatement §28, Comment a, at 274; see id., §28, Reporter’s Note, at 284 (noting “the pervasive importance of reviewability in the application of preclusion doctrine”). In significant part, preclusion doctrine is premised on “an underlying confidence that the result achieved in the initial litigation was substantially correct.” Standefer v. United States, 447 U. S. 10 , n. 18 (1980); see Restatement §29, Comment f, at 295. “In the absence of appellate review,” we have observed, “such confidence is often unwarranted.” Standefer, 447 U. S., at 23, n. 18. In civil suits, inability to obtain review is exceptional; it occurs typically when the controversy has become moot. In criminal cases, however, only one side (the defendant) has recourse to an appeal from an adverse judgment on the merits. The Government “cannot secure appellate review” of an acquittal, id., at 22, even one “based upon an egregiously erroneous foundation,” Arizona v. Washington, 434 U. S. 497, 503 (1978) . Juries enjoy an “unreviewable power . . . to return a verdict of not guilty for impermissible reasons,” for “the Government is precluded from appealing or otherwise upsetting such an acquittal by the Constitution’s Double Jeopardy Clause.” United States v. Powell, 469 U. S. 57, 63, 65 (1984) . The absence of appellate review of acquittals, we have cautioned, calls for guarded application of preclusion doctrine in criminal cases. See Standefer, 447 U. S., at 22–23, and n. 18. Particularly where it appears that a jury’s verdict is the result of compromise, compassion, lenity, or misunderstanding of the governing law, the Government’s inability to gain review “strongly militates against giving an acquittal [issue] preclusive effect.” Id., at 23. See also Restatement §29, Comment g, at 295 (Where circumstances suggest that an issue was resolved on erroneous considerations, “taking the prior determination at face value for purposes of the second action would [impermissibly] extend the . . . imperfections in the adjudicative process.”); id., §28, Comment j, at 283 (Issue preclusion may be denied where it is “evident from the jury’s verdict that the verdict was the result of compromise.”); Wright & Miller §4423, at 617 (same). B This case requires us to determine whether an appellate court’s vacatur of a conviction alters issue-preclusion analysis under the Double Jeopardy Clause. Three prior decisions guide our disposition. This Court first interpreted the Double Jeopardy Clause to incorporate the principle of issue preclusion in Ashe v. Swenson, 397 U. S. 436 (1970) .[2] Ashe involved a robbery of six poker players by a group of masked men. Ashe was charged with robbing one of the players, but a jury acquitted him “due to insufficient evidence.” Id., at 439. The State then tried Ashe again, this time for robbing another of the poker players. Aided by “substantially stronger” testimony from “witnesses [who] were for the most part the same,” id., at 439–440, the State secured a conviction. We held that the second prosecution violated the Double Jeopardy Clause. Because the sole issue in dispute in the first trial was whether Ashe had been one of the robbers, the jury’s acquittal verdict precluded the State from trying to convince a different jury of that very same fact in a second trial. Id., at 445. Our decision in Ashe explained that issue preclusion in criminal cases must be applied with “realism and rationality.” Id., at 444. To identify what a jury in a previous trial necessarily decided, we instructed, a court must “examine the record of a prior proceeding, taking into account the pleadings, evidence, charge, and other relevant matter.” Ibid. (quoting Mayers & Yarbrough, Bis Vexari: New Trials and Successive Prosecutions, 74 Harv. L. Rev. 1, 38 (1960)). This inquiry, we explained, “must be set in a practical frame and viewed with an eye to all the circumstances of the proceedings.” 397 U. S., at 444 (quoting Sealfon v. United States, 332 U. S. 575, 579 (1948) ). We have also made clear that “[t]he burden is on the defendant to demonstrate that the issue whose relitigation he seeks to foreclose was actually decided” by a prior jury’s verdict of acquittal. Schiro v. Farley, 510 U. S. 222, 233 (1994) (internal quotation marks omitted); accord Dowling v. United States, 493 U. S. 342, 350 (1990) . In United States v. Powell, 469 U. S. 57 , we held that a defendant cannot meet this burden when the same jury returns irreconcilably inconsistent verdicts on the question she seeks to shield from reconsideration. Powell’s starting point was our holding in Dunn v. United States, 284 U. S. 390 (1932) , that a criminal defendant may not attack a jury’s finding of guilt on one count as inconsistent with the jury’s verdict of acquittal on another count. Powell, 469 U. S., at 58–59. The Court’s opinion in Dunn stated no exceptions to this rule, and after Dunn the Court had several times “alluded to [the] rule as an established principle,” 469 U. S., at 63. Nevertheless, several Courts of Appeals had “recogniz[ed] exceptions to the rule,” id., at 62, and Powell sought an exception for the verdicts of guilt she faced. At trial, a jury had acquitted Powell of various substantive drug charges but convicted her of using a telephone in “causing and facilitating” those same offenses. Id., at 59–60. She appealed, arguing that “the verdicts were inconsistent, and that she therefore was entitled to reversal of the telephone facilitation convictions.” Id., at 60. Issue preclusion, she maintained, barred “acceptance of [the] guilty verdict[s]” on the auxiliary offenses because the same jury had acquitted her of the predicate felonies. Id., at 64. Rejecting Powell’s argument, we noted that issue preclusion is “predicated on the assumption that the jury acted rationally.” Id., at 68. When a jury returns irreconcilably inconsistent verdicts, we said, one can glean no more than that “either in the acquittal or the conviction the jury did not speak their real conclusions.” Id., at 64 (quoting Dunn, 284 U. S., at 393). Although it is impos-sible to discern which verdict the jurors arrived at rationally, we observed, “that does not show that they were not convinced of the defendant’s guilt.” Powell, 469 U. S., at 64–65 (quoting Dunn, 284 U. S., at 393). In the event of inconsistent verdicts, we pointed out, it is just as likely that “the jury, convinced of guilt, properly reached its conclusion on [one count], and then through mistake, compromise, or lenity, arrived at an inconsistent conclusion on the [related] offense.” Powell, 469 U. S., at 65. Because a court would be at a loss to know which verdict the jury “really meant,” we reasoned, principles of issue preclusion are not useful, for they are “predicated on the assumption that the jury acted rationally and found certain facts in reaching its verdict.” Id., at 68. Holding that the acquittals had no preclusive effect on the counts of conviction, we reaffirmed Dunn’s rule, under which both Powell’s convictions and her acquittals, albeit inconsistent, remained undisturbed. 469 U. S., at 69. Finally, in Yeager v. United States, 557 U. S. 110 (2009) , we clarified that Powell’s holding on inconsistent verdicts does not extend to an apparent inconsistency between a jury’s verdict of acquittal on one count and its inability to reach a verdict on another count. See 557 U. S., at 124 (“[I]nconsistent verdicts” present an “entirely different context” than one involving “both verdicts and seemingly inconsistent hung counts.”). Yeager was tried on charges of fraud and insider trading. Id., at 114. The jury acquitted him of the fraud offenses, which the Court of Appeals concluded must have reflected a finding that he “did not have any insider information that contradicted what was presented to the public.” Id., at 116. Yet the jury failed to reach a verdict on the insider-trading charges, as to which “the possession of insider information was [likewise] a critical issue of ultimate fact.” Id., at 123. Arguing that the jury had therefore acted inconsistently, the Government sought to retry Yeager on the hung counts. We ruled that retrial was barred by the Double Jeopardy Clause. A jury “speaks only through its verdict,” we noted. Id., at 121. Any number of reasons—including confusion about the issues and sheer exhaustion, we observed—could cause a jury to hang. Ibid. Accordingly, we said, only “a jury’s decisions, not its failures to decide,” identify “what a jury necessarily determined at trial.” Id., at 122. Because a hung count reveals nothing more than a jury’s failure to reach a decision, we further reasoned, it supplies no evidence of the jury’s irrationality. Id., at 124–125. Hung counts, we therefore held, “ha[ve] no place in the issue-preclusion analysis,” id., at 122: When a jury acquits on one count while failing to reach a verdict on another count concerning the same issue of ultimate fact, the acquittal, and only the acquittal, counts for preclusion purposes. Given the preclusive effect of the acquittal, the Court concluded, Yeager could not be retried on the hung count. Id., at 122–125. C With our controlling precedent in view, we turn to the inconsistent verdicts rendered in this case. The prosecution stemmed from an alleged bribe paid by petitioner Juan Bravo-Fernandez (Bravo), an entrepreneur, to petitioner Hector Martínez-Maldonado (Martínez), then a senator serving the Commonwealth of Puerto Rico. The alleged bribe took the form of an all-expenses-paid trip to Las Vegas, including a $1,000 seat at a professional boxing match featuring a popular Puerto Rican contender. United States v. Fernandez, 722 F. 3d 1, 6 (CA1 2013). According to the Government, Bravo intended the bribe to secure Martínez’ help in shepherding legislation through the Puerto Rico Senate that, if enacted, would “provid[e] substantial financial benefits” to Bravo’s enterprise. Ibid. In the leadup to the Las Vegas trip, Martínez submitted the legislation for the Senate’s consideration and issued a committee report supporting it; within a week of returning from Las Vegas, Martínez issued another favorable report and voted to enact the legislation. Id., at 6–7. Based on these events, a federal grand jury in Puerto Rico indicted petitioners for, inter alia, federal-program bribery, in violation of 18 U. S. C. §666; conspiracy to violate §666, in violation of §371; and traveling in interstate commerce to further violations of §666, in violation of the Travel Act, §1952(a)(3)(A).[3] Following a three-week trial, a jury convicted Bravo and Martínez of the standalone §666 bribery offense, but acquitted them of the related conspiracy and Travel Act charges. Fernandez, 722 F. 3d, at 7. Each received a sentence of 48 months in prison. Id., at 8. The Court of Appeals for the First Circuit vacated the §666 convictions for instructional error. Id., at 27. In the First Circuit’s view, the jury had been erroneously charged on what constitutes criminal conduct under that statute. Id., at 22–27. The charge permitted the jury to find Bravo and Martínez “guilty of offering and receiving a gratuity,” id., at 16, but, the appeals court held, §666 proscribes only quid pro quo bribes, and not gratuities, id., at 6, 22.[4] True, the court acknowledged, the jury was instructed on both theories of bribery, and the evidence at trial sufficed to support a guilty verdict on either theory. Id., at 19–20. But the Court of Appeals could not say with confidence that the erroneous charge was harmless, so it vacated the §666 convictions and remanded for further proceedings. Id., at 27, 39. On remand, relying on the issue-preclusion component of the Double Jeopardy Clause, Bravo and Martínez moved for judgments of acquittal on the standalone §666 charges. 988 F. Supp. 2d 191 (PR 2013). They could not be retried on the bribery offense, they insisted, because the jury necessarily determined that they were not guilty of violating §666 when it acquitted them of conspiring to violate §666 and traveling in interstate commerce to further violations of §666. Id., at 193. That was so, petitioners maintained, because the only contested issue at trial was whether Bravo had offered, and Martínez had accepted, a bribe within the meaning of §666. Id., at 196; see Tr.of Oral Arg. 4 (“There was no dispute that they agreed to go to a boxing match together”; nor was there any dispute “that to get to Las Vegas from Puerto Rico, you have to travel” across state lines.). The District Court denied the motions for acquittal. 988 F. Supp. 2d, at 196–198. If the sole issue disputed at trial was whether Bravo and Martínez had violated §666, the court explained, then “the jury [had] acted irrationally.” Id., at 196. Because the same jury had simultaneously convicted Bravo and Martínez on the standalone §666 charges, “the verdict simply was inconsistent.” Ibid. The First Circuit affirmed the denial of petitioners’ motions for acquittal, agreeing that the jury’s inconsistent returns were fatal to petitioners’ issue-preclusion plea. 790 F. 3d 41. The jury received the same bribery instructions for each count involving §666, the court noted, so the §666-based verdicts—convicting on the standalone bribery charges but acquitting on the related Travel Act and conspiracy counts—could not be reconciled. Id., at 54–55.[5] The Court of Appeals rejected petitioners’ argument that the eventual invalidation of the bribery convictions rendered Powell’s inconsistent-verdicts rule inapplicable. Ashe, the court reminded, calls for a practical appraisal based on the complete record of the prior proceeding; the §666 bribery convictions, like the §666-based acquittals, were part of that record. See 790 F. 3d, at 50. Nor are vacated convictions like hung counts for issue-preclusion purposes, the court continued. Informed by our decision in Yeager, the First Circuit recognized that a hung count reveals only a jury’s failure to decide, and therefore cannot evidence actual inconsistency with a jury’s decision. 790 F. 3d, at 50–51. In contrast, the court said, vacated convictions “are jury decisions, through which the jury has spoken.” Id., at 51. The later upset of a conviction on an unrelated ground, the court reasoned, does not undermine Powell’s recognition that “inconsistent verdicts make it impossible to determine what a jury necessarily decided.” 790 F. 3d, at 51. The First Circuit therefore concluded that “vacated convictions, unlike hung counts, are relevant to the Ashe [issue-preclusion] inquiry.” Ibid. We granted certiorari to resolve a conflict among courts on this question: Does the issue-preclusion component of the Double Jeopardy Clause bar the Government from retrying defendants, like Bravo and Martínez, after a jury has returned irreconcilably inconsistent verdicts of conviction and acquittal, and the convictions are later vacated for legal error unrelated to the inconsistency?[6] 557 U. S. ___ (2016). Holding that the Double Jeopardy Clause does not bar retrial in these circumstances, we affirm the First Circuit’s judgment. II When a conviction is overturned on appeal, “[t]he general rule is that the [Double Jeopardy] Clause does not bar reprosecution.” Justices of Boston Municipal Court v. Lydon, 466 U. S. 294, 308 (1984) . The ordinary consequence of vacatur, if the Government so elects, is a new trial shorn of the error that infected the first trial. This “continuing jeopardy” rule neither gives effect to the vacated judgment nor offends double jeopardy principles. Rather, it reflects the reality that the “criminal proceedings against an accused have not run their full course.” Ibid. And by permitting a new trial post vacatur, the continuing-jeopardy rule serves both society’s and criminal defendants’ interests in the fair administration of justice. “It would be a high price indeed for society to pay,” we have recognized, “were every accused granted immunity from punishment because of any defect sufficient to constitute reversible error in the proceedings leading to conviction.” United States v. Tateo, 377 U. S. 463, 466 (1964) . And the rights of criminal defendants would suffer too, for “it is at least doubtful that appellate courts would be as zealous as they now are in protecting against the effects of improprieties at the trial or pretrial stage if they knew that reversal of a conviction would put the accused irrevocably beyond the reach of further prosecution.” Ibid. Bravo and Martínez ask us to deviate from the general rule that, post vacatur of a conviction, a new trial is in order. When a conviction is vacated on appeal, they maintain, an acquittal verdict simultaneously returned should preclude the Government from retrying the defendant on the vacated count. Our precedent, harmonious with issue-preclusion doctrine, opposes the foreclosure petitioners seek. A Bravo and Martínez bear the burden of demonstrating that the jury necessarily resolved in their favor the question whether they violated §666. Schiro, 510 U. S., at 233. But, as we have explained, see supra, at 7, a defendant cannot meet that burden where the trial yielded incompatible jury verdicts on the issue the defendant seeks to insulate from relitigation. Here, the jury convicted Bravo and Martínez of violating §666 but acquitted them of conspiring, and traveling with the intent, to violate §666. The convictions and acquittals are irreconcilable because other elements of the Travel Act and conspiracy counts were not disputed. See supra, at 10–11, and n. 5. It is unknowable “which of the inconsistent verdicts—the acquittal[s] or the conviction[s]—‘the jury really meant.’ ” 790 F. 3d, at 47 (quoting Powell, 469 U. S., at 68); see Restatement §29, Comment f, at 295 (“Where a determination relied on as preclusive is itself inconsistent with some other adjudication of the same issue, . . . confidence [in that determination] is generally unwarranted.”). In view of the Government’s inability to obtain review of the acquittals, Powell, 469 U. S., at 68, the inconsistent jury findings weigh heavily against according those acquittals issue-preclusive effect. See Standefer, 447 U. S., at 23, n. 17. That petitioners’ bribery convictions were later vacated for trial error does not alter our analysis. The critical inquiry is whether the jury actually decided that Bravo and Martínez did not violate §666. Ashe counsels us to approach that task with “realism and rationality,” 397 U. S., at 444, in particular, to examine the trial record “with an eye to all the circumstances of the proceedings,” ibid. As the Court of Appeals explained, “the fact [that] the jury . . . convicted [Bravo and Martínez] of violating §666 would seem to be of quite obvious relevance” to this practical inquiry, “even though the convictions were later vacated.” 790 F. 3d, at 50. Because issue preclusion “depends on the jury’s assessment of the facts in light of the charges as presented at trial,” a conviction overturned on appeal is “appropriately considered in our assessment of [an acquittal] verdict’s preclusive effect.” United States v. Citron, 853 F. 2d 1055, 1061 (CA2 1988). Indeed, the jurors in this case might not have acquitted on the Travel Act and conspiracy counts absent their belief that the §666 bribery convictions would stand. See ibid. Bravo and Martínez could not be retried on the bribery counts, of course, if the Court of Appeals had vacated their §666 convictions because there was insufficient evidence to support those convictions. For double jeopardy purposes, a court’s evaluation of the evidence as insufficient to convict is equivalent to an acquittal and therefore bars a second prosecution for the same offense. See Burks v. United States, 437 U. S. 1 –11 (1978); cf. Powell, 469 U. S., at 67 (noting that defendants are “afforded protection against jury irrationality or error by [courts’] independent review of the sufficiency of the evidence”). But this is scarcely a case in which the prosecution “failed to muster” sufficient evidence in the first proceeding. Burks, 437 U. S., at 11. Quite the opposite. The evidence presented at petitioners’ trial, the Court of Appeals determined, supported a guilty verdict on the gratuity theory (which the First Circuit ruled impermissible) as well as the quid pro quo theory (which the First Circuit approved). 790 F. 3d, at 44. Vacatur was compelled for the sole reason that the First Circuit found the jury charge erroneous to the extent that it encompassed gratuities. See supra, at 9–10, and n. 4. Therefore, the general rule of “allowing a new trial to rectify trial error” applied. Burks, 437 U. S., at 14 (emphasis deleted). Nor, as the Government acknowledges, would retrial be tolerable if the trial error could resolve the apparent inconsistency in the jury’s verdicts. See Brief for United States 30 (If, for example, “a jury receives an erroneous instruction on the count of conviction but the correct instruction on the charge on which it acquits, the instructional error may reconcile the verdicts.”). But the instructional error here cannot account for the jury’s contradictory determinations because the error applied equally to every §666-related count. See supra, at 11. As in Powell, so in this case, “[t]he problem is that the same jury reached inconsistent results.” 469 U. S., at 68. The convictions’ later invalidation on an unrelated ground does not erase or reconcile that inconsistency: It does not bear on “the factual determinations actually and necessarily made by the jury,” nor does it “serv[e] to turn the jury’s otherwise inconsistent and irrational verdict into a consistent and rational verdict.” People v. Wilson, 496 Mich. 91, 125, 852 N. W. 2d 134, 151 (2014) (Markman, J., dissenting). Bravo and Martínez, therefore, cannot establish the factual predicate necessary to preclude the Government from retrying them on the standalone §666 charges—namely, that the jury in the first proceeding actually decided that they did not violate the federal bribery statute. B To support their argument for issue preclusion, Bravo and Martínez highlight our decision in Yeager. In Yeager, they point out, we recognized that hung counts “have never been accorded respect as a matter of law or history.” 557 U. S., at 124. That is also true of vacated convictions, they urge, so vacated convictions, like hung counts, should be excluded from the Ashe inquiry into what the jury necessarily determined. Brief for Petitioners 20–24. Asserting that we have “never held an invalid conviction . . . relevant to or evidence of anything,” Tr. of Oral Arg. 5, Bravo and Martínez argue that taking account of a vacated conviction in our issue-preclusion analysis would im-permissibly give effect to “a legal nullity,” Brief for Petitioners 39; see Wilson, 496 Mich., at 107, 852 N. W. 2d, at 142 (majority opinion) (considering a vacated count would impermissibly “bring that legally vacated conviction back to life”). This argument misapprehends the Ashe inquiry. It is undisputed that petitioners’ convictions are invalid judgments that may not be used to establish their guilt. The question is whether issue preclusion stops the Government from prosecuting them anew. On that question, Bravo and Martínez bear the burden of showing that the issue whether they violated §666 has been “determined by a valid and final judgment of acquittal.” Yeager, 557 U. S., at 119 (internal quotation marks omitted). To judge whether they carried that burden, a court must realisti-cally examine the record to identify the ground for the §666-based acquittals. Ashe, 397 U. S., at 444. A conviction that contradicts those acquittals is plainly relevant to that determination, no less so simply because it is later overturned on appeal for unrelated legal error: The split verdict—finding §666 violated on the standalone counts, but not violated on the related Travel Act and conspiracy counts—tells us that, on one count or the other, “the jury [did] not follo[w] the court’s instructions,” whether because of “mistake, compromise, or lenity.” Powell, 469 U. S., at 65; see supra, at 7. Petitioners’ acquittals therefore do not support the application of issue preclusion here.[7] Further relying on Yeager, Bravo and Martínez contend that their vacated convictions should be ignored because, as with hung counts, “there is no way to decipher” what they represent. Brief for Petitioners 28 (quoting Yeager, 557 U. S., at 121). The §666 convictions are meaningless, they maintain, because the jury was allowed to convict on the basis of conduct not criminal in the First Circuit—payment of a gratuity. Brief for Petitioners 24. This argument trips on Yeager’s reasoning. Yeager did not rest on a court’s inability to detect the basis for a jury’s decision. Rather, this Court reasoned that, when a jury hangs, there is no decision, hence no evidence of irrationality. 557 U. S., at 124–125. A verdict of guilt, by contrast, is a jury decision, even if subsequently vacated on appeal. It therefore can evince irrationality. That is the case here. Petitioners do not dispute that the Government’s evidence at trial supported a guilty verdict on the quid pro quo theory, or that the gratuity instruction held erroneous by the Court of Appeals applied to every §666-based offense. Because no rational jury could have reached conflicting verdicts on those counts, petitioners’ §666 convictions “reveal the jury’s inconsis-tency—which is the relevant issue here—even if they do not reveal which theory of liability jurors relied upon in reaching those inconsistent verdicts.” Brief for United States 31. In other words, because we do not know what the jury would have concluded had there been no instructional error, Brief for Petitioners 28–29, a new trial on the counts of conviction is in order. Bravo and Martínez have succeeded on appeal to that extent, but they are entitled to no more. The split verdict does not impede the Government from renewing the prosecution.[8] The Double Jeopardy Clause, as the First Circuit explained, forever bars the Government from again prosecuting Bravo and Martínez on the §666-based conspiracy and Travel Act offenses; “the acquittals themselves remain inviolate.” 790 F. 3d, at 51, n. 6. Bravo and Martínez have also gained “the benefit of their appellate victory,” ibid.: a second trial on the standalone bribery charges, in which the Government may not invoke a gratuity theory. But issue preclusion is not a doctrine they can commandeer when inconsistent verdicts shroud in mystery what the jury necessarily decided. * * * For the reasons stated, the judgment of the Court of Appeals for the First Circuit is Affirmed.Notes 1 The parties use the expression “collateral estoppel component,” but as this Court has observed, “issue preclusion” is the more descriptive term. Yeager v. United States, 557 U. S. 110 , n. 4 (2009); see Restatement (Second) of Judgments §27, Comment b, pp. 251–252 (1980). 2 Though we earlier recognized that res judicata (which embraces both claim and issue preclusion) applies in criminal as well as civil proceedings, we did not link the issue-preclusion inquiry to the Double Jeopardy Clause. See Sealfon v. United States, 332 U. S. 575, 578 (1948) ; Frank v. Mangum, 237 U. S. 309, 334 (1915) (The principle that “a question of fact or of law distinctly put in issue and directly determined by a court of competent jurisdiction cannot afterwards be dis-puted between the same parties” applies to “the decisions of criminal courts.”). 3 Petitioners were indicted on several other charges not relevant here. See United States v. Fernandez, 722 F. 3d 1, 7 (CA1 2013). 4 As the First Circuit acknowledged, this holding is contrary to the rulings of “most circuits to have addressed th[e] issue.” Id., at 6. Three other Federal Courts of Appeals have considered the question; each has held that §666 prohibits gratuities as well as quid pro quo bribes. See United States v. Bahel, 662 F. 3d 610, 636 (CA2 2011); United States v. Hawkins, 777 F. 3d 880, 881 (CA7 2015); United States v. Zimmerman, 509 F. 3d 920, 927 (CA8 2007). 5 As just observed, see supra, at 10, petitioners urge that §666 bribery was the sole issue in controversy, and that there was no dispute on other elements of the Travel Act and conspiracy counts. See Tr. of Oral Arg. 4. See also Brief for United States 13 (accepting that the jury “returned irreconcilably inconsistent verdicts”). If another element could explain the acquittals, then there would be no inconsistency and no argument against a new trial on bribery. See infra, at 12–13. 6 Compare United States v. Citron, 853 F. 2d 1055, 1058–1061 (CA2 1988) (holding that retrial does not violate Double Jeopardy Clause under these circumstances); United States v. Price, 750 F. 2d 363, 366 (CA5 1985) (same); Evans v. United States, 987 A. 2d 1138, 1141–1142 (D. C. 2010) (same); and State v. Kelly, 201 N. J. 471, 493–494, 992 A. 2d 776, 789 (2010) (same), with People v. Wilson, 496 Mich. 91, 105–107, 852 N. W. 2d 134, 141–142 (2014) (holding that Double Jeopardy Clause bars retrial in this situation). As the First Circuit explained, “[a]lthough Citron and Price predate Yeager, both the Second and Fifth Circuits decided that vacated counts are relevant to the Ashe analysis at a time when those circuits had already ruled that hung counts should be disregarded for purposes of the Ashe inquiry.” 790 F. 3d 41, 51, n. 7 (2015) (citing United States v. Mespoulede, 597 F. 2d 329, 332, 335–336 (CA2 1979); United States v. Nelson, 599 F. 2d 714, 716–717 (CA5 1979)). The Second Circuit, moreover, has adhered to Citron since Yeager. See United States v. Bruno, 531 Fed. Appx. 47, 49 (2013). 7 Nor is this the first time we have looked to a vacated conviction to ascertain what a jury decided in a prior proceeding. Our holding in Morris v. Mathews, 475 U. S. 237 (1986) , that a conviction vacated on double jeopardy grounds may be “reduced to a conviction for a lesser included offense which is not jeopardy barred,” id., at 246–247, rested on exactly that rationale. See id., at 247 (relying on a jeopardy-barred vacated conviction for aggravated murder to conclude that the jury “necessarily found that the defendant’s conduct satisfie[d] the elements of the lesser included offense” of simple murder). 8 A number of lower courts have reached the same conclusion. See Citron, 853 F. 2d, at 1059 (If the defendant “was convicted of the offense that is the subject of the retrial,” the case is materially different from one with “an acquittal accompanied by a failure to reach a verdict.”); Price, 750 F. 2d, at 366 (a case in which “the jury returned no verdict of conviction” on the compound count, “but only a verdict of acquittal on the substantive count,” is not instructive on whether the Government may retry a defendant after an inconsistent verdict has been vacated); Evans, 987 A. 2d, at 1142 (“Yeager does nothing to undermine” the conclusion that a defendant may be retried after an inconsistent verdict is overturned.); Kelly, 201 N. J., at 494, 992 A. 2d, at 789 (explaining in the context of retrial following vacatur that “Yeager has no application to a case . . . involving an inconsistent verdict of acquittals and convictions returned by the same jury”).

582.US.2016_16-466

A group of plaintiffs, most of whom are not California residents, sued Bristol-Myers Squibb Company (BMS) in California state court, alleging that the pharmaceutical company’s drug Plavix had damaged their health. BMS is incorporated in Delaware and headquartered in New York, and it maintains substantial operations in both New York and New Jersey. Although it engages in business activities in California and sells Plavix there, BMS did not develop, create a marketing strategy for, manufacture, label, package, or work on the regulatory approval for Plavix in the State. And the nonresident plaintiffs did not allege that they obtained Plavix from a California source, that they were injured by Plavix in California, or that they were treated for their injuries in California. The California Superior Court denied BMS’s motion to quash service of summons on the nonresidents’ claims for lack of personal jurisdiction, concluding that BMS’s extensive activities in the State gave the California courts general jurisdiction. Following this Court’s decision in Daimler AG v. Bauman, 571 U. S. ___, the State Court of Appeal found that the California courts lacked general jurisdiction. But the Court of Appeal went on to find that the California courts had specific jurisdiction over the claims brought by the nonresident plaintiffs. Affirming, the State Supreme Court applied a “sliding scale approach” to specific jurisdiction, concluding that BMS’s “wide ranging” contacts with the State were enough to support a finding of specific jurisdiction over the claims brought by the nonresident plaintiffs. That attenuated connection was met, the court held, in part because the nonresidents’ claims were similar in many ways to the California residents’ claims and because BMS engaged in other activities in the State. Held: California courts lack specific jurisdiction to entertain the nonresidents’ claims. Pp. 4–12. (a) The personal jurisdiction of state courts is “subject to review for compatibility with the Fourteenth Amendment’s Due Process Clause.” Goodyear Dunlop Tires Operations, S. A. v. Brown, 564 U. S. 915 . This Court’s decisions have recognized two types of personal jurisdiction: general and specific. For general jurisdiction, the “paradigm forum” is an “individual’s domicile,” or, for corporations, “an equivalent place, one in which the corporation is fairly regarded as at home.” Id., at 924. Specific jurisdiction, however, requires “the suit” to “aris[e] out of or relat[e] to the defendant’s contacts with the forum.” Daimler, supra, at ___ (internal quotation marks omitted). The “primary concern” in assessing personal jurisdiction is “the burden on the defendant.” World-Wide Volkswagen Corp. v. Woodson, 444 U. S. 286 . Assessing this burden obviously requires a court to consider the practical problems resulting from litigating in the forum, but it also encompasses the more abstract matter of submitting to the coercive power of a State that may have little legitimate interest in the claims in question. At times, “the Due Process Clause, acting as an instrument of interstate federalism, may . . . divest the State of its power to render a valid judgment.” Id., at 294. Pp. 4–7. (b) Settled principles of specific jurisdiction control this case. For a court to exercise specific jurisdiction over a claim there must be an “affiliation between the forum and the underlying controversy, principally, [an] activity or an occurrence that takes place in the forum State.” Goodyear, supra, at 919 (internal quotation marks and brackets omitted). When no such connection exists, specific jurisdiction is lacking regardless of the extent of a defendant’s unconnected activities in the State. The California Supreme Court’s “sliding scale approach”—which resembles a loose and spurious form of general jurisdiction—is thus difficult to square with this Court’s precedents. That court found specific jurisdiction without identifying any adequate link between the State and the nonresidents’ claims. The mere fact that other plaintiffs were prescribed, obtained, and ingested Plavix in California does not allow the State to assert specific jurisdiction over the nonresidents’ claims. Nor is it sufficient (or relevant) that BMS conducted research in California on matters unrelated to Plavix. What is needed is a connection between the forum and the specific claims at issue. Cf. Walden v. Fiore, 571 U. S. ___. Pp. 7–9. (c) The nonresident plaintiffs’ reliance on Keeton v. Hustler Magazine, Inc., 465 U. S. 770 , and Phillips Petroleum Co. v. Shutts, 472 U. S. 797 , is misplaced. Keeton concerned jurisdiction to determine the scope of a claim involving in-state injury and injury to residents of the State, not, as here, jurisdiction to entertain claims involving no in-state injury and no injury to residents of the forum State. And Shutts, which concerned the due process rights of plaintiffs, has no bearing on the question presented here. Pp. 9–11. (d) BMS’s decision to contract with McKesson, a California company, to distribute Plavix nationally does not provide a sufficient basis for personal jurisdiction. It is not alleged that BMS engaged in relevant acts together with McKesson in California or that BMS is derivatively liable for McKesson’s conduct in California. The bare fact that BMS contracted with a California distributor is not enough to establish personal jurisdiction in the State. Pp. 11–12. (e) The Court’s decision will not result in the parade of horribles that respondents conjure up. It does not prevent the California and out-of-state plaintiffs from joining together in a consolidated action in the States that have general jurisdiction over BMS. Alternatively, the nonresident plaintiffs could probably sue together in their respective home States. In addition, since this decision concerns the due process limits on the exercise of specific jurisdiction by a State, the question remains open whether the Fifth Amendment imposes the same restrictions on the exercise of personal jurisdiction by a federal court. P. 12. 1 Cal. 5th 783, 377 P. 3d 874, reversed and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, Breyer, Kagan, and Gorsuch, JJ., joined. Sotomayor, J., filed a dissenting opinion.

More than 600 plaintiffs, most of whom are not California residents, filed this civil action in a California state court against Bristol-Myers Squibb Company (BMS), asserting a variety of state-law claims based on injuries allegedly caused by a BMS drug called Plavix. The California Supreme Court held that the California courts have specific jurisdiction to entertain the nonresidents’ claims. We now reverse. I A BMS, a large pharmaceutical company, is incorporated in Delaware and headquartered in New York, and it maintains substantial operations in both New York and New Jersey. 1 Cal. 5th 783, 790, 377 P. 3d 874, 879 (2016). Over 50 percent of BMS’s work force in the United States is employed in those two States. Ibid. BMS also engages in business activities in other jurisdictions, including California. Five of the company’s research and laboratory facilities, which employ a total of around 160 employees, are located there. Ibid. BMS also employs about 250 sales representatives in California and maintains a small state-government advocacy office in Sacramento. Ibid. One of the pharmaceuticals that BMS manufactures and sells is Plavix, a prescription drug that thins the blood and inhibits blood clotting. BMS did not develop Plavix in California, did not create a marketing strategy for Plavix in California, and did not manufacture, label, package, or work on the regulatory approval of the product in California. Ibid. BMS instead engaged in all of these activities in either New York or New Jersey. Ibid. But BMS does sell Plavix in California. Between 2006 and 2012, it sold almost 187 million Plavix pills in the State and took in more than $900 million from those sales. 1 Cal. 5th, at 790–791, 377 P. 3d, at 879. This amounts to a little over one percent of the company’s nationwide sales revenue. Id., at 790, 377 P. 3d, at 879. B A group of plaintiffs—consisting of 86 California residents and 592 residents from 33 other States—filed eight separate complaints in California Superior Court, alleging that Plavix had damaged their health. Id., at 789, 377 P. 3d, at 878. All the complaints asserted 13 claims under California law, including products liability, negligent misrepresentation, and misleading advertising claims. Ibid. The nonresident plaintiffs did not allege that they obtained Plavix through California physicians or from any other California source; nor did they claim that they were injured by Plavix or were treated for their injuries in California. Asserting lack of personal jurisdiction, BMS moved to quash service of summons on the nonresidents’ claims, but the California Superior Court denied this motion, finding that the California courts had general jurisdiction over BMS “[b]ecause [it] engages in extensive activities in California.” App. to Pet. for Cert. 150. BMS unsuccess-fully petitioned the State Court of Appeal for a writ of mandate, but after our decision on general jurisdiction in Daimler AG v. Bauman, 571 U. S. ___ (2014), the California Supreme Court instructed the Court of Appeal “to vacate its order denying mandate and to issue an order to show cause why relief sought in the petition should not be granted.” App. 9–10. The Court of Appeal then changed its decision on the question of general jurisdiction. 228 Cal. App. 4th 605, 175 Cal. Rptr. 3d 412 (2014). Under Daimler, it held, general jurisdiction was clearly lacking, but it went on to find that the California courts had specific jurisdiction over the nonresidents’ claims against BMS. 228 Cal. App. 4th 605, 175 Cal. Rptr. 3d, at 425–439. The California Supreme Court affirmed. The court unanimously agreed with the Court of Appeal on the issue of general jurisdiction, but the court was divided on the question of specific jurisdiction. The majority applied a “sliding scale approach to specific jurisdiction.” 1 Cal. 5th, at 806, 377 P. 3d, at 889. Under this approach, “the more wide ranging the defendant’s forum contacts, the more readily is shown a connection between the forum contacts and the claim.” Ibid. (internal quotation marks omitted). Applying this test, the majority concluded that “BMS’s extensive contacts with California” permitted the exercise of specific jurisdiction “based on a less direct connection between BMS’s forum activities and plaintiffs’ claims than might otherwise be required.” Ibid. This attenuated requirement was met, the majority found, because the claims of the nonresidents were similar in several ways to the claims of the California residents (as to which specific jurisdiction was uncontested). Id., at 803–806, 377 P. 3d, at 887–889. The court noted that “[b]oth the resident and nonresident plaintiffs’ claims are based on the same allegedly defective product and the assertedly misleading marketing and promotion of that product.” Id., at 804, 377 P. 3d, at 888. And while acknowledging that “there is no claim that Plavix itself was designed and developed in [BMS’s California research facilities],” the court thought it significant that other research was done in the State. Ibid. Three justices dissented. “The claims of . . . nonresidents injured by their use of Plavix they purchased and used in other states,” they wrote, “in no sense arise from BMS’s marketing and sales of Plavix in California,” and they found that the “mere similarity” of the residents’ and nonresidents’ claims was not enough. Id., at 819, 377 P. 3d, at 898 (opinion of Werdegar, J.). The dissent accused the majority of “expand[ing] specific jurisdiction to the point that, for a large category of defendants, it becomes indistinguishable from general jurisdiction.” Id., at 816, 377 P. 3d, at 896. We granted certiorari to decide whether the California courts’ exercise of jurisdiction in this case violates the Due Process Clause of the Fourteenth Amendment. 580 U. S. ___ (2017).[1] II A It has long been established that the Fourteenth Amendment limits the personal jurisdiction of state courts. See, e.g., Daimler, supra, at ___–___ (slip op., at 6–13); World-Wide Volkswagen Corp. v. Woodson, 444 U. S. 286, 291 (1980) ; International Shoe Co. v. Washington, 326 U. S. 310 –317 (1945); Pennoyer v. Neff, 95 U. S. 714, 733 (1878). Because “[a] state court’s assertion of jurisdiction exposes defendants to the State’s coercive power,” it is “subject to review for compatibility with the Fourteenth Amendment’s Due Process Clause,” Goodyear Dunlop Tires Operations, S. A. v. Brown, 564 U. S. 915, 918 (2011) , which “limits the power of a state court to render a valid personal judgment against a nonresident defendant,” World-Wide Volkswagen, supra, at 291. The primary focus of our personal jurisdiction inquiry is the defendant’s relationship to the forum State. See Walden v. Fiore, 571 U. S. ___, ___–___ (2014) (slip op., at 5–8); Phillips Petroleum Co. v. Shutts, 472 U. S. 797 –807 (1985). Since our seminal decision in International Shoe, our decisions have recognized two types of personal jurisdiction: “general” (sometimes called “all-purpose”) jurisdiction and “specific” (sometimes called “case-linked”) jurisdiction. Goodyear, 564 U. S., at 919. “For an individual, the paradigm forum for the exercise of general jurisdiction is the individual’s domicile; for a corporation, it is an equivalent place, one in which the corporation is fairly regarded as at home.” Id., at 924. A court with general jurisdiction may hear any claim against that defendant, even if all the incidents underlying the claim occurred in a different State. Id., at 919. But “only a limited set of affiliations with a forum will render a defendant amenable to” general jurisdiction in that State. Daimler, 571 U. S., at ___ (slip op., at 18). Specific jurisdiction is very different. In order for a state court to exercise specific jurisdiction, “the suit” must “aris[e] out of or relat[e] to the defendant’s contacts with the forum.” Id., at ___ (slip op., at 8) (internal quotation marks omitted; emphasis added); see Burger King Corp. v. Rudzewicz, 471 U. S. 462 –473 (1985); Helicopteros Nacionales de Colombia, S. A. v. Hall, 466 U. S. 408, 414 (1984) . In other words, there must be “an affiliation between the forum and the underlying controversy, principally, [an] activity or an occurrence that takes place in the forum State and is therefore subject to the State’s regulation.” Goodyear, 564 U. S., at 919 (internal quotation marks and brackets omitted). For this reason, “specific jurisdiction is confined to adjudication of issues deriv-ing from, or connected with, the very controversy that establishes jurisdiction.” Ibid. (internal quotation marks omitted). B In determining whether personal jurisdiction is present, a court must consider a variety of interests. These include “the interests of the forum State and of the plaintiff in proceeding with the cause in the plaintiff’s forum of choice.” Kulko v. Superior Court of Cal., City and County of San Francisco, 436 U. S. 84, 92 (1978) ; see Daimler, supra, at ___–___, n. 20 (slip op., at 21–22, n. 20); Asahi Metal Industry Co. v. Superior Court of Cal., Solano Cty., 480 U. S. 102, 113 (1987) ; World-Wide Volkswagen, 444 U. S., at 292. But the “primary concern” is “the burden on the defendant.” Id., at 292. Assessing this burden obviously requires a court to consider the practical problems resulting from litigating in the forum, but it also encompasses the more abstract matter of submitting to the coercive power of a State that may have little legitimate interest in the claims in question. As we have put it, restrictions on personal jurisdiction “are more than a guarantee of immunity from inconvenient or distant litigation. They are a consequence of territorial limitations on the power of the respective States.” Hanson v. Denckla, 357 U. S. 235, 251 (1958) . “[T]he States retain many essential attributes of sovereignty, including, in particular, the sovereign power to try causes in their courts. The sovereignty of each State . . . implie[s] a limitation on the sovereignty of all its sister States.” World-Wide Volkswagen, 444 U. S., at 293. And at times, this federalism interest may be decisive. As we explained in World-Wide Volkswagen, “[e]ven if the defendant would suffer minimal or no inconvenience from being forced to litigate before the tribunals of another State; even if the forum State has a strong interest in applying its law to the controversy; even if the forum State is the most convenient location for litigation, the Due Process Clause, acting as an instrument of interstate federalism, may sometimes act to divest the State of its power to render a valid judgment.” Id., at 294. III A Our settled principles regarding specific jurisdiction control this case. In order for a court to exercise specific jurisdiction over a claim, there must be an “affiliation between the forum and the underlying controversy, principally, [an] activity or an occurrence that takes place in the forum State.” Goodyear, 564 U. S., at 919 (internal quotation marks and brackets in original omitted). When there is no such connection, specific jurisdiction is lacking regardless of the extent of a defendant’s unconnected activities in the State. See id., at 931, n. 6 (“[E]ven regularly occurring sales of a product in a State do not justify the exercise of jurisdiction over a claim unrelated to those sales”). For this reason, the California Supreme Court’s “sliding scale approach” is difficult to square with our precedents. Under the California approach, the strength of the requisite connection between the forum and the specific claims at issue is relaxed if the defendant has extensive forum contacts that are unrelated to those claims. Our cases provide no support for this approach, which resembles a loose and spurious form of general jurisdiction. For spe-cific jurisdiction, a defendant’s general connections with the forum are not enough. As we have said, “[a] corporation’s ‘continuous activity of some sorts within a state . . . is not enough to support the demand that the corporation be amenable to suits unrelated to that activity.’ ” Id., at 927 (quoting International Shoe, 326 U. S., at 318). The present case illustrates the danger of the California approach. The State Supreme Court found that specific jurisdiction was present without identifying any adequate link between the State and the nonresidents’ claims. As noted, the nonresidents were not prescribed Plavix in California, did not purchase Plavix in California, did not ingest Plavix in California, and were not injured by Plavix in California. The mere fact that other plaintiffs were prescribed, obtained, and ingested Plavix in California—and allegedly sustained the same injuries as did the nonresidents—does not allow the State to assert specific jurisdiction over the nonresidents’ claims. As we have explained, “a defendant’s relationship with a . . . third party, standing alone, is an insufficient basis for jurisdiction.” Walden, 571 U. S., at ___ (slip op., at 8). This remains true even when third parties (here, the plaintiffs who reside in California) can bring claims similar to those brought by the nonresidents. Nor is it sufficient—or even relevant—that BMS conducted research in California on matters unrelated to Plavix. What is needed—and what is missing here—is a connection between the forum and the specific claims at issue. Our decision in Walden, supra, illustrates this requirement. In that case, Nevada plaintiffs sued an out-of-state defendant for conducting an allegedly unlawful search of the plaintiffs while they were in Georgia preparing to board a plane bound for Nevada. We held that the Nevada courts lacked specific jurisdiction even though the plaintiffs were Nevada residents and “suffered foreseeable harm in Nevada.” Id., at ___ (slip op., at 11). Because the “relevant conduct occurred entirely in Georgi[a] . . . the mere fact that [this] conduct affected plaintiffs with connections to the forum State d[id] not suffice to authorize jurisdiction.” Id., at ___ (slip op., at 14) (emphasis added). In today’s case, the connection between the nonresidents’ claims and the forum is even weaker. The relevant plaintiffs are not California residents and do not claim to have suffered harm in that State. In addition, as in Walden, all the conduct giving rise to the nonresidents’ claims occurred elsewhere. It follows that the California courts cannot claim specific jurisdiction. See World-Wide Volkswagen, supra, at 295 (finding no personal jurisdiction in Oklahoma because the defendant “carr[ied] on no activ-ity whatsoever in Oklahoma” and dismissing “the fortuitous circumstance that a single Audi automobile, sold [by defendants] in New York to New York residents, happened to suffer an accident while passing through Oklahoma” as an “isolated occurrence”). B The nonresidents maintain that two of our cases sup-port the decision below, but they misinterpret those precedents. In Keeton v. Hustler Magazine, Inc., 465 U. S. 770 (1984) , a New York resident sued Hustler in New Hampshire, claiming that she had been libeled in five issues of the magazine, which was distributed throughout the country, including in New Hampshire, where it sold 10,000 to 15,000 copies per month. Concluding that specific jurisdiction was present, we relied principally on the connection between the circulation of the magazine in New Hampshire and damage allegedly caused within the State. We noted that “[f]alse statements of fact harm both the subject of the falsehood and the readers of the statement.” Id., at 776 (emphasis deleted). This factor amply distinguishes Keeton from the present case, for here the nonresidents’ claims involve no harm in California and no harm to California residents. The nonresident plaintiffs in this case point to our holding in Keeton that there was jurisdiction in New Hampshire to entertain the plaintiff’s request for damages suffered outside the State, id., at 774, but that holding concerned jurisdiction to determine the scope of a claim involving in-state injury and injury to residents of the State, not, as in this case, jurisdiction to entertain claims involving no in-state injury and no injury to residents of the forum State. Keeton held that there was jurisdiction in New Hampshire to consider the full measure of the plaintiff’s claim, but whether she could actually recover out-of-state damages was a merits question governed by New Hampshire libel law. Id., at 778, n. 9. The Court’s decision in Phillips Petroleum Co. v. Shutts, 472 U. S. 797 (1985) , which involved a class action filed in Kansas, is even less relevant. The Kansas court exercised personal jurisdiction over the claims of nonresident class members, and the defendant, Phillips Petroleum, argued that this violated the due process rights of these class members because they lacked minimum contacts with the State.[2] According to the defendant, the out-of-state class members should not have been kept in the case unless they affirmatively opted in, instead of merely failing to opt out after receiving notice. Id., at 812. Holding that there had been no due process violation, the Court explained that the authority of a State to entertain the claims of nonresident class members is entirely different from its authority to exercise jurisdiction over an out-of-state defendant. Id., at 808–812. Since Shutts concerned the due process rights of plaintiffs, it has no bearing on the question presented here. Respondents nevertheless contend that Shutts supports their position because, in their words, it would be “absurd to believe that [this Court] would have reached the exact opposite result if the petitioner [Phillips] had only invoked its own due-process rights, rather than those of the non-resident plaintiffs.” Brief for Respondents 28–29, n. 6 (emphasis deleted). But the fact remains that Phillips did not assert that Kansas improperly exercised personal jurisdiction over it, and the Court did not address that issue.[3] Indeed, the Court stated specifically that its “discussion of personal jurisdiction [did not] address class actions where the jurisdiction is asserted against a defendant class.” Shutts, supra, at 812, n. 3. C In a last ditch contention, respondents contend that BMS’s “decision to contract with a California company [McKesson] to distribute [Plavix] nationally” provides a sufficient basis for personal jurisdiction. Tr. of Oral Arg. 32. But as we have explained, “[t]he requirements of International Shoe . . . must be met as to each defendant over whom a state court exercises jurisdiction.” Rush v. Savchuk, 444 U. S. 320, 332 (1980) ; see Walden, 571 U. S., at ___ (slip op, at 8) (“[A] defendant’s relationship with a . . . third party, standing alone, is an insufficient basis for jurisdiction”). In this case, it is not alleged that BMS engaged in relevant acts together with McKesson in California. Nor is it alleged that BMS is derivatively liable for McKesson’s conduct in California. And the nonresidents “have adduced no evidence to show how or by whom the Plavix they took was distributed to the pharmacies that dispensed it to them.” 1 Cal. 5th, at 815, 377 P. 3d, at 895 (Werdegar, J., dissenting) (emphasis deleted). See Tr. of Oral Arg. 33 (“It is impossible to trace a particular pill to a particular person . . . . It’s not possible for us to track particularly to McKesson”). The bare fact that BMS contracted with a California distributor is not enough to establish personal jurisdiction in the State. IV Our straightforward application in this case of settled principles of personal jurisdiction will not result in the parade of horribles that respondents conjure up. See Brief for Respondents 38–47. Our decision does not prevent the California and out-of-state plaintiffs from joining together in a consolidated action in the States that have general jurisdiction over BMS. BMS concedes that such suits could be brought in either New York or Delaware. See Brief for Petitioner 13. Alternatively, the plaintiffs who are residents of a particular State—for example, the 92 plaintiffs from Texas and the 71 from Ohio—could probably sue together in their home States. In addition, since our decision concerns the due process limits on the exercise of specific jurisdiction by a State, we leave open the question whether the Fifth Amendment imposes the same restrictions on the exercise of personal jurisdiction by a federal court. See Omni Capital Int’l, Ltd. v. Rudolf Wolff & Co., 484 U. S. 97 , n. 5 (1987). * * * The judgment of the California Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 California law provides that its courts may exercise jurisdiction “on any basis not inconsistent with the Constitution . . . of the United States,” Cal. Civ. Proc. Code Ann. §410.10 (West 2004); see Daimler AG v. Bauman, 571 U. S. ___, ___ (2014) (slip op., at 6). 2 The Court held that the defendant had standing to argue that the Kansas court had improperly exercised personal jurisdiction over the claims of the out-of-state class members because that holding materially affected the defendant’s own interests, specifically, the res judicata effect of an adverse judgment. 472 U. S., at 803–806. 3 Petitioner speculates that Phillips did not invoke its own due process rights because it was believed at the time that the Kansas court had general jurisdiction. See Reply Brief 7, n. 1.

582.US.2016_16-373

Section 11 of the Securities Act of 1933 gives purchasers of securities “a right of action against an issuer or designated individuals,” including securities underwriters, for any material misstatements or omissions in a registration statement. Omnicare, Inc. v. Laborers Dist. Council Constr. Industry Pension Fund, 575 U. S. ___, ___; see 15 U. S. C. §77k(a). Section 13 provides two time limits for §11 suits. The first sentence states that an action “must be brought within one year after the discovery of the untrue statement or the omission, or after such discovery should have been made by the exercise of reasonable diligence . . . .” The second sentence provides that “[i]n no event shall any such action be brought . . . more than three years after the security was bona fide offered to the public . . . .” §77m. In 2007 and 2008, Lehman Brothers Holdings Inc. raised capital through several public securities offerings. Petitioner, the largest public pension fund in the country, purchased some of those securities; and it is alleged that respondents, various financial firms, are liable under the Act for their participation as underwriters in the transactions. In 2008, a putative class action was filed against respondents in the Southern District of New York. The complaint raised §11 claims, alleging that the registration statements for certain of Lehman’s 2007 and 2008 securities offerings included material misstatements or omissions. Because the complaint was filed on behalf of all persons who purchased the identified securities, petitioner was a member of the putative class. In February 2011, more than three years after the relevant securities offerings, petitioner filed a separate complaint against respondents in the Northern District of California, alleging violations identical to those in the class action on petitioner’s own behalf. Soon thereafter, a proposed settlement was reached in the putative class action, but petitioner opted out of the class. Respondents then moved to dismiss petitioner’s individual suit, alleging that the §11 violations were untimely under the 3-year bar in the second sentence of §13. Petitioner countered that the 3-year period was tolled during the pendency of the class-action filing, relying on American Pipe & Construction Co. v. Utah, 414 U. S. 538 . The trial court disagreed, and the Second Circuit affirmed, holding that American Pipe’s tolling principle is inapplicable to the 3-year bar. It also rejected petitioner’s alternative argument that the timely filing of the class action made petitioner’s individual claims timely as well. Held: Petitioner’s untimely filing of its individual complaint more than three years after the relevant securities offering is ground for dismissal. Pp. 4–17. (a) Section 13’s 3-year time limit is a statute of repose not subject to equitable tolling. Pp. 4–14. (1) The two categories of statutory time bars—statutes of limitations and statutes of repose—each have “a distinct purpose.” CTS Corp. v. Waldburger, 573 U. S. ___, ___. Statutes of limitations are designed to encourage plaintiffs “ ‘to pursue diligent prosecution of known claims,’ ” id., at ___, while statutes of repose “effect a legislative judgment that a defendant should ‘be free from liability after the legislatively determined period of time,’ ” id., at ___. For this reason, statutes of limitations begin to run “when the cause of action accrues,” while statutes of repose begin to run on “the date of the last culpable act or omission of the defendant.” Id., at ___. From the structure of §13, and the language of its second sentence, it is evident that the 3-year bar is a statute of repose. The instruction that “[i]n no event” shall an action be brought more than three years after the relevant securities offering admits of no exception. The statute also runs from the defendant’s last culpable act (the securities offering), not from the accrual of the claim (the plaintiff’s discovery of the defect). This view is confirmed by §13’s two-sentence structure. The pairing of a shorter statute of limitations and a longer statute of repose is a common feature of statutory time limits. See, e.g., Gabelli v. SEC, 568 U. S. 442 . Section 13’s history also supports the classification. The 1933 Securities Act’s original 2-year discovery period and 10-year outside limit were shortened a year later. The evident design of the shortened period was to protect defendants’ financial security by reducing the open period for potential liability. Pp. 4–7. (2) The determination that the 3-year period is a statute of repose is critical here, for the question whether a tolling rule applies to a given statutory time bar is one “of statutory intent.” Lozano v. Montoya Alvarez, 572 U. S. 1 , ___. In light of the purpose of a statute of repose, the provision is in general not subject to tolling. Tolling is permissible only where there is a particular indication that the legislature did not intend the statute to provide complete repose but instead anticipated the extension of the statutory period under certain circumstances. A statute of repose implements a “ ‘legislative decisio[n] that . . . there should be a specific time beyond which a defendant should no longer be subjected to protracted liability.’ ” CTS, 573 U. S., at ___. The unqualified nature of that determination supersedes the courts’ residual authority and forecloses the extension of the statutory period based on equitable principles. Thus, the Court repeatedly has stated that statutes of repose are not subject to equitable tolling. See, e.g., id., at ___–___. Pp. 7–8. (3) The tolling decision in American Pipe derived from equity principles and therefore cannot alter the unconditional language and purpose of the 3-year statute of repose. The source of the tolling rule applied in American Pipe is the judicial power to promote equity, not the power to interpret and enforce statutory provisions. Nothing in the decision suggests that its tolling rule was mandated by a statute or federal rule. Moreover, the Court relied on cases that are paradigm applications of equitable tolling principles, see 414 U. S., at 559. Thus, the Court has previously referred to American Pipe as “equitable tolling.” See, e.g., Irwin v. Department of Veterans Affairs, 498 U. S. 89 , and n. 3. Pp. 8–11. (4) Petitioner’s counterarguments are unpersuasive. First, petitioner contends that this case is indistinguishable from American Pipe, but the statute there was one of limitations, which may be tolled by equitable considerations even where a statute of repose may not. Second, petitioner argues that the timely filing of a class-action complaint fulfills the purposes of a statutory time limit with regard to later filed suits by individual members of the class. But by permitting a class action to splinter into individual suits, the application of American Pipe tolling here would threaten to alter and expand a defendant’s accountability, contradicting the substance of a statute of repose. Third, petitioner contends that dismissal of its individual suit as untimely would eviscerate its ability to opt out, but it does not follow from any privilege to opt out that an ensuing suit can be filed without regard to mandatory time limits. Fourth, petitioner argues that declining to apply American Pipe tolling to statutes of repose will create inefficiencies, but this Court “lack[s] the authority to rewrite” the statute of repose or to ignore its plain import. Baker Botts L. L. P. v. ASARCO LLC, 576 U. S. ___, ___. And petitioner’s practical concerns likely are overstated. Pp. 11–14. (b) Also unpersuasive is petitioner’s alternative argument: that §13’s requirement that an “action” be “brought” within three years of the relevant securities offering is met here because the filing of the class-action complaint “brought” petitioner’s individual “action” within the statutory time period. This argument presumes that an “action” is “brought” when substantive claims are presented to any court, rather than when a particular complaint is filed in a particular court. The term “action,” however, refers to a judicial “proceeding,” or perhaps a “suit”—not to the general content of claims. Taken to its logical limit, petitioner’s argument would make an individual action timely even if it were filed decades after the original securities offering—provided a class-action complaint had been filed within the initial 3-year period. Congress would not have intended this result. This argument is also inconsistent with the reasoning in American Pipe itself. If the filing of a class action made all subsequent actions by putative class members timely, there would be no need for tolling at all. Pp. 14–15. (c) The final analysis is straightforward. Because §13’s 3-year time bar is a statute of repose, it displaces the traditional power of courts to modify statutory time limits in the name of equity. And because the American Pipe tolling rule is rooted in those equitable powers, it cannot extend the 3-year period. Petitioner’s untimely filing of its individual action is thus ground for dismissal. Pp. 16–17. 655 Fed. Appx. 13, affirmed. Kennedy, J., delivered the opinion of the Court, in which Roberts, C. J., and Thomas, Alito, and Gorsuch, JJ., joined. Ginsburg, J., filed a dissenting opinion, in which Breyer, Sotomayor, and Kagan, JJ., joined.

The suit giving rise to the case before the Court was filed by a plaintiff who was a member of a putative class in a class action but who later elected to withdraw and proceed in this separate suit, seeking recovery for the same illegalities that were alleged in the class suit. The class-action suit had been filed within the time permitted by statute. Whether the later, separate suit was also timely is the controlling question. I A The Securities Act of 1933 “protects investors by ensuring that companies issuing securities . . . make a ‘full and fair disclosure of information’ relevant to a public offering.” Omnicare, Inc. v. Laborers Dist. Council Constr. Industry Pension Fund, 575 U. S. ___, ___ (2015) (slip op., at 1) (quoting Pinter v. Dahl, 486 U. S. 622, 646 (1988) ); see 48Stat. 74, as amended, 15 U. S. C. §77a et seq. Companies may offer securities to the public only after filing a registration statement, which must contain information about the company and the security for sale. Omnicare, 575 U. S., at ___–___ (slip op., at 1–2). Section 11 of the Securities Act “promotes compliance with these disclosure provisions by giving purchasers a right of action against an issuer or designated individuals,” including securities underwriters, for any material misstatements or omissions in a registration statement. Id., at ___ (slip op., at 2); see 15 U. S. C. §77k(a). The Act provides time limits for §11 suits. These time limits are set forth in a two-sentence section of the Act, §13. It provides as follows: “No action shall be maintained to enforce any liability created under [§11] unless brought within one year after the discovery of the untrue statement or the omission, or after such discovery should have been made by the exercise of reasonable diligence . . . . In no event shall any such action be brought to enforce a liability created under [§11] more than three yearsafter the security was bona fide offered to thepublic . . . .” 15 U. S. C. §77m. So there are two time bars in the quoted provision; and the second one, the 3-year bar, is central to this case. B Lehman Brothers Holdings Inc. formerly was one of the largest investment banks in the United States. In 2007 and 2008, Lehman raised capital through a number of public securities offerings. Petitioner, California Public Employees’ Retirement System (sometimes called CalPERS), is the largest public pension fund in the country. Petitioner purchased securities in some of these Lehman offerings; and it is alleged that respondents, various financial firms, are liable under the Act for their participation as underwriters in the transactions. The separate respondents are listed in an appendix to this opinion. In September 2008, Lehman filed for bankruptcy. Around the same time, a putative class action concerning Lehman securities was filed against respondents in the United States District Court for the Southern District of New York. The operative complaint raised claims under §11, alleging that the registration statements for certain of Lehman’s 2007 and 2008 securities offerings included material misstatements or omissions. The complaint was filed on behalf of all persons who purchased the identified securities, making petitioner a member of the putative class. Petitioner, however, was not one of the named plaintiffs in the suit. The class action was consolidated with other securities suits against Lehman in a single multidistrict litigation. In February 2011, petitioner filed a separate complaint against respondents in the United States District Court for the Northern District of California. This suit was filed more than three years after the relevant transactions occurred. The complaint alleged identical securities law violations as the class-action complaint, but the claims were on petitioner’s own behalf. The suit was transferred and consolidated with the multidistrict litigation in the Southern District of New York. Soon thereafter, a proposed settlement was reached in the putative class action. Petitioner, apparently convinced it could obtain a more favorable recovery in its separate suit, opted out of the class. Respondents then moved to dismiss petitioner’s indi-vidual suit alleging §11 violations as untimely under the 3-year bar in the second sentence of §13. Petitioner countered that its individual suit was timely because that 3-year period was tolled during the pendency of the class-action filing. The principal authority cited to support petitioner’s argument that the 3-year period was tolled was American Pipe & Constr. Co. v. Utah, 414 U. S. 538 (1974). The District Court disagreed with petitioner’s argument, holding that the 3-year bar in §13 is not subject to tolling. The Court of Appeals for the Second Circuit affirmed. In agreement with the District Court, the Court of Appeals held that the tolling principle discussed in American Pipe is inapplicable to the 3-year time bar. In re Lehman Brothers Securities and ERISA Litigation, 655 Fed. Appx. 13, 15 (2016). As the Court of Appeals noted, there is disagreement about whether the tolling rule of American Pipe applies to the 3-year time bar in §13. Compare Joseph v. Wiles, 223 F. 3d 1155, 1166–1168 (CA10 2000), with Stein v. Regions Morgan Keegan Select High Income Fund, Inc., 821 F. 3d 780, 792–795 (CA6 2016), and Dusek v. JPMorgan Chase & Co., 832 F. 3d 1243, 1246–1249 (CA11 2016). The Court of Appeals also rejected petitioner’s alternative argument that its individual claims were “essentially ‘filed’ in the putative class complaint,” so that the filing of the class action within three years made the individual claims timely. 655 Fed. Appx., at 15. This Court granted certiorari. 580 U. S. ___ (2017). II The question then is whether §13 permits the filing of an individual complaint more than three years after the relevant securities offering, when a class-action complaint was timely filed, and the plaintiff filing the individual complaint would have been a member of the class but for opting out of it. The answer turns on the nature and purpose of the 3-year bar and of the tolling rule that petitioner seeks to invoke. Each will be addressed in turn. A As the Court explained in CTS Corp. v. Waldburger, 573 U. S. ___ (2014), statutory time bars can be divided into two categories: statutes of limitations and statutes of repose. Both “are mechanisms used to limit the temporal extent or duration of liability for tortious acts,” but “each has a distinct purpose.” Id., at ___–___ (slip op., at 5–6). Statutes of limitations are designed to encourage plaintiffs “to pursue diligent prosecution of known claims.” Id., at ___ (slip op., at 6) (internal quotation marks omitted). In accord with that objective, limitations periods begin to run “when the cause of action accrues”—that is, “when the plaintiff can file suit and obtain relief.” Id., at ___ (slip op., at 5) (internal quotation marks omitted). In a personal-injury or property-damage action, for example, more often than not this will be “ ‘when the injury occurred or was discovered.’ ” Ibid. In contrast, statutes of repose are enacted to give more explicit and certain protection to defendants. These statutes “effect a legislative judgment that a defendant should be free from liability after the legislatively determined period of time.” Id., at ___–___ (slip op., at 6–7) (internal quotation marks omitted). For this reason, statutes of repose begin to run on “the date of the last culpable act or omission of the defendant.” Id., at ___ (slip op., at 6). The 3-year time bar in §13 reflects the legislative objective to give a defendant a complete defense to any suit after a certain period. From the structure of §13, and the language of its second sentence, it is evident that the 3-year bar is a statute of repose. In fact, this Court has already described the provision as establishing “a period of repose,” which “ ‘impose[s] an outside limit’ ” on temporal liability. Lampf, Pleva, Lipkind, Prupis & Petigrow v. Gilbertson, 501 U. S. 350, 363 (1991) . The statute provides in clear terms that “[i]n no event” shall an action be brought more than three years after the securities offering on which it is based. 15 U. S. C. §77m. This instruction admits of no exception and on its face creates a fixed bar against future liability. See CTS, supra, at ___–___ (slip op., at 6–7); cf. United States v. Brockamp, 519 U. S. 347, 350 (1997) (noting that a statute that “sets forth its time limitations in unusually emphatic form . . . cannot easily be read as containing implicit exceptions”). The statute, furthermore, runs from the defendant’s last culpable act (the offering of the securities), not from the accrual of the claim (the plaintiff’s discovery of the defect in the registration statement). Under CTS, this point is close to a dispositive indication that the statute is one of repose. This view is confirmed by the two-sentence structure of §13. In addition to the 3-year time bar, §13 contains a 1-year statute of limitations. The limitations statute runs from the time when the plaintiff discovers (or should have discovered) the securities-law violation. The pairing of a shorter statute of limitations and a longer statute of repose is a common feature of statutory time limits. See, e.g., Gabelli v. SEC, 568 U. S. 442, 453 (2013) (“[S]tatutes applying a discovery rule . . . often couple that rule with an absolute provision for repose”). The two periods work together: The discovery rule gives leeway to a plaintiff who has not yet learned of a violation, while the rule of repose protects the defendant from an interminable threat of liability. Cf. Merck & Co. v. Reynolds, 559 U. S. 633, 650 (2010) (reasoning that 2-year discovery rule would not “subject defendants to liability for acts taken long ago,” because the statute also included an “unqualified bar on actions instituted ‘5 years after such violation’ ”). The history of the 3-year provision also supports its classification as a statute of repose. It is instructive to note that the statute was not enacted in its current form. The original version of the 1933 Securities Act featured a 2-year discovery period and a 10-year outside limit, see §13, 48Stat. 84, but Congress changed this framework just one year after its enactment. The discovery period was changed to one year and the outside limit to three years. See Securities Exchange Act of 1934, §207, 48Stat. 908. The evident design of the shortened statutory period was to protect defendants’ financial security in fast-changing markets by reducing the open period for potential liability. B The determination that the 3-year period is a statute of repose is critical in this case, for the question whether a tolling rule applies to a given statutory time bar is one “of statutory intent.” Lozano v. Montoya Alvarez, 572 U. S. 1 , ___ (2014) (slip op., at 8). The purpose of a statute of repose is to create “an absolute bar on a defendant’s temporal liability,” CTS, 573 U. S., at ___ (slip op., at 6) (alteration and internal quotation marks omitted); and that purpose informs the assessment of whether, and when, tolling rules may apply. In light of the purpose of a statute of repose, the provision is in general not subject to tolling. Tolling is permissible only where there is a particular indication that the legislature did not intend the statute to provide complete repose but instead anticipated the extension of the statutory period under certain circumstances. For example, if the statute of repose itself contains an express exception, this demonstrates the requisite intent to alter the operation of the statutory period. See 1 C. Corman, Limitation of Actions §1.1, pp. 4–5 (1991) (Corman); see, e.g., 29 U. S. C. §1113 (establishing a 6-year statute of repose, but stipulating that, in case of fraud, the 6-year period runs from the plaintiff’s discovery of the violation). In contrast, where the legislature enacts a general tolling rule in a different part of the code—e.g., a rule that suspends time limits until the plaintiff reaches the age of majority—courts must analyze the nature and relation of the legislative purpose of each provision to determine which controls. See 2 Corman §10.2.1, at 108. In keeping with the statute-specific nature of that analysis, courts have reached different conclusions about whether general tolling statutes govern particular periods of repose. Ibid., n. 15. Of course, not all tolling rules derive from legislative enactments. Some derive from the traditional power of the courts to “ ‘apply the principles . . . of equity jurisprudence.’ ” Young v. United States, 535 U. S. 43, 50 (2002) (alteration omitted). The classic example is the doctrine of equitable tolling, which permits a court to pause a statu-tory time limit “when a litigant has pursued his rights diligently but some extraordinary circumstance prevents him from bringing a timely action.” Lozano, 572 U. S., at ___ (slip op., at 7). Tolling rules of that kind often apply to statutes of limitations based on the presumption that Congress “ ‘legislate[s] against a background of common-law adjudicatory principles.’ ” Id., at ___ (slip op., at 8). The purpose and effect of a statute of repose, by contrast, is to override customary tolling rules arising from the equitable powers of courts. By establishing a fixed limit, a statute of repose implements a “ ‘legislative decisio[n] that as a matter of policy there should be a specific time beyond which a defendant should no longer be subjected to protracted liability.’ ” CTS, 573 U. S., at ___ (slip op., at 7). The unqualified nature of that determination supersedes the courts’ residual authority and forecloses the extension of the statutory period based on equitable principles. For this reason, the Court repeatedly has stated in broad terms that statutes of repose are not subject to equitable tolling. See, e.g., id., at ___–___ (slip op., at 7–8); Lampf, Pleva, 501 U. S., at 363. C Petitioner contends that the 3-year provision is subject to tolling based on the rationale and holding in the Court’s decision in American Pipe. The language of the 3-year statute does not refer to or impliedly authorize any exceptions for tolling. If American Pipe had itself been grounded in a legislative enactment, perhaps an argument couldbe made that the enactment expressed a legislative objective to modify the 3-year period. If, however, the tolling decision in American Pipe derived from equity principles, it cannot alter the unconditional language and purpose of the 3-year statute of repose. In American Pipe, a timely class-action complaint was filed asserting violations of federal antitrust law. 414 U. S., at 540. Class certification was denied because the class was not large enough, see Fed. Rule Civ. Proc. 23(a)(1), and individuals who otherwise would have been members of the class then filed motions to intervene as individual plaintiffs. The motions were denied on the grounds that the applicable 4-year time bar had expired. See 15 U. S. C. §15b. The Court of Appeals reversed, permitting intervention. This Court affirmed. It held the individual plaintiffs’ motions to intervene were timely because “the commencement of a class action suspends the applicable statute of limitations as to all asserted members of the class.” American Pipe, 414 U. S., at 554. The Court reasoned that this result was consistent “both with the procedures of Rule 23 and with the proper function of the limitations statute” at issue. Id., at 555. First, the tolling furthered “the purposes of litigative efficiency and economy” served by Rule 23. Id., at 556. Without the tolling, “[p]otential class members would be induced to file protective motions to intervene or to join in the event that a class was later found unsuitable,” which would “breed needless duplication of motions.” Id., at 553–554. Second, the tolling was in accord with “the functional operation of a statute of limitations.” Id., at 554. By filing a class complaint within the statutory period, the named plaintiff “notifie[d] thedefendants not only of the substantive claims being brought against them, but also of the number and generic identities of the potential plaintiffs who may participate in the judgment.” Id., at 555. As this discussion indicates, the source of the tolling rule applied in American Pipe is the judicial power to promote equity, rather than to interpret and enforce statutory provisions. Nothing in the American Pipe opinion suggests that the tolling rule it created was mandated by the text of a statute or federal rule. Nor could it have. The central text at issue in American Pipe was Rule 23, and Rule 23 does not so much as mention the extension or suspension of statutory time bars. The Court’s holding was instead grounded in the traditional equitable powers of the judiciary. The Court described its rule as authorized by the “judicial power to toll statutes of limitations.” Id., at 558; see also id., at 555 (“the tolling rule we establish here” (emphasis added)). The Court also relied on cases that are paradigm applications of equitable tolling principles, explaining with approval that tolling in one such case was based on “considerations ‘deeply rooted in our jurisprudence.’ ” Id., at 559 (quoting Glus v. Brooklyn Eastern Dist. Terminal, 359 U. S. 231, 232 (1959) ; alteration omitted); see also 414 U. S., at 559 (citing Holmberg v. Armbrecht, 327 U. S. 392 (1946) ). The Court noted too that “bad faith” was not the cause of the District Court’s denial of class certification. 414 U. S., at 553 (internal quotation marks omitted). Perhaps for these reasons, this Court has referred to American Pipe as “equitable tolling.” See Irwin v. Department of Veterans Affairs, 498 U. S. 89 , and n. 3 (1990); see also Young, supra, at 49; Greyhound Corp. v. Mt. Hood Stages, Inc., 437 U. S. 322 , n. (1978) (Burger, C. J., concurring) (using American Pipe as an example of “[t]he authority of a federal court, sitting as a chancellor, to toll a statute of limitations on equitable grounds”). It is true, however, that the American Pipe Court did not consider the criteria of the formal doctrine of equitable tolling in any direct manner. It did not analyze, for example, whether the plaintiffs pursued their rights with special care; whether some extraordinary circumstance prevented them from intervening earlier; or whether the defendant engaged in misconduct. See Holland v. Florida, 560 U. S. 631, 649 (2010) (identifying these considerations); Young, 535 U. S., at 50 (same). The balance of the Court’s reasoning nonetheless reveals a rule based on traditional equitable powers, designed to modify a stat-utory time bar where its rigid application would create injustice. D This analysis shows that the American Pipe tolling rule does not apply to the 3-year bar mandated in §13. As explained above, the 3-year limit is a statute of repose. See supra, at 5–7. And the object of a statute of repose, to grant complete peace to defendants, supersedes the application of a tolling rule based in equity. See supra, at 7–8. No feature of §13 provides that deviation from its time limit is permissible in a case such as this one. To the contrary, the text, purpose, structure, and history of the statute all disclose the congressional purpose to offer defendants full and final security after three years. Petitioner raises four counterarguments, but they are not persuasive. First, petitioner contends that this case is indistinguishable from American Pipe itself. If the 3-year bar here cannot be tolled, petitioner reasons, then there was no justification for the American Pipe Court’s contrary decision to suspend the time bar in that case. American Pipe, however, is distinguishable. The statute in American Pipe was one of limitations, not of repose; it began to run when “ ‘the cause of action accrued.’ ” 414 U. S., at 541, n. 2 (quoting 15 U. S. C. §15b). The statute in the instant case, however, is a statute of repose. Consistent with the different purposes embodied in statutes of limitations and statutes of repose, it is reasonable that the former may be tolled by equitable considerations even though the latter in most circumstances may not. See supra, at 7–8. Second, petitioner argues that the filing of a class-action complaint within three years fulfills the purposes of a statutory time limit with regard to later filed suits by individual members of the class. That is because, according to petitioner, the class complaint puts a defendant on notice as to the content of the claims against it and the set of potential plaintiffs who might assert those claims. It is true that the American Pipe Court, in permitting tolling, suggested that generic notice satisfied the purposes of the statute of limitations in that case. See 414 U. S., at 554–555. While this was deemed sufficient in balancing the equities to allow tolling under the antitrust statute, it must be noted that here the analysis differs because the purpose of a statute of repose is to give the defendant full protection after a certain time. If the number and identity of individual suits, where they may be filed, and the litigation strategies they will use are unknown, a defendant cannot calculate its potential liability or set its own plans for litigation with much precision. The initiation of separate individual suits may thus increase a defendant’s practical burdens. See, e.g., Cottreau, Note, The Due Process Right To Opt Out of Class Actions, 73 N. Y. U. L. Rev. 480, 486, and n. 29 (1998) (“A defendant’s transaction costs are likely to be reduced by having to defend just one action”). The emergence of individual suits, furthermore, may increase a defendant’s financial liability; for plaintiffs who opt out have considerable leverage and, as a result, may obtain outsized recoveries. See, e.g., Coffee, Accountability and Competition in Securities Class Actions: Why “Exit” Works Better Than “Voice,” 30 Cardozo L. Rev. 407, 417, 432–433 (2008); Perino, Class Action Chaos? The Theory of the Core and an Analysis of Opt-Out Rights in Mass Tort Class Actions, 46 Emory L. J. 85, 97 (1997). These uncertainties can put defendants at added risk in conducting business going forward, causing destabilization in markets which react with sensitivity to these matters. By permitting a class action to splinter into individual suits, the application of American Pipe tolling would threaten to alter and expand a defendant’s accountability, contradicting the substance of a statute of repose. All this is not to suggest how best to further equity under these circumstances but simply to support the recognition that a statute of repose supersedes a court’s equitable balancing powers by setting a fixed time period for claims to end. Third, petitioner contends that dismissal of its individ-ual suit as untimely would eviscerate its ability to opt out, an ability this Court has indicated should not be disregarded. See Wal-Mart Stores, Inc. v. Dukes, 564 U. S. 338, 363 (2011) . It does not follow, however, from any privilege to opt out that an ensuing suit can be filed without regard to mandatory time limits set by statute. Fourth, petitioner argues that declining to apply American Pipe tolling to statutes of repose will create inefficiencies. It contends that nonnamed class members will inundate district courts with protective filings. Even if petitioner were correct, of course, this Court “lack[s] the authority to rewrite” the statute of repose or to ignore its plain import. Baker Botts L. L. P. v. ASARCO LLC, 576 U. S. ___, ___ (2015) (slip op., at 12). And petitioner’s concerns likely are overstated. Petitioner has not offered evidence of any recent influx of protective filings in the Second Circuit, where the rule affirmed here has been the law since 2013. This is not surprising. The very premise of class actions is that “ ‘small recoveries do not provide the incentive for any individual to bring a solo action prosecuting his or her rights.’ ” Amchem Products, Inc. v. Windsor, 521 U. S. 591, 617 (1997) . Many individual class members may have no interest in protecting their right to litigate on an individ-ual basis. Even assuming that they do, the process is un-likely to be as onerous as petitioner claims. A simple motion to intervene or request to be included as a named plaintiff in the class-action complaint may well suffice. See, e.g., Brief for Washington Legal Foundation as Amicus Curiae 6–11 (describing procedures); Brief for Securities Industry and Financial Markets Association et al. as Amici Curiae 16, 19–20 (same). District courts, furthermore, have ample means and methods to administer their dockets and to ensure that any additional filings proceed in an orderly fashion. Cf. Dietz v. Bouldin, 579 U. S. ___, ___ (2016) (slip op., at 6) (“[D]istrict courts have the inherent authority to manage their dockets and courtrooms with a view toward the efficient and expedient resolution of cases”). III Petitioner makes an alternative argument that does not depend on tolling. Petitioner submits its individual suit was timely in any event. Section 13 provides that an “action” must be “brought” within three years of the relevant securities offering. See 15 U. S. C. §77m. Petitioner argues that requirement is met here because the filing of the class-action complaint “brought” petitioner’s individual “action” within the statutory time period. This argument rests on the premise that an “action” is “brought” when substantive claims are presented to any court, rather than when a particular complaint is filed in a particular court. The term “action,” however, refers to a judicial “proceeding,” or perhaps to a “suit”—not to the general content of claims. See Black’s Law Dictionary 41 (3d ed. 1933) (defining “action” as, inter alia, “an ordinary proceeding in a court of justice”); see also id., at 43 (“The terms ‘action’ and ‘suit’ are . . . nearly, if not entirely, synonymous”). Whether or not petitioner’s individual complaint alleged the same securities law violations as the class-action complaint, it defies ordinary understanding to suggest that its filing—in a separate forum, on a separate date, by a separate named party—was the same “action,” “proceeding,” or “suit.” The limitless nature of petitioner’s argument, furthermore, reveals its implausibility. It appears that, in petitioner’s view, the bringing of the class action would make any subsequent action raising the same claims timely. Taken to its logical limit, an individual action would be timely even if it were filed decades after the original securities offering—provided a class-action complaint had been filed at some point within the initial 3-year period. Congress would not have intended this result. Petitioner’s argument also fails because it is inconsistent with the reasoning in American Pipe itself. If the filing of a class action made all subsequent actions by putative class members timely, there would be no need for tolling at all. Yet this Court has described American Pipe as creating a tolling rule, necessary to permit the ensuing individual actions to proceed. See, e.g., American Pipe, 414 U. S., at 555; Irwin, 498 U. S., at 96, n. 3; Crown, Cork & Seal Co. v. Parker, 462 U. S. 345, 350 (1983) . Indeed, the American Pipe Court reasoned that the class-action complaint “was filed with 11 days yet to run” in the statutory period, so the motions for intervention were timely only if filed within 11 days after the denial of class certification. 414 U. S., at 561. If the filing of the class action “brought” any included individual actions, it would have sufficed for the Court to note the date on which the class action was filed and deem all subsequent individual actions proper, regardless when filed. * * * Tolling may be of great value to allow injured persons to recover for injuries that, through no fault of their own, they did not discover because the injury or the perpetrator was not evident until the limitations period otherwise would have expired. This is of obvious utility in the securities market, where complex transactions and events can be obscure and difficult for a market participant to analyze or apprehend. In a similar way, tolling as allowed in American Pipe may protect plaintiffs who anticipated their interests would be protected by a class action but later learned that a class suit could not be maintained for reasons outside their control. The purpose of a statute of repose, on the other hand, is to allow more certainty and reliability. These ends, too, are a necessity in a marketplace where stability and reliance are essential components of valuation and expectation for financial actors. The statute in this case reconciles these different ends by its two-tier structure: a conventional statute of limitations in the first clause and a statute of repose in the second. The statute of repose transforms the analysis. In a hypothetical case with a different statutory scheme, consisting of a single limitations period without an additional outer limit, a court’s equitable power under American Pipe in many cases would authorize the relief petitioner seeks. Here, however, the Court need not consider how equitable considerations should be formulated or balanced, for the mandate of the statute of repose takes the case outside the bounds of the American Pipe rule. The final analysis, then, is straightforward. The 3-year time bar in §13 of the Securities Act is a statute of repose. Its purpose and design are to protect defendants against future liability. The statute displaces the traditional power of courts to modify statutory time limits in the name of equity. Because the American Pipe tolling rule is rooted in those equitable powers, it cannot extend the 3-year period. Petitioner’s untimely filing of its individual action is ground for dismissal. The judgment of the Court of Appeals for the Second Circuit is affirmed. It is so ordered. APPENDIX Respondents are the following financial securities firms: ANZ Securities, Inc.; Bankia, S. A.; BBVA Securities Inc.; BMO Capital Markets Corp.; BNP Paribas FS, LLC; BNP Paribas S. A.; BNY Mellon Capital Markets, LLC; CIBC World Markets Corp.; Citigroup Global Markets Inc.; Daiwa Capital Markets Europe Limited; DZ Financial Markets LLC; HSBC Securities (USA) Inc.; HVB Capital Markets, Inc.; ING Financial Markets LLC; Mizuho Securities USA Inc.; M. R. Beal & Company; Muriel Siebert & Co. Inc.; nabSecurities LLC; Natixis Securities Americas LLC; RBC Capital Markets LLC; RBS Securities, Inc.; RBS WCS Holding Company; Santander Investment Securities Inc.; Scotia Capital (USA) Inc.; SG Americas Securities, LLC; Sovereign Securities Corporation LLC; SunTrust Capital Markets, Inc.; Utendahi Capital Partners, L. P.; and Wells Fargo Securities, LLC.

581.US.2016_15-1262

The Equal Protection Clause of the Fourteenth Amendment prevents a State, in the absence of “sufficient justification,” from “separating its citizens into different voting districts on the basis of race.” Bethune-Hill v. Virginia State Bd. of Elections, 580 U. S. ___, ___. When a voter sues state officials for drawing such race-based lines, this Court’s decisions call for a two-step analysis. First, the plaintiff must prove that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller v. Johnson, 515 U. S. 900 . Second, if racial considerations did predominate, the State must prove that its race-based sorting of voters serves a “compelling interest” and is “narrowly tailored” to that end, Bethune-Hill, 580 U. S., at ___. This Court has long assumed that one compelling interest is compliance with the Voting Rights Act of 1965 (VRA or Act). When a State invokes the VRA to justify race-based districting, it must show (to meet the “narrow tailoring” requirement) that it had “good reasons” for concluding that the statute required its action. Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___. A district court’s factual findings made in the course of this two-step inquiry are reviewed only for clear error. See Fed. Rule Civ. Proc. 52(a)(6); Easley v. Cromartie, 532 U. S. 234 (Cromartie II). This case concerns North Carolina’s redrawing of two congressional districts, District 1 and District 12, after the 2010 census. Prior to that redistricting, neither district had a majority black voting-age population (BVAP), but both consistently elected the candidates preferred by most African-American voters. The new map significantly altered both District 1 and District 12. The State needed to add almost 100,000 people to District 1 to comply with the one-person-one-vote principle, and it chose to take most of those people from heavily black areas of Durham—increasing the district’s BVAP from 48.6% to 52.7%. The State also reconfigured District 12, increasing its BVAP from 43.8% to 50.7%. Registered voters in those districts (here called “the plaintiffs”) filed suit against North Carolina officials (collectively, “the State” or “North Carolina”), complaining of impermissible racial gerrymanders. A three-judge District Court held both districts unconstitutional. It found that racial considerations predominated in the drawing of District 1’s lines and rejected the State’s claim that this action was justified by the VRA. As for District 12, the court again found that race predominated, and it explained that the State made no attempt to justify its attention to race in designing that district. Held: 1. North Carolina’s victory in a similar state-court lawsuit does not dictate the disposition of this case or alter the applicable standard of review. Before this case was filed, a state trial court rejected a claim by several civil rights groups that Districts 1 and 12 were unlawful racial gerrymanders. The North Carolina Supreme Court affirmed that decision under the state-court equivalent of clear error review. The State claims that the plaintiffs are members of the same organizations that brought the earlier case, and thus precluded from raising the same questions anew. But the State never satisfied the District Court that the alleged affiliation really existed. And because the District Court’s factual finding was reasonable, it defeats North Carolina’s attempt to argue for claim or issue preclusion here. The State’s backup argument about the proper standard of review also falls short. The rule that a trial court’s factual findings are reviewed only for clear error contains no exception for findings that diverge from those made in another court. See Fed. Rule Civ. Proc. 52(a)(6). Although the state court’s decision is certainly relevant, the premise of clear error review is that there are often “two permissible views of the evidence.” Anderson v. Bessemer City, 470 U. S. 564 . Even assuming that the state court’s findings capture one such view, the only question here is whether the District Court’s assessment represents another. Pp. 7–10. 2. The District Court did not err in concluding that race furnished the predominant rationale for District 1’s redesign and that the State’s interest in complying with the VRA could not justify that consideration of race. Pp. 10–18. (a) The record shows that the State purposefully established a racial target for the district and that the target “had a direct and significant impact” on the district’s configuration, Alabama, 575 U. S., at ___, subordinating other districting criteria. Faced with this body of evidence, the District Court did not clearly err in finding that race predominated in drawing District 1; indeed, it could hardly have concluded anything but. Pp. 10–12. (b) North Carolina’s use of race as the predominant factor in designing District 1 does not withstand strict scrutiny. The State argues that it had good reasons to believe that it had to draw a majority-minority district to avoid liability for vote dilution under §2 of the VRA. Thornburg v. Gingles, 478 U. S. 30 , identifies three threshold conditions for proving such a vote-dilution claim: (1) A “minority group” must be “sufficiently large and geographically compact to constitute a majority” in some reasonably configured legislative district, id., at 50; (2) the minority group must be “politically cohesive,” id., at 51; and (3) a district’s white majority must “vote[ ] sufficiently as a bloc” to usually “defeat the minority’s preferred candidate,” ibid. If a State has good reason to think that all three of these conditions are met, then so too it has good reason to believe that §2 requires drawing a majority-minority district. But if not, then not. Here, electoral history provided no evidence that a §2 plaintiff could demonstrate the third Gingles prerequisite. For nearly 20 years before the new plan’s adoption, African-Americans made up less than a majority of District 1’s voters, but their preferred candidates scored consistent victories. District 1 thus functioned as a “crossover” district, in which members of the majority help a “large enough” minority to elect its candidate of choice. Bartlett v. Strickland, 556 U. S. 1 (plurality opinion). So experience gave the State no reason to think that the VRA required it to ramp up District 1’s BVAP. The State counters that because it needed to substantially increase District 1’s population, the question facing the state mapmakers was not whether the then-existing District 1 violated §2, but whether the future District 1 would do so if drawn without regard to race. But that reasoning, taken alone, cannot justify the State’s race-based redesign of the district. Most important, the State points to no meaningful legislative inquiry into the key issue it identifies: whether a new, enlarged District 1, created without a focus on race, could lead to §2 liability. To have a strong basis to conclude that §2 demands race-based measures to augment a district’s BVAP, the State must evaluate whether a plaintiff could establish the Gingles preconditions in a new district created without those measures. Nothing in the legislative record here fits that description. And that is no accident: The redistricters believed that this Court’s decision in Strickland mandated a 50%-plus BVAP in District 1. They apparently reasoned that if, as Strickland held, §2 does not require crossover districts (for groups insufficiently large under Gingles), then §2 also cannot be satisfied by crossover districts (for groups meeting Gingles’ size condition). But, as this Court’s §2 jurisprudence makes clear, unless each of the three Gingles prerequisites is established, “there neither has been a wrong nor can be a remedy.” Growe v. Emison, 507 U. S. 25 . North Carolina’s belief that it was compelled to redraw District 1 (a successful crossover district) as a majority-minority district thus rested on a pure error of law. Accordingly, the Court upholds the District Court’s conclusion that the State’s use of race as the predominant factor in designing District 1 does not withstand strict scrutiny. Pp. 12–18. 3. The District Court also did not clearly err by finding that race predominated in the redrawing of District 12. Pp. 18–34. (a) The district’s legality turns solely on which of two possible reasons predominantly explains its reconfiguration. The plaintiffs contended at trial that North Carolina intentionally increased District 12’s BVAP in the name of ensuring preclearance under §5 of the VRA. According to the State, by contrast, the mapmakers moved voters in and out of the district as part of a “strictly” political gerrymander, without regard to race. After hearing evidence supporting both parties’ accounts, the District Court accepted the plaintiffs’. Getting to the bottom of a dispute like this one poses special challenges for a trial court, which must make “ ‘a sensitive inquiry’ ” into all “ ‘circumstantial and direct evidence of intent’ ” to assess whether the plaintiffs have proved that race, not politics, drove a district’s lines. Hunt v. Cromartie, 526 U. S. 541 (Cromartie I). This Court’s job is different—and generally easier. It affirms a trial court’s factual finding as to racial predominance so long as the finding is “plausible”; it reverses only when “left with the definite and firm conviction that a mistake has been committed.” Anderson, 470 U. S., at 573–574. In assessing a finding’s plausibility, moreover, the Court gives singular deference to a trial court’s judgments about the credibility of witnesses. See Fed. Rule Civ. Proc. 52(a)(6). Applying those principles here, the evidence at trial—including live witness testimony subject to credibility determinations—adequately supports the District Court’s conclusion that race, not politics, accounted for District 12’s reconfiguration. And contrary to the State’s view, the court had no call to dismiss this challenge just because the plaintiffs did not proffer an alternative design for District 12. Pp. 18–21. (b) By slimming the district and adding a couple of knobs to its snakelike body, North Carolina added 35,000 African-Americans and subtracted 50,000 whites, turning District 12 into a majority-minority district. State Senator Robert Rucho and State Representative David Lewis—the chairs of the two committees responsible for preparing the revamped plan—publicly stated that racial considerations lay behind District 12’s augmented BVAP. Specifically, Rucho and Lewis explained that because part of Guilford County, a jurisdiction covered by §5 of the VRA, lay in the district, they had increased the district’s BVAP to ensure preclearance of the plan. Dr. Thomas Hofeller, their hired mapmaker, confirmed that intent. The State’s preclearance submission to the Justice Department indicated a similar determination to concentrate black voters in District 12. And, in testimony that the District Court found credible, Congressman Mel Watt testified that Rucho disclosed a majority-minority target to him in 2011. Hofeller testified that he had drawn District 12’s lines based on political data, and that he checked the racial data only after he drew a politics-based line between adjacent areas in Guilford County. But the District Court disbelieved Hofeller’s asserted indifference to the new district’s racial composition, pointing to his contrary deposition testimony and a significant contradiction in his trial testimony. Finally, an expert report lent circumstantial support to the plaintiffs’ case, showing that, regardless of party, a black voter in the region was three to four times more likely than a white voter to cast a ballot within District 12’s borders. The District Court’s assessment that all this evidence proved racial predominance clears the bar of clear error review. Maybe this Court would have evaluated the testimony differently had it presided over the trial; or then again, maybe it would not have. Either way, the Court is far from having a “definite and firm conviction” that the District Court made a mistake in concluding from the record before it that racial considerations predominated in District 12’s design. Pp. 21–28. (c) Finally, North Carolina argues that when race and politics are competing explanations of a district’s lines, plaintiffs must introduce an alternative map that achieves a State’s asserted political goals while improving racial balance. Such a map can serve as key evidence in a race-versus-politics dispute, but it is hardly the only means to disprove a State’s contention that politics drove a district’s lines. In this case, the plaintiffs’ introduction of mostly direct and some circumstantial evidence gave the District Court a sufficient basis, sans any map, to resolve the race-or-politics question. Although a plaintiff will sometimes need an alternative map, as a practical matter, to make his case, such a map is merely an evidentiary tool to show that an equal protection violation has occurred; neither its presence nor its absence can itself resolve a racial gerrymandering claim. North Carolina claims that a passage of this Court’s opinion in Cromartie II makes an alternative map essential in cases like this one, but the reasoning of Cromartie II belies that reading. The Court’s opinion nowhere attempts to explicate or justify the categorical rule that the State claims to find there, and the entire thrust of the opinion runs counter to an inflexible counter-map requirement. Rightly understood, the passage on which the State relies had a different and narrower point: Given the weak evidence of a racial gerrymander offered in Cromartie II, only maps that would actually show what the plaintiffs’ had not could carry the day. This case, in contrast, turned not on the possibility of creating more optimally constructed districts, but on direct evidence of the General Assembly’s intent in creating the actual District 12—including many hours of trial testimony subject to credibility determinations. That evidence, the District Court plausibly found, itself satisfied the plaintiffs’ burden of debunking North Carolina’s politics defense. Pp. 28–34. 159 F. Supp. 3d 600, affirmed. Kagan, J., delivered the opinion of the Court, in which Thomas, Ginsburg, Breyer, and Sotomayor, JJ., joined. Thomas, J., filed a concurring opinion. Alito, J., filed an opinion concurring in the judgment in part and dissenting in part, in which Roberts, C. J., and Kennedy, J., joined. Gorsuch, J., took no part in the consideration or decision of the case.

The Constitution entrusts States with the job of designing congressional districts. But it also imposes an important constraint: A State may not use race as the predominant factor in drawing district lines unless it has a compelling reason. In this case, a three-judge District Court ruled that North Carolina officials violated that bar when they created two districts whose voting-age populations were majority black. Applying a deferential standard of review to the factual findings underlying thatdecision, we affirm. I A The Equal Protection Clause of the Fourteenth Amendment limits racial gerrymanders in legislative districting plans. It prevents a State, in the absence of “sufficient justification,” from “separating its citizens into different voting districts on the basis of race.” Bethune-Hill v. Virginia State Bd. of Elections, 580 U. S. ___, ___ (2017) (slip op., at 6) (internal quotation marks and alteration omitted). When a voter sues state officials for drawing such race-based lines, our decisions call for a two-step analysis. First, the plaintiff must prove that “race was the predominant factor motivating the legislature’s decision to place a significant number of voters within or without a particular district.” Miller v. Johnson, 515 U. S. 900, 916 (1995) . That entails demonstrating that the legislature “subordinated” other factors—compactness, respect for political subdivisions, partisan advantage, what have you—to “racial considerations.” Ibid. The plaintiff may make the required showing through “direct evidence” of legislative intent, “circumstantial evidence of a district’s shape and demographics,” or a mix of both. Ibid.[1] Second, if racial considerations predominated over others, the design of the district must withstand strict scrutiny. See Bethune-Hill, 580 U. S., at ____ (slip op., at 13). The burden thus shifts to the State to prove that its race-based sorting of voters serves a “compelling interest” and is “narrowly tailored” to that end. Ibid. This Court has long assumed that one compelling interest is complying with operative provisions of the Voting Rights Act of 1965 (VRA or Act), 79Stat. 437, as amended, 52 U. S. C. §10301 et seq. See, e.g., Shaw v. Hunt, 517 U. S. 899, 915 (1996) (Shaw II). Two provisions of the VRA—§2 and §5—are involved in this case. §§10301, 10304. Section 2 prohibits any “standard, practice, or procedure” that “results in a denial or abridgement of the right . . . to vote on account of race.” §10301(a). We have construed that ban to extend to “vote dilution”—brought about, most relevantly here, by the “dispersal of [a group’s members] into districts in which they constitute an ineffective minority of voters.” Thornburg v. Gingles, 478 U. S. 30 , n. 11 (1986). Section 5, at the time of the districting in dispute, worked through a different mechanism. Before this Court invalidated its coverage formula, see Shelby County v. Holder, 570 U. S. __ (2013), that section required certain jurisdictions (including various North Carolina counties) to pre-clear voting changes with the Department of Justice, so as to forestall “retrogression” in the ability of racial minorities to elect their preferred candidates, Beer v. United States, 425 U. S. 130, 141 (1976) . When a State invokes the VRA to justify race-based districting, it must show (to meet the “narrow tailoring” requirement) that it had “a strong basis in evidence” for concluding that the statute required its action. Alabama Legislative Black Caucus v. Alabama, 575 U. S. ___, ___ (2015) (slip op., at 22). Or said otherwise, the State must establish that it had “good reasons” to think that it would transgress the Act if it did not draw race-based district lines. Ibid. That “strong basis” (or “good reasons”) standard gives States “breathing room” to adopt reasonable compliance measures that may prove, in perfect hindsight, not to have been needed. Bethune-Hill, 580 U. S., at ___ (slip op., at 16). A district court’s assessment of a districting plan, in accordance with the two-step inquiry just described, warrants significant deference on appeal to this Court.[2] We of course retain full power to correct a court’s errors of law, at either stage of the analysis. But the court’s findings of fact—most notably, as to whether racial considerations predominated in drawing district lines—are subject to review only for clear error. See Fed. Rule Civ. Proc. 52(a)(6); Easley v. Cromartie, 532 U. S. 234, 242 (2001) (Cromartie II); id., at 259 (Thomas, J., dissenting). Under that standard, we may not reverse just because we “would have decided the [matter] differently.” Anderson v. Bessemer City, 470 U. S. 564, 573 (1985) . A finding that is “plausible” in light of the full record—even if another is equally or more so—must govern. Id., at 574. B This case concerns North Carolina’s most recent redrawing of two congressional districts, both of which have long included substantial populations of black voters. In its current incarnation, District 1 is anchored in the northeastern part of the State, with appendages stretching both south and west (the latter into Durham). District 12 begins in the south-central part of the State (where it takes in a large part of Charlotte) and then travels northeast, zig-zagging much of the way to the State’s northern border. (Maps showing the districts are included in an appendix to this opinion.) Both have quite the history before this Court. We first encountered the two districts, in their 1992 versions, in Shaw v. Reno, 509 U. S. 630 (1993) . There, we held that voters stated an equal protection claim by alleging that Districts 1 and 12 were unwarranted racial gerrymanders. See id., at 642, 649. After a remand to the District Court, the case arrived back at our door. See Shaw II, 517 U. S. 899 . That time, we dismissed the challenge to District 1 for lack of standing, but struck down District 12. The design of that “serpentine” district, we held, was nothing if not race-centric, and could not be justified as a reasonable attempt to comply with the VRA. Id., at 906; see id., at 911–918. The next year, the State responded with a new districting plan, including a new District 12—and residents of that district brought another lawsuit alleging an impermissible racial gerrymander. A District Court sustained the claim twice, but both times this Court reversed. See Hunt v. Cromartie, 526 U. S. 541 (1999) (Cromartie I); Cromartie II, 532 U. S. 234 . Racial considerations, we held, did not predominate in designing the revised District 12. Rather, that district was the result of a political gerrymander—an effort to engineer, mostly “without regard to race,” a safe Democratic seat. Id., at 245. The State redrew its congressional districts again in 2001, to account for population changes revealed in the prior year’s census. Under the 2001 map, which went unchallenged in court, neither District 1 nor District 12 had a black voting-age population (called a “BVAP”) that was a majority of the whole: The former had a BVAP of around 48%, the latter a BVAP of around 43%. See App. 312, 503. Nonetheless, in five successive general elections conducted in those reconfigured districts, all the candidates preferred by most African-American voters won their contests—and by some handy margins. In District 1, black voters’ candidates of choice garnered as much as 70% of the total vote, and never less than 59%. See 5 Record 636, 638, 641, 645, 647 (Pls. Exh. 112). And in District 12, those candidates won with 72% of the vote at the high end and 64% at the low. See id., at 637, 640, 643, 646, 650. Another census, in 2010, necessitated yet another congressional map—(finally) the one at issue in this case. State Senator Robert Rucho and State Representative David Lewis, both Republicans, chaired the two committees jointly responsible for preparing the revamped plan. They hired Dr. Thomas Hofeller, a veteran political mapmaker, to assist them in redrawing district lines. Several hearings, drafts, and revisions later, both chambers of the State’s General Assembly adopted the scheme the three men proposed. The new map (among other things) significantly altered both District 1 and District 12. The 2010 census had revealed District 1 to be substantially underpopulated: To comply with the Constitution’s one-person-one-vote principle, the State needed to place almost 100,000 new people within the district’s boundaries. See App. 2690; Evenwel v. Abbott, 578 U. S. ___, ___ (2016) (slip op., at 3) (explaining that “[s]tates must draw congressional districts with populations as close to perfect equality as possible”). Rucho, Lewis, and Hofeller chose to take most of those people from heavily black areas of Durham, requiring a finger-like extension of the district’s western line. See Appendix, infra. With that addition, District 1’s BVAP rose from 48.6% to 52.7%. See App. 312–313. District 12, for its part, had no need for significant total-population changes: It was overpopulated by fewer than 3,000 people out of over 730,000. See id., at 1150. Still, Rucho, Lewis, and Hofeller decided to reconfigure the district, further narrowing its already snakelike body while adding areas at either end—most relevantly here, in Guilford County. See Appendix, infra; App. 1164. Those changes appreciably shifted the racial composition of District 12: As the district gained some 35,000 African-Americans of voting age and lost some 50,000 whites of that age, its BVAP increased from 43.8% to 50.7%. See 2 Record 349 (Fourth Affidavit of Dan Frey, Exh. 5); id., at 416 (Exh. 11). Registered voters in the two districts (David Harris and Christine Bowser, here called “the plaintiffs”) brought this suit against North Carolina officials (collectively, “the State” or “North Carolina”), complaining of impermissible racial gerrymanders. After a bench trial, a three-judge District Court held both districts unconstitutional. All the judges agreed that racial considerations predominated in the design of District 1. See Harris v. McCrory, 159 F. Supp. 3d 600, 611 (MDNC 2016). And in then applying strict scrutiny, all rejected the State’s argument that it had a “strong basis” for thinking that the VRA compelled such a race-based drawing of District 1’s lines. Id., at 623. As for District 12, a majority of the panel held that “race predominated” over all other factors, including partisanship. Id., at 622. And the court explained that the State had failed to put forward any reason, compelling or otherwise, for its attention to race in designing that district. See ibid. Judge Osteen dissented from the conclusion that race, rather than politics, drove District 12’s lines—yet still characterized the majority’s view as “[e]minently reasonable.” Id., at 640. The State filed a notice of appeal, and we noted probable jurisdiction. McCrory v. Harris, 579 U. S. ___ (2016). II We address at the outset North Carolina’s contention that a victory it won in a very similar state-court lawsuit should dictate (or at least influence) our disposition of this case. As the State explains, the North Carolina NAACP and several other civil rights groups challenged Districts 1 and 12 in state court immediately after their enactment, charging that they were unlawful racial gerrymanders. See Brief for Appellants 19–20. By the time the plaintiffs before us filed this action, the state trial court, in Dickson v. Rucho, had rejected those claims—finding that in District 1 the VRA justified the General Assembly’s use of race and that in District 12 race was not a factor at all. See App. 1969. The North Carolina Supreme Court then affirmed that decision by a 4–3 vote, applying the state-court equivalent of clear error review. See Dickson v. Rucho, 368 N. C. 481, 500, 781 S. E. 2d 404, 419 (2015), modified on denial of reh’g, 368 N. C. 673, 789 S. E. 2d 436 (2016), cert. pending, No. 16–24. In this Court, North Carolina makes two related arguments based on the Dickson litigation: first, that the state trial court’s judgment should have barred this case altogether, under familiar principles of claim and issue preclusion; and second, that the state court’s conclusions should cause us to conduct a “searching review” of the decision below, rather than deferring (as usual) to its factual findings. Reply Brief 6. The State’s preclusion theory rests on an assertion about how the plaintiffs in the two cases are affiliated. As the State acknowledges, one person’s lawsuit generally does not bar another’s, no matter how similar they are in substance. See Taylor v. Sturgell, 553 U. S. 880 –893 (2008) (noting the “deep-rooted historic tradition that everyone should have his own day in court”). But when plaintiffs in two cases have a special relationship, a judgment against one can indeed bind both. See id., at 893–895 (describing six categories of qualifying relationships). The State contends that Harris and Bowser, the plaintiffs here, are members of organizations that were plaintiffs in Dickson. And according to North Carolina, that connection prevents the pair from raising anew the questions that the state court previously resolved against those groups. See Brief for Appellants 20–21. But North Carolina never satisfied the District Court that the alleged affiliation really existed. When the State argued that its preclusion theory entitled it to summary judgment, Harris and Bowser responded that they were not members of any of the organizations that had brought the Dickson suit. See 3 Record 1577–1582 (Defs. Motion for Summary Judgment); 4 Record 101–106 (Pls. Opposition to Motion for Summary Judgment). The parties’ dueling contentions turned on intricate issues about those groups’ membership policies (e.g., could Harris’s payment of dues to the national NAACP, or Bowser’s financial contribution to the Mecklenburg County NAACP, have made either a member of the state branch?). Because of those unresolved “factual disputes,” the District Court denied North Carolina’s motion for summary judgment. 4 Record 238 (July 29, 2014 Order). And nothing in the subsequent trial supported the State’s assertion about Harris’s and Bowser’s organizational ties: Indeed, the State chose not to present any further evidence relating to the membership issue. Based on the resulting record, the District Court summarily rejected the State’s claim that Harris and Bowser were something other than independent plaintiffs. See 159 F. Supp. 3d, at 609. That conclusion defeats North Carolina’s attempt to argue for claim or issue preclusion here. We have no basis for assessing the factual assertions underlying the State’s argument any differently than the District Court did. Nothing in the State’s evidence clearly rebuts Harris’s and Bowser’s testimony that they never joined any of the Dickson groups. We need not decide whether the alleged memberships would have supported preclusion if they had been proved. It is enough that the District Court reason-ably thought they had not. The State’s back-up argument about our standard of review also falls short. The rule that we review a trial court’s factual findings for clear error contains no exception for findings that diverge from those made in another court. See Fed. Rule Civ. Proc. 52(a)(6) (“Findings of fact . . . must not be set aside unless clearly erroneous”); see also Hernandez v. New York, 500 U. S. 352, 369 (1991) (plurality opinion) (applying the same standard to a state court’s findings). Whatever findings are under review receive the benefit of deference, without regard to whether a court in a separate suit has seen the matter differently. So here, we must ask not which court considering Districts 1 and 12 had the better view of the facts, but simply whether the court below’s view is clearly wrong. That does not mean the state court’s decision is wholly irrelevant: It is common sense that, all else equal, a finding is more likely to be plainly wrong if some judges disagree with it. Cf. Glossip v. Gross, 576 U. S. ___, ___ (2015) (slip op., at 17) (noting that we are even less likely to disturb a factual determination when “multiple trial courts have reached the same finding”). But the very premise of clear error review is that there are often “two permissible”—because two “plausible”—“views of the evidence.” Anderson, 470 U. S., at 574; see supra, at 4. Even assuming the state court’s findings capture one such view, the District Court’s assessment may yet represent another. And the permissibility of the District Court’s account is the only question before us. III With that out of the way, we turn to the merits of this case, beginning (appropriately enough) with District 1. As noted above, the court below found that race furnished the predominant rationale for that district’s redesign. See supra, at 6–7. And it held that the State’s interest in complying with the VRA could not justify that consideration of race. See supra, at 7. We uphold both conclusions. A Uncontested evidence in the record shows that the State’s mapmakers, in considering District 1, purposefully established a racial target: African-Americans should make up no less than a majority of the voting-age population. See 159 F. Supp. 3d, at 611–614. Senator Rucho and Representative Lewis were not coy in expressing that goal. They repeatedly told their colleagues that District 1 had to be majority-minority, so as to comply with the VRA. During a Senate debate, for example, Rucho explained that District 1 “must include a sufficient number of African-Americans” to make it “a majority black district.” App. 689–690. Similarly, Lewis informed the House and Senate redistricting committees that the district must have “a majority black voting age population.” Id., at 610. And that objective was communicated in no uncertain terms to the legislators’ consultant. Dr. Hofeller testified multiple times at trial that Rucho and Lewis instructed him “to draw [District 1] with a [BVAP] in excess of 50 percent.” 159 F. Supp. 3d, at 613; see, e.g., ibid. (“Once again, my instructions [were] that the district had to be drawn at above 50 percent”). Hofeller followed those directions to the letter, such that the 50%-plus racial target “had a direct and significant impact” on District 1’s configuration. Alabama, 575 U. S., at __ (slip op., at 17). In particular, Hofeller moved the district’s borders to encompass the heavily black parts of Durham (and only those parts), thus taking in tens of thousands of additional African-American voters. That change and similar ones, made (in his words) to ensure that the district’s racial composition would “add[ ] up correctly,” deviated from the districting practices he other-wise would have followed. App. 2802. Hofeller candidly admitted that point: For example, he testified, he sometimes could not respect county or precinct lines as he wished because “the more important thing” was to create a majority-minority district. Id., at 2807; see id., at 2809. The result is a district with stark racial borders: Within the same counties, the portions that fall inside District 1 have black populations two to three times larger than the portions placed in neighboring districts. See Brief for United States as Amicus Curiae 19; cf. Alabama, 575 U. S., at ___–___ (slip op., at 17–18) (relying on similar evidence to find racial predominance). Faced with this body of evidence—showing an announced racial target that subordinated other districting criteria and produced boundaries amplifying divisions between blacks and whites—the District Court did not clearly err in finding that race predominated in drawing District 1. Indeed, as all three judges recognized, the court could hardly have concluded anything but. See 159 F. Supp. 3d, at 611 (calling District 1 a “textbook example” of race-based districting).[3] B The more substantial question is whether District 1 can survive the strict scrutiny applied to racial gerrymanders. As noted earlier, we have long assumed that complying with the VRA is a compelling interest. See supra, at 2. And we have held that race-based districting is narrowly tailored to that objective if a State had “good reasons” for thinking that the Act demanded such steps. See supra, at 3. North Carolina argues that District 1 passes muster under that standard: The General Assembly (so says the State) had “good reasons to believe it needed to draw [District 1] as a majority-minority district to avoid Section 2 liability” for vote dilution. Brief for Appellants 52. We now turn to that defense. This Court identified, in Thornburg v. Gingles, three threshold conditions for proving vote dilution under §2 of the VRA. See 478 U. S., at 50–51. First, a “minority group” must be “sufficiently large and geographically compact to constitute a majority” in some reasonably configured legislative district. Id., at 50. Second, the minority group must be “politically cohesive.” Id., at 51. And third, a district’s white majority must “vote[ ] sufficiently as a bloc” to usually “defeat the minority’s preferred candidate.” Ibid. Those three showings, we have explained, are needed to establish that “the minority [group] has the potential to elect a representative of its own choice” in a possible district, but that racially polarized voting prevents it from doing so in the district as actually drawn because it is “submerg[ed] in a larger white voting population.” Growe v. Emison, 507 U. S. 25, 40 (1993) . If a State has good reason to think that all the “Gingles preconditions” are met, then so too it has good reason to believe that §2 requires drawing a majority-minority district. See Bush v. Vera, 517 U. S. 952, 978 (1996) (plurality opinion). But if not, then not. Here, electoral history provided no evidence that a §2 plaintiff could demonstrate the third Gingles prerequisite—effective white bloc-voting.[4] For most of the twenty years prior to the new plan’s adoption, African-Americans had made up less than a majority of District 1’s voters; the district’s BVAP usually hovered between 46% and 48%. See 159 F. Supp. 3d, at 606; App. 312. Yet throughout those two decades, as the District Court noted, District 1 was “an extraordinarily safe district for African-American preferred candidates.” 159 F. Supp. 3d, at 626. In the closest election during that period, African-Americans’ candidate of choice received 59% of the total vote; in other years, the share of the vote garnered by those candidates rose to as much as 70%. See supra, at 5. Those victories (indeed, landslides) occurred because the district’s white population did not “vote[ ] sufficiently as a bloc” to thwart black voters’ preference, Gingles, 478 U. S., at 51; rather, a meaningful number of white voters joined a politically cohesive black community to elect that group’s favored candidate. In the lingo of voting law, District 1 functioned, election year in and election year out, as a “cross-over” district, in which members of the majority help a “large enough” minority to elect its candidate of choice. Bartlett v. Strickland, 556 U. S. 1, 13 (2009) (plurality opinion). When voters act in that way, “[i]t is difficult to see how the majority-bloc-voting requirement could be met”—and hence how §2 liability could be established. Id., at 16. So experience gave the State no reason to think that the VRA required it to ramp up District 1’s BVAP. The State counters that, in this context, past performance is no guarantee of future results. See Brief for Appellants 57–58; Reply Brief 19–20. Recall here that the State had to redraw its whole congressional map following the 2010 census. See supra, at 5. And in particular, the State had to add nearly 100,000 new people to District 1 to meet the one-person-one-vote standard. See supra, at 6. That meant about 13% of the voters in the new district would never have voted there before. See App. 2690; Reply Brief 20. So, North Carolina contends, the question facing the state mapmakers was not whether the then-existing District 1 violated §2. Rather, the question was whether the future District 1 would do so if drawn without regard to race. And that issue, the State claims, could not be resolved by “focusing myopically on past elections.” Id., at 19. But that reasoning, taken alone, cannot justify North Carolina’s race-based redesign of District 1. True enough, a legislature undertaking a redistricting must assess whether the new districts it contemplates (not the old ones it sheds) conform to the VRA’s requirements. And true too, an inescapable influx of additional voters into a district may suggest the possibility that its former track record of compliance can continue only if the legislature intentionally adjusts its racial composition. Still, North Carolina too far downplays the significance of a longtime pattern of white crossover voting in the area that would form the core of the redrawn District 1. See Gingles, 478 U. S., at 57 (noting that longtime voting patterns are highly probative of racial polarization). And even more important, North Carolina can point to no meaningful legislative inquiry into what it now rightly identifies as the key issue: whether a new, enlarged District 1, created without a focus on race but however else the State would choose, could lead to §2 liability. The prospect of a significant population increase in a district only raises—it does not answer—the question whether §2 requires deliberate measures to augment the district’s BVAP. (Indeed, such population growth could cut in either direction, depending on who comes into the district.) To have a strong basis in evidence to conclude that §2 demands such race-based steps, the State must carefully evaluate whether a plaintiff could establish the Gingles preconditions—including effective white bloc-voting—in a new district created without those measures. We see nothing in the legislative record that fits that description.[5] And that absence is no accident: Rucho and Lewis proceeded under a wholly different theory—arising not from Gingles but from Bartlett v. Strickland—of what §2 demanded in drawing District 1. Strickland involved a geographic area in which African-Americans could not form a majority of a reasonably compact district. See 556 U. S., at 8 (plurality opinion). The African-American community, however, was sizable enough to enable the formation of a crossover district, in which a substantial bloc of black voters, if receiving help from some white ones, could elect the candidates of their choice. See supra, at 14. A plurality of this Court, invoking the first Gingles precondition, held that §2 did not require creating that district: When a minority group is not sufficiently large to make up a majority in a reasonably shaped district, §2 simply does not apply. See 556 U. S., at 18–20. Over and over in the legislative record, Rucho and Lewis cited Strickland as mandating a 50%-plus BVAP in District 1. See App. 355–356, 363–364, 472–474, 609–610, 619, 1044. They apparently reasoned that if, as Strickland held, §2 does not require crossover districts (for groups insufficiently large under Gingles), then §2 also cannot be satisfied by crossover districts (for groups in fact meeting Gingles’ size condition). In effect, they concluded, whenever a legislature can draw a majority-minority district, it must do so—even if a crossover district would also allow the minority group to elect its favored candidates. See 1 Tr. 21–22 (counsel’s explanation that “the [S]tate interpreted” Strickland to say that, in order to protect African-Americans’ electoral strength and thus avoid §2 liability, the BVAP in District 1 “need[ed] to be above 50 percent”). That idea, though, is at war with our §2 jurisprudence—Strickland included. Under the State’s view, the third Gingles condition is no condition at all, because even in the absence of effective white bloc-voting, a §2 claim could succeed in a district (like the old District 1) with an under-50% BVAP. But this Court has made clear that unless each of the three Gingles prerequisites is established, “there neither has been a wrong nor can be a remedy.” Growe, 507 U. S., at 41. And Strickland, far from supporting North Carolina’s view, underscored the necessity of demonstrating effective white bloc-voting to prevail in a §2 vote-dilution suit. The plurality explained that “[i]n areas with substantial crossover voting,” §2 plaintiffs would not “be able to establish the third Gingles precondition” and so “majority-minority districts would not be required.” 556 U. S., at 24; see also ibid. (noting that States can “defend against alleged §2 violations by pointing to crossover voting patterns and to effective crossover districts”). Thus, North Carolina’s belief that it was compelled to redraw District 1 (a successful crossover district) as a majority-minority district rested not on a “strong basis in evidence,” but instead on a pure error of law. Alabama, 575 U. S., at ___ (slip op., at 22). In sum: Although States enjoy leeway to take race-based actions reasonably judged necessary under a proper interpretation of the VRA, that latitude cannot rescue District 1. We by no means “insist that a state legislature, when redistricting, determine precisely what percent minority population [§2 of the VRA] demands.” Ibid. But neither will we approve a racial gerrymander whose necessity is supported by no evidence and whose raison d’être is a legal mistake. Accordingly, we uphold the District Court’s conclusion that North Carolina’s use of race as the predominant factor in designing District 1 does not withstand strict scrutiny. IV We now look west to District 12, making its fifth(!) appearance before this Court. This time, the district’s legality turns, and turns solely, on which of two possible reasons predominantly explains its most recent reconfiguration. The plaintiffs contended at trial that the General Assembly chose voters for District 12, as for District 1, because of their race; more particularly, they urged that the Assembly intentionally increased District 12’s BVAP in the name of ensuring preclearance under the VRA’s §5. But North Carolina declined to mount any defense (similar to the one we have just considered for District 1) that §5’s requirements in fact justified race-based changes to District 12—perhaps because §5 could not reasonably be understood to have done so, see n. 10, infra. Instead, the State altogether denied that racial considerations accounted for (or, indeed, played the slightest role in) District 12’s redesign. According to the State’s version of events, Senator Rucho, Representative Lewis, and Dr. Hofeller moved voters in and out of the district as part of a “strictly” political gerrymander, without regard to race. 6 Record 1011. The mapmakers drew their lines, in other words, to “pack” District 12 with Democrats, not African-Americans. After hearing evidence supporting both parties’ accounts, the District Court accepted the plaintiffs’.[6] Getting to the bottom of a dispute like this one poses special challenges for a trial court. In the more usual case alleging a racial gerrymander—where no one has raised a partisanship defense—the court can make real headway by exploring the challenged district’s conformity to traditional districting principles, such as compactness and respect for county lines. In Shaw II, for example, this Court emphasized the “highly irregular” shape of then-District 12 in concluding that race predominated in its design. 517 U. S., at 905 (internal quotation marks omitted). But such evidence loses much of its value when the State asserts partisanship as a defense, because a bizarre shape—as of the new District 12—can arise from a “political motivation” as well as a racial one. Cromartie I, 526 U. S., at 547, n. 3. And crucially, political and racial reasons are capable of yielding similar oddities in a district’s boundaries. That is because, of course, “racial identification is highly correlated with political affiliation.” Cromartie II, 532 U. S., at 243. As a result of those redistricting realities, a trial court has a formidable task: It must make “a sensitive inquiry” into all “circumstantial and direct evidence of intent” to assess whether the plaintiffs have managed to disentangle race from politics and prove that the former drove a district’s lines. Cromartie I, 526 U. S., at 546 (internal quotation marks omitted).[7] Our job is different—and generally easier. As described earlier, we review a district court’s finding as to racial predominance only for clear error, except when the court made a legal mistake. See supra, at 3–4. Under that standard of review, we affirm the court’s finding so long as it is “plausible”; we reverse only when “left with the definite and firm conviction that a mistake has been committed.” Anderson, 470 U. S., at 573–574; see supra, at 4. And in deciding which side of that line to come down on, we give singular deference to a trial court’s judgments about the credibility of witnesses. See Fed. Rule Civ. Proc. 52(a)(6). That is proper, we have explained, because the various cues that “bear so heavily on the listener’s understanding of and belief in what is said” are lost on an appellate court later sifting through a paper record. Anderson, 470 U. S., at 575.[8] In light of those principles, we uphold the District Court’s finding of racial predominance respecting District 12. The evidence offered at trial, including live witness testimony subject to credibility determinations, adequately supports the conclusion that race, not politics, accounted for the district’s reconfiguration. And no error of law infected that judgment: Contrary to North Carolina’s view, the District Court had no call to dismiss this challenge just because the plaintiffs did not proffer an alternative design for District 12 as circumstantial evidence of the legislature’s intent. A Begin with some facts and figures, showing how the redistricting of District 12 affected its racial composition. As explained above, District 12 (unlike District 1) was approximately the right size as it was: North Carolina did not—indeed, could not—much change its total population. See supra, at 6. But by further slimming the district and adding a couple of knobs to its snakelike body (including in Guilford County), the General Assembly incorporated tens of thousands of new voters and pushed out tens of thousands of old ones. And those changes followed racial lines: To be specific, the new District 12 had 35,000 more African-Americans of voting age and 50,000 fewer whites of that age. (The difference was made up of voters from other racial categories.) See ibid. Those voter exchanges produced a sizable jump in the district’s BVAP, from 43.8% to 50.7%. See ibid. The Assembly thus turned District 12 (as it did District 1, see supra, at 10–11) into a majority-minority district. As the plaintiffs pointed out at trial, Rucho and Lewis had publicly stated that racial considerations lay behind District 12’s augmented BVAP. In a release issued along with their draft districting plan, the two legislators ascribed that change to the need to achieve preclearance of the plan under §5 of the VRA. See App. 358. At that time, §5 covered Guilford County and thus prohibited any “retrogression in the [electoral] position of racial minorities” there. Beer, 425 U. S., at 141; see 31 Fed. Reg. 5081 (1966). And part of Guilford County lay within District 12, which meant that the Department of Justice would closely scrutinize that district’s new lines. In light of those facts, Rucho and Lewis wrote: “Because of the presence of Guilford County in the Twelfth District, we have drawn our proposed Twelfth District at a [BVAP] level that is above the percentage of [BVAP] found in the current Twelfth District.” App. 358. According to the two legislators, that race-based “measure w[ould] ensure preclearance of the plan.” Ibid. Thus, the District Court found, Rucho’s and Lewis’s own account “evince[d] intentionality” as to District 12’s racial composition: Because of the VRA, they increased the number of African-Americans. 159 F. Supp. 3d, at 617. Hofeller confirmed that intent in both deposition testimony and an expert report. Before the redistricting, Hofeller testified, some black residents of Guilford County fell within District 12 while others fell within neighboring District 13. The legislators, he continued, “decided to reunite the black community in Guilford County into the Twelfth.” App. 558; see id., at 530–531. Why? Hofeller responded, in language the District Court emphasized: “[I]n order to be cautious and draw a plan that would pass muster under the Voting Rights Act.” Id., at 558; see 159 F. Supp. 3d, at 619. Likewise, Hofeller’s expert report highlighted the role of the VRA in altering District 12’s lines. “[M]indful that Guilford County was covered” by §5, Hofeller explained, the legislature “determined that it was prudent to reunify [the county’s] African-American community” into District 12. App. 1103. That change caused the district’s compactness to decrease (in expert-speak, it “lowered the Reock Score”), but that was a sacrifice well worth making: It would “avoid the possibility of a [VRA] charge” that would “inhibit[ ] preclearance.” Ibid. The State’s preclearance submission to the Justice Department indicated a similar determination to concentrate black voters in District 12. “One of the concerns of the Redistricting Chairs,” North Carolina there noted, had to do with the Justice Department’s years-old objection to “a failure by the State to create a second majority minority district” (that is, in addition to District 1). Id., at 478. The submission then went on to explain that after considering alternatives, the redistricters had designed a version of District 12 that would raise its BVAP to 50.7%. Thus, concluded the State, the new District 12 “increases[ ] the African-American community’s ability to elect their candidate of choice.” Id., at 479. In the District Court’s view, that passage once again indicated that making District 12 majority-minority was no “mere coincidence,” but a deliberate attempt to avoid perceived obstacles to preclearance. 159 F. Supp. 3d, at 617.[9] And still there was more: Perhaps the most dramatic testimony in the trial came when Congressman Mel Watt (who had represented District 12 for some 20 years) recounted a conversation he had with Rucho in 2011 about the district’s future make-up. According to Watt, Rucho said that “his leadership had told him that he had to ramp the minority percentage in [District 12] up to over 50 percent to comply with the Voting Rights Law.” App. 2369; see id., at 2393. And further, that it would then be Rucho’s “job to go and convince the African-American community” that such a racial target “made sense” under the Act. Ibid.; see id., at 2369.[10] The District Court cred-ited Watt’s testimony about the conversation, citing his courtroom demeanor and “consistent recollection” under “probing cross-examination.” 159 F. Supp. 3d, at 617–618.[11] In the court’s view, Watt’s account was of a piece with all the other evidence—including the redistricters’ on-the-nose attainment of a 50% BVAP—indicating that the General Assembly, in the name of VRA compliance, deliberately redrew District 12 as a majority-minority district. See id., at 618.[12] The State’s contrary story—that politics alone drove decisionmaking—came into the trial mostly through Hofeller’s testimony. Hofeller explained that Rucho and Lewis instructed him, first and foremost, to make the map as a whole “more favorable to Republican candidates.” App. 2682. One agreed-on stratagem in that effort was to pack the historically Democratic District 12 with even more Democratic voters, thus leaving surrounding districts more reliably Republican. See id., at 2682–2683, 2696–2697. To that end, Hofeller recounted, he drew District 12’s new boundaries based on political data—specifically, the voting behavior of precincts in the 2008 Presidential election between Barack Obama and John McCain. See id., at 2701–2702. Indeed, he claimed, he displayed only this data, and no racial data, on his computer screen while mapping the district. See id., at 2721. In part of his testimony, Hofeller further stated that the Obama-McCain election data explained (among other things) his incorporation of the black, but not the white, parts of Guilford County then located in District 13. See id., at 2824. Only after he drew a politics-based line between those adjacent areas, Hofeller testified, did he “check[ ]” the racial data and “f[ind] out” that the resulting configuration of District 12 “did not have a [§5] issue.” Id., at 2822. The District Court, however, disbelieved Hofeller’s asserted indifference to the new district’s racial composition. The court recalled Hofeller’s contrary deposition testimony—his statement (repeated in only slightly different words in his expert report) that Rucho and Lewis “decided” to shift African-American voters into District 12 “in order to” ensure preclearance under §5. See 159 F. Supp. 3d, at 619–620; App. 558. And the court explained that even at trial, Hofeller had given testimony that undermined his “blame it on politics” claim. Right after asserting that Rucho and Lewis had told him “[not] to use race” in designing District 12, Hofeller added a qualification: “except perhaps with regard to Guilford County.” Id., at 2791; see id., at 2790. As the District Court understood, that is the kind of “exception” that goes pretty far toward swallowing the rule. District 12 saw a net increase of more than 25,000 black voters in Guilford County, relative to a net gain of fewer than 35,000 across the district: So the newly added parts of that county played a major role in pushing the district’s BVAP over 50%. See id., at 384, 500–502.[13] The District Court came away from Hofeller’s self-contradictory testimony unpersuaded that this decisive influx of black voters was an accident. Whether the racial make-up of the county was displayed on his computer screen or just fixed in his head, the court thought, Hofeller’s denial of race-based districting “r[ang] hollow.” 159 F. Supp. 3d, at 620, n. 8. Finally, an expert report by Dr. Stephen Ansolabehere lent circumstantial support to the plaintiffs’ race-not-politics case. Ansolabehere looked at the six counties overlapping with District 12—essentially the region from which the mapmakers could have drawn the district’s population. The question he asked was: Who from those counties actually ended up in District 12? The answer he found was: Only 16% of the region’s white registered voters, but 64% of the black ones. See App. 321–322. Ansolabehere next controlled for party registration, but discovered that doing so made essentially no difference: For example, only 18% of the region’s white Democrats wound up in District 12, whereas 65% of the black Democrats did. See id., at 332. The upshot was that, regardless of party, a black voter was three to four times more likely than a white voter to cast his ballot within District 12’s borders. See ibid. Those stark disparities led Ansolabehere to conclude that “race, and not party,” was “the dominant factor” in District 12’s design. Id., at 337.[14] His report, as the District Court held, thus tended to confirm the plaintiffs’ direct evidence of racial predominance. See 159 F. Supp. 3d, at 620–621. The District Court’s assessment that all this evidence proved racial predominance clears the bar of clear error review. The court emphasized that the districting plan’s own architects had repeatedly described the influx of African-Americans into District 12 as a §5 compliance measure, not a side-effect of political gerrymandering. And those contemporaneous descriptions comported with the court’s credibility determinations about the trial testimony—that Watt told the truth when he recounted Rucho’s resolve to hit a majority-BVAP target; and conversely that Hofeller skirted the truth (especially as to Guilford County) when he claimed to have followed only race-blind criteria in drawing district lines. We cannot disrespect such credibility judgments. See Anderson, 470 U. S., at 575 (A choice to believe “one of two or more witnesses, each of whom has told a coherent and facially plausible story that is not contradicted by extrinsic evidence,” can “virtually never be clear error”). And more generally, we will not take it upon ourselves to weigh the trial evidence as if we were the first to hear it. See id., at 573 (A “reviewing court oversteps” under Rule 52(a) “if it undertakes to duplicate the role of the lower court”). No doubt other interpretations of that evidence were permissible. Maybe we would have evaluated the testimony differently had we presided over the trial; or then again, maybe we would not have. Either way—and it is only this which matters—we are far from having a “definite and firm conviction” that the District Court made a mistake in concluding from the record before it that racial considerations predominated in District 12’s design. B The State mounts a final, legal rather than factual, attack on the District Court’s finding of racial predominance. When race and politics are competing explanations of a district’s lines, argues North Carolina, the party challenging the district must introduce a particular kind of circumstantial evidence: “an alternative [map] that achieves the legislature’s political objectives while improving racial balance.” Brief for Appellants 31 (emphasis deleted). That is true, the State says, irrespective of what other evidence is in the case—so even if the plaintiff offers powerful direct proof that the legislature adopted the map it did for racial reasons. See Tr. of Oral Arg. 8. Because the plaintiffs here (as all agree) did not present such a counter-map, North Carolina concludes that they cannot prevail. The dissent echoes that argument. See post, at 6–11. We have no doubt that an alternative districting plan, of the kind North Carolina describes, can serve as key evidence in a race-versus-politics dispute. One, often highly persuasive way to disprove a State’s contention that politics drove a district’s lines is to show that the legislature had the capacity to accomplish all its partisan goals without moving so many members of a minority group into the district. If you were really sorting by political behavior instead of skin color (so the argument goes) you would have done—or, at least, could just as well have done—this. Such would-have, could-have, and (to round out the set) should-have arguments are a familiar means of undermining a claim that an action was based on a permissible, rather than a prohibited, ground. See, e.g., Miller-El v. Dretke, 545 U. S. 231, 249 (2005) (“If that were the [real] explanation for striking [juror] Warren[,] the prosecutors should have struck [juror] Jenkins” too). But they are hardly the only means. Suppose that the plaintiff in a dispute like this one introduced scores of leaked emails from state officials instructing their mapmaker to pack as many black voters as possible into a district, or telling him to make sure its BVAP hit 75%. Based on such evidence, a court could find that racial rather than political factors predominated in a district’s design, with or without an alternative map. And so too in cases lacking that kind of smoking gun, as long as the evidence offered satisfies the plaintiff’s burden of proof. In Bush v. Vera, for example, this Court upheld a finding of racial predominance based on “substantial direct evidence of the legislature’s racial motivations”—including credible testimony from political figures and statements made in a §5 preclearance submission—plus circumstantial evidence that redistricters had access to racial, but not political, data at the “block-by-block level” needed to explain their “intricate” designs. See 517 U. S., at 960–963 (plurality opinion). Not a single Member of the Court thought that the absence of a counter-map made any difference. Similarly, it does not matter in this case, where the plaintiffs’ introduction of mostly direct and some circumstantial evidence—documents issued in the redistricting process, testimony of government officials, expert analysis of demographic patterns—gave the District Court a sufficient basis, sans any map, to resolve the race-or-politics question. A plaintiff’s task, in other words, is simply to persuade the trial court—without any special evidentiary prerequisite—that race (not politics) was the “predominant consideration in deciding to place a significant number of voters within or without a particular district.” Alabama, 575 U. S., at ___ (slip op., at 4) (internal quotation marks omitted); cf. Bethune-Hill, 580 U. S., at ___, ___ (slip op., at 8, 10) (rejecting a similar effort to elevate one form of “persuasive circumstantial evidence” in a dispute respecting racial predominance to a “mandatory precondition” or “threshold requirement” of proof). That burden of proof, we have often held, is “demanding.” E.g., Cromartie II, 532 U. S., at 241. And because that is so, a plaintiff will sometimes need an alternative map, as a practical matter, to make his case. But in no area of our equal protection law have we forced plaintiffs to submit one particular form of proof to prevail. See Arlington Heights v. Metropolitan Housing Development Corp., 429 U. S. 252 –268 (1977) (offering a varied and non-exhaustive list of “subjects of proper inquiry in determining whether racially discriminatory intent existed”). Nor would it make sense to do so here. The Equal Protection Clause prohibits the unjustified drawing of district lines based on race. An alternative map is merely an evidentiary tool to show that such a substantive violation has occurred; neither its presence nor its absence can itself resolve a racial gerrymandering claim.[15] North Carolina insists, however, that we have already said to the contrary—more particularly, that our decision in Cromartie II imposed a non-negotiable “alternative-map requirement.” Brief for Appellants 31. As the State observes, Cromartie II reversed as clearly erroneous a trial court’s finding that race, rather than politics, predominated in the assignment of voters to an earlier incarnation of District 12. See 532 U. S., at 241; supra, at 5. And as the State emphasizes, a part of our opinion faulted the Cromartie plaintiffs for failing to offer a convincing account of how the legislature could have accomplished its political goals other than through the map it chose. See 532 U. S., at 257–258. We there stated: “In a case such as this one where majority-minority districts . . . are at issue and where racial identification correlates highly with political affiliation, the party attacking the legislatively drawn boundaries must show at the least that the legislature could have achieved its legitimate political objectives in alternative ways that are comparably consistent with traditional districting principles. That party must also show that those districting alternatives would have brought about significantly greater racial balance.” Id., at 258. According to North Carolina, that passage alone settles this case, because it makes an alternative map “essential” to a finding that District 12 (a majority-minority district in which race and partisanship are correlated) was a racial gerrymander. Reply Brief 11. Once again, the dissent says the same. See post, at 7. But the reasoning of Cromartie II belies that reading. The Court’s opinion nowhere attempts to explicate or justify the categorical rule that the State claims to find there. (Certainly the dissent’s current defense of that rule, see post, at 8–11, was nowhere in evidence.) And given the strangeness of that rule—which would treat a mere form of evidence as the very substance of a constitutional claim, see supra, at 29–31—we cannot think that the Court adopted it without any explanation. Still more, the entire thrust of the Cromartie II opinion runs counter to an inflexible counter-map requirement. If the Court had adopted that rule, it would have had no need to weigh each piece of evidence in the case and determine whether, taken together, they were “adequate” to show “the predominance of race in the legislature’s line-drawing process.” 532 U. S., at 243–244. But that is exactly what Cromartie II did, over a span of 20 pages and in exhaustive detail. Item by item, the Court discussed and dismantled the supposed proof, both direct and circumstantial, of race-based redistricting. All that careful analysis would have been superfluous—that dogged effort wasted—if the Court viewed the absence or inadequacy of a single form of evidence as necessarily dooming a gerrymandering claim. Rightly understood, the passage from Cromartie II had a different and narrower point, arising from and reflecting the evidence offered in that case. The direct evidence of a racial gerrymander, we thought, was extremely weak: We said of one piece that it “says little or nothing about whether race played a predominant role” in drawing district lines; we said of another that it “is less persuasive than the kinds of direct evidence we have found significant in other redistricting cases.” Id., at 253–254 (emphasis deleted). Nor did the report of the plaintiffs’ expert impress us overmuch: In our view, it “offer[ed] little insight into the legislature’s true motive.” Id., at 248. That left a set of arguments of the would-have-could-have variety. For example, the plaintiffs offered several maps purporting to “show how the legislature might have swapped” some mostly black and mostly white precincts to obtain greater racial balance “without harming [the legislature’s] political objective.” Id., at 255 (internal quotation marks omitted). But the Court determined that none of those proposed exchanges would have worked as advertised—essentially, that the plaintiffs’ “you could have redistricted differently” arguments failed on their own terms. See id., at 254–257. Hence emerged the demand quoted above, for maps that would actually show what the plaintiffs’ had not. In a case like Cromartie II—that is, one in which the plaintiffs had meager direct evidence of a racial gerrymander and needed to rely on evidence of forgone alternatives—only maps of that kind could carry the day. Id., at 258. But this case is most unlike Cromartie II, even though it involves the same electoral district some twenty years on. This case turned not on the possibility of creating more optimally constructed districts, but on direct evidence of the General Assembly’s intent in creating the actual District 12, including many hours of trial testimony subject to credibility determinations. That evidence, the District Court plausibly found, itself satisfied the plaintiffs’ burden of debunking North Carolina’s “it was really politics” defense; there was no need for an alternative map to do the same job. And we pay our precedents no respect when we extend them far beyond the circumstances for which they were designed. V Applying a clear error standard, we uphold the District Court’s conclusions that racial considerations predominated in designing both District 1 and District 12. For District 12, that is all we must do, because North Carolina has made no attempt to justify race-based districting there. For District 1, we further uphold the District Court’s decision that §2 of the VRA gave North Carolina no good reason to reshuffle voters because of their race. We accordingly affirm the judgment of the District Court. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case. Congressional District 1 (Enacted 2011) Congressional District 12 (Enacted 2011)Notes 1 A plaintiff succeeds at this stage even if the evidence reveals that a legislature elevated race to the predominant criterion in order to advance other goals, including political ones. See Bush v. Vera, 517 U. S. 952 –970 (1996) (plurality opinion) (holding that race predominated when a legislature deliberately “spread[ ] the Black population” among several districts in an effort to “protect[ ] Democratic incumbents”); Miller v. Johnson, 515 U. S. 900, 914 (1995) (stating that the “use of race as a proxy” for “political interest[s]” is “prohibit[ed]”). 2 Challenges to the constitutionality of congressional districts are heard by three-judge district courts, with a right of direct appeal to this Court. See 28 U. S. C. §§2284(a), 1253. 3 The State’s argument to the contrary rests on a legal proposition that was foreclosed almost as soon as it was raised in this Court. According to the State, racial considerations cannot predominate in drawing district lines unless there is an “actual conflict” between those lines and “traditional districting principles.” Brief for Appellants 45. But we rejected that view earlier this Term, holding that when (as here) race furnished “the overriding reason for choosing one map over others,” a further showing of “inconsistency between the enacted plan and traditional redistricting criteria” is unnecessary to a finding of racial predominance. Bethune-Hill v. Virginia State Bd. of Elections, 580 U. S. ___, ___ (2017) (slip op., at 10). And in any event, the evidence recounted in the text indicates that District 1’s boundaries did conflict with traditional districting principles—for example, by splitting numerous counties and precincts. See supra, at 11. So we would uphold the District Court’s finding of racial predominance even under the (incorrect) legal standard the State proposes. 4 In the District Court, the parties also presented arguments relating to the first Gingles prerequisite, contesting whether the African-American community in the region was sufficiently large and compact to form a majority of a reasonably shaped district. The court chose not to decide that fact-intensive question. And aside from the State’s unelaborated assertion that “[t]here is no question that the first factor was satisfied,” Brief for Appellants 52, the parties have not briefed or argued the issue before us. We therefore have no occasion to address it. 5 North Carolina calls our attention to two expert reports on voting patterns throughout the State, but neither casts light on the relevant issue. The first (by Dr. Thomas Brunell) showed that some elections in many of the State’s counties exhibited “statistically significant” racially polarized voting. App. 1001. The second (by Dr. Ray Block) found that in various elections across the State, white voters were “noticeably” less likely than black voters to support black candidates. Id., at 959. From those far-flung data points—themselves based only on past elections—the experts opined (to no one’s great surprise) that in North Carolina, as in most States, there are discernible, non-random relationships between race and voting. But as the District Court found, see Harris v. McCrory, 159 F. Supp. 3d 600, 624 (MDNC 2016), that generalized conclusion fails to meaningfully (or indeed, at all) address the relevant local question: whether, in a new version of District 1 created without a focus on race, black voters would encounter “sufficient[ ]” white bloc-voting to “cancel [their] ability to elect representatives of their choice,” Gingles, 478 U. S., at 56. And so the reports do not answer whether the legislature needed to boost District 1’s BVAP to avoid potential §2 liability. 6 Justice Alito charges us with “ignor[ing]” the State’s political-gerrymander defense, making our analysis “like Hamlet without the prince.” Post, at 20 (opinion concurring in judgment in part and dissenting in part) (hereinafter dissent); see post, at 20, 34. But we simply take the State’s account for what it is: one side of a thoroughly two-sided case (and, as we will discuss, the side the District Court rejected, primarily on factual grounds). By contrast, the dissent consistently treats the State’s version of events (what it calls “the Legislature’s political strategy and the relationship between that strategy and [District 12’s] racial composition,” post, at 20) as if it were a simple “fact of the matter”—the premise of, rather than a contested claim in, this case. See post, at 12–14, 16, 20, 26, 27–29, 33. The dissent’s narrative thus tracks, top-to-bottom and point-for-point, the testimony of Dr. Hofeller, the State’s star witness at trial—so much so that the dissent could just have block-quoted that portion of the transcript and saved itself a fair bit of trouble. Compare post, at 12–20, with App. 2671–2755. Imagine (to update the dissent’s theatrical reference) Inherit the Wind retold solely from the perspective of William Jennings Bryan, with nary a thought given to the competing viewpoint of Clarence Darrow. 7 As earlier noted, that inquiry is satisfied when legislators have “place[d] a significant number of voters within or without” a district predominantly because of their race, regardless of their ultimate objective in taking that step. See supra, at 2, and n. 1. So, for example, if legislators use race as their predominant districting criterion with the end goal of advancing their partisan interests—perhaps thinking that a proposed district is more “sellable” as a race-based VRA compliance measure than as a political gerrymander and will accomplish much the same thing—their action still triggers strict scrutiny. See Vera, 517 U. S., at 968–970 (plurality opinion). In other words, the sorting of voters on the grounds of their race remains suspect even if race is meant to function as a proxy for other (including political) characteristics. See Miller, 515 U. S., at 914. 8 Undeterred by these settled principles, the dissent undertakes to refind the facts of this case at every turn. See post, at 11–33. Indeed, the dissent repeatedly flips the appropriate standard of review—arguing, for example, that the District Court’s is not “the only plausible interpretation” of one piece of contested evidence and that the State offered an “entirely natural” view of another. Post, at 24, 31; see also post, at 20, 26, 27, 33. Underlying that approach to the District Court’s factfinding is an elemental error: The dissent mistakes the rule that a legislature’s good faith should be presumed “until a claimant makes a showing sufficient to support th[e] allegation” of “race-based decisionmaking,” Miller, 515 U. S., at 915, for a kind of super-charged, pro-State presumption on appeal, trumping clear-error review. See post, at 11–12, n. 7. 9 The dissent’s contrary reading of the preclearance submission—as reporting the redistricters’ “decis[ion] not to construct District 12 as a majority-minority district,” post, at 24—is difficult to fathom. The language the dissent cites explains only why Rucho and Lewis rejected one particular way of creating such a district; the submission then relates their alternative (and, of course, successful) approach to attaining an over-50% BVAP. See App. 478–479. 10 Watt recalled that he laughed in response because the VRA required no such target. See id., at 2369. And he told Rucho that “the African-American community will laugh at you” too. Ibid. Watt explained to Rucho: “I’m getting 65 percent of the vote in a 40 percent black district. If you ramp my [BVAP] to over 50 percent, I’ll probably get 80 percent of the vote, and[ ] that’s not what the Voting Rights Act was designed to do.” Ibid. 11 The court acknowledged that, in the earlier state-court trial involving District 12, Rucho denied making the comments that Watt recalled. See 159 F. Supp. 3d, at 617–618. But the court explained that it could not “assess [the] credibility” of Rucho’s contrary account because even though he was listed as a defense witness and present in the courtroom throughout the trial, the State chose not to put him on the witness stand. Id., at 618. 12 The dissent conjures a different way of explaining Watt’s testi-mony. Perhaps, the dissent suggests, Rucho disclosed a majority-minority target to Watt, but Watt then changed Rucho’s mind—and perhaps it was just a coincidence (or a mistake?) that Rucho still created a 50.7%-BVAP district. See post, at 25–26. But nothing in the record supports that hypothesis. See ibid. (relying exclusively on the State’s preclearance submission to back up this story); supra, at 23, and n. 9 (correcting the dissent’s misreading of that submission). And the State, lacking the dissent’s creativity, did not think to present it at trial. 13 The dissent charges that this comparison is misleading, but offers no good reason why that is so. See post, at 29–30. It is quite true, as the dissent notes, that another part of District 12 (in Mecklenburg County) experienced a net increase in black voters even larger than the one in Guilford County. See post, at 30. (The net increases in the two counties thus totaled more than 35,000; they were then partially offset by net decreases in other counties in District 12.) But that is irrelevant to the point made here: Without the numerous black voters added to District 12 in Guilford County—where the evidence most clearly indicates voters were chosen based on race—the district would have fallen well shy of majority-minority status. 14 Hofeller did not dispute Ansolabehere’s figures, but questioned his inference. Those striking patterns, the mapmaker claimed, were nothing more than the result of his own reliance on voting data from the 2008 Presidential election—because that information (i.e., who voted for Obama and who for McCain) tracked race better than it did party registration. See App. 1101, 1111–1114; cf. Cromartie II, 532 U. S. 234, 245 (2001) (recognizing that “party registration and party preference do not always correspond”). As we have just recounted, however, the District Court had other reasons to disbelieve Hofeller’s testimony that he used solely that electoral data to draw District 12’s lines. See supra, at 25–26. And Ansolabehere contended that even if Hofeller did so, that choice of data could itself suggest an intent to sort voters by race. Voting results from a “single [Presidential] election with a Black candidate,” Ansolabehere explained, would be a “problematic and unusual” indicator of future party preference, because of the racial dynamics peculiar to such a match-up. App. 341; see id., at 342–343. That data would, indeed, be much more useful as a reflection of an area’s racial composition: “The Obama vote,” Ansolabehere found, is “an extremely strong positive indicator of the location of Black registered voters” and, conversely, an “extremely strong negative indicator of the location of White registered voters.” Id., at 342; see id., at 2546–2550. 15 The dissent responds that an alternative-map requirement “should not be too hard” for plaintiffs (or at least “sophisticated” litigants “like those in the present case”) to meet. Post, at 10–11. But if the plaintiffs have already proved by a preponderance of the evidence that race predominated in drawing district lines, then we have no warrant to demand that they jump through additional evidentiary hoops (whether the exercise would cost a hundred dollars or a million, a week’s more time or a year’s). Or at least that would be so if we followed the usual rules. Underlying the dissent’s view that we should not—that we should instead create a special evidentiary burden—is its belief that “litigation of this sort” often seeks to “obtain in court what [a political party] could not achieve in the political arena,” post, at 9, and so that little is lost by making suits like this one as hard as possible. But whatever the possible motivations for bringing such suits (and the dissent says it is not questioning “what occurred here,” ibid.), they serve to prevent legislatures from taking unconstitutional districting action—which happens more often than the dissent must suppose. State lawmakers sometimes misunderstand the VRA’s requirements (as may have occurred here with respect to §5), leading them to employ race as a predominant districting criterion when they should not. See supra, at 22–24, and n. 10. Or they may resort to race-based districting for ultimately political reasons, leveraging the strong correlation between race and voting behavior to advance their partisan interests. See nn. 1, 7, supra. Or, finally—though we hope less commonly—they may simply seek to suppress the electoral power of minority voters. When plaintiffs meet their burden of showing that such conduct has occurred, there is no basis for subjecting them to additional—and unique—evidentiary hurdles, preventing them from receiving the remedy to which they are entitled.

581.US.2016_16-369

The Los Angeles County Sheriff’s Department received word from a confidential informant that a potentially armed and dangerous parolee-at-large had been seen at a certain residence. While other officers searched the main house, Deputies Conley and Pederson searched the back of the property where, unbeknownst to the deputies, respondents Mendez and Garcia were napping inside a shack where they lived. Without a search warrant and without announcing their presence, the deputies opened the door of the shack. Mendez rose from the bed, holding a BB gun that he used to kill pests. Deputy Conley yelled, “Gun!” and the deputies immediately opened fire, shooting Mendez and Garcia multiple times. Officers did not find the parolee in the shack or elsewhere on the property. Mendez and Garcia sued Deputies Conley and Pederson and the County under 42 U. S. C. §1983, pressing three Fourth Amendment claims: a warrantless entry claim, a knock-and-announce claim, and an excessive force claim. On the first two claims, the District Court awarded Mendez and Garcia nominal damages. On the excessive force claim, the court found that the deputies’ use of force was reasonable under Graham v. Connor, 490 U. S. 386 , but held them liable nonetheless under the Ninth Circuit’s provocation rule, which makes an officer’s otherwise reasonable use of force unreasonable if (1) the officer “intentionally or recklessly provokes a violent confrontation” and (2) “the provocation is an independent Fourth Amendment violation,” Billington v. Smith, 292 F. 3d 1177, 1189. On appeal, the Ninth Circuit held that the officers were entitled to qualified immunity on the knock-and-announce claim and that the warrantless entry violated clearly established law. It also affirmed the District Court’s application of the provocation rule, and held, in the alternative, that basic notions of proximate cause would support liability even without the provocation rule. Held: The Fourth Amendment provides no basis for the Ninth Circuit’s “provocation rule.” Pp. 5–10. (a) The provocation rule is incompatible with this Court’s excessive force jurisprudence, which sets forth a settled and exclusive framework for analyzing whether the force used in making a seizure complies with the Fourth Amendment. See Graham, supra, at 395. The operative question in such cases is “whether the totality of the circumstances justifie[s] a particular sort of search or seizure.” Tennessee v. Garner, 471 U. S. 1 –9. When an officer carries out a seizure that is reasonable, taking into account all relevant circumstances, there is no valid excessive force claim. The provocation rule, however, instructs courts to look back in time to see if a different Fourth Amendment violation was somehow tied to the eventual use of force, an approach that mistakenly conflates distinct Fourth Amendment claims. The proper framework is set out in Graham. To the extent that a plaintiff has other Fourth Amendment claims, they should be analyzed separately. The Ninth Circuit attempts to cabin the provocation rule by defining a two-prong test: First, the separate constitutional violation must “creat[e] a situation which led to” the use of force; and second, the separate constitutional violation must be committed recklessly or intentionally. 815 F. 3d 1178, 1193. Neither limitation, however, solves the fundamental problem: namely, that the provocation rule is an unwarranted and illogical expansion of Graham. In addition, each limitation creates problems of its own. First, the rule relies on a vague causal standard. Second, while the reasonableness of a search or seizure is almost always based on objective factors, the provocation rule looks to the subjective intent of the officers who carried out the seizure. There is no need to distort the excessive force inquiry in this way in order to hold law enforcement officers liable for the foreseeable consequences of all their constitutional torts. Plaintiffs can, subject to qualified immunity, generally recover damages that are proximately caused by any Fourth Amendment violation. See, e.g., Heck v. Humphrey, 512 U. S. 477 . Here, if respondents cannot recover on their excessive force claim, that will not foreclose recovery for injuries proximately caused by the warrantless entry. Pp. 5–10. (b) The Ninth Circuit’s proximate-cause holding is similarly tainted. Its analysis appears to focus solely on the risks foreseeably associated with the failure to knock and announce—the claim on which the court concluded that the deputies had qualified immunity—rather than the warrantless entry. On remand, the court should revisit the question whether proximate cause permits respondents to recover damages for their injuries based on the deputies’ failure to secure a warrant at the outset. Pp. 10–11. 815 F. 3d 1178, vacated and remanded. Alito, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case.

If law enforcement officers make a “seizure” of a person using force that is judged to be reasonable based on a consideration of the circumstances relevant to that determination, may the officers nevertheless be held liable for injuries caused by the seizure on the ground that they committed a separate Fourth Amendment violation that contributed to their need to use force? The Ninth Circuit has adopted a “provocation rule” that imposes liability in such a situation. We hold that the Fourth Amendment provides no basis for such a rule. A different Fourth Amendment violation cannot transform a later, reasonable use of force into an unreasonable seizure. I A In October 2010, deputies from the Los Angeles County Sheriff’s Department were searching for a parolee-at-large named Ronnie O’Dell. A felony arrest warrant had been issued for O’Dell, who was believed to be armed and dangerous and had previously evaded capture. Findings of Fact and Conclusions of Law, No. 2:11–cv–04771 (CD Cal.), App. to Pet. for Cert. 56a, 64a. Deputies Christopher Conley and Jennifer Pederson were assigned to assist the task force searching for O’Dell. Id., at 57a–58a. The task force received word from a confidential informant that O’Dell had been seen on a bicycle at a home in Lancaster, California, owned by Paula Hughes, and the officers then mapped out a plan for apprehending O’Dell. Id., at 58a. Some officers would approach the front door of the Hughes residence, while Deputies Conley and Pederson would search the rear of the property and cover the back door of the residence. Id., at 59a. During this briefing, it was announced that a man named Angel Mendez lived in the backyard of the Hughes home with a pregnant woman named Jennifer Garcia (now Mrs. Jennifer Mendez). Ibid. Deputy Pederson heard this announcement, but at trial Deputy Conley testified that he did not remember it. Ibid. When the officers reached the Hughes residence around midday, three of them knocked on the front door while Deputies Conley and Pederson went to the back of the property. Id., at 63a. At the front door, Hughes asked if the officers had a warrant. Ibid. A sergeant responded that they did not but were searching for O’Dell and had a warrant for his arrest. Ibid. One of the officers heard what he thought were sounds of someone running inside the house. Id., at 64a. As the officers prepared to open the door by force, Hughes opened the door and informed them that O’Dell was not in the house. Ibid. She was placed under arrest, and the house was searched, but O’Dell was not found. Ibid. Meanwhile, Deputies Conley and Pederson, with guns drawn, searched the rear of the residence, which was cluttered with debris and abandoned automobiles. Id., at 60a, 65a. The property included three metal storage sheds and a one-room shack made of wood and plywood. Id., at 60a. Mendez had built the shack, and he and Garcia had lived inside for about 10 months. Id., at 61a. The shack had a single doorway covered by a blue blanket. Ibid. Amid the debris on the ground, an electrical cord ran into the shack, and an air conditioner was mounted on the side. Id., at 62a. A gym storage locker and clothes and other possessions were nearby. Id., at 61a. Mendez kept a BB rifle in the shack for use on rats and other pests. Id., at 62a. The BB gun “closely resembled a small caliber rifle.” Ibid. Deputies Conley and Pederson first checked the three metal sheds and found no one inside. Id., at 65a. They then approached the door of the shack. Id., at 66a. Unbeknownst to the officers, Mendez and Garcia were in the shack and were napping on a futon. Id., at 67a. The deputies did not have a search warrant and did not knock and announce their presence. Id., at 66a. When Deputy Conley opened the wooden door and pulled back the blanket, Mendez thought it was Ms. Hughes and rose from the bed, picking up the BB gun so he could stand up and place it on the floor. Id., at 68a. As a result, when the deputies entered, he was holding the BB gun, and it was “point[ing] somewhat south towards Deputy Conley.” Id., at 69a. Deputy Conley yelled, “Gun!” and the deputies immediately opened fire, discharging a total of 15 rounds. Id., at 69a–70a. Mendez and Garcia “were shot multiple times and suffered severe injuries,” and Mendez’s right leg was later amputated below the knee. Id., at 70a. O’Dell was not in the shack or anywhere on the property. Ibid. B Mendez and his wife (respondents here) filed suit under Rev. Stat. §1976, 42 U. S. C. §1983, against petitioners, the County of Los Angeles and Deputies Conley and Pederson. As relevant here, they pressed three Fourth Amendment claims. First, they claimed that the deputies executed an unreasonable search by entering the shack without a warrant (the “warrantless entry claim”); second, they asserted that the deputies performed an unreason-able search because they failed to announce their presence before entering the shack (the “knock-and-announce claim”); and third, they claimed that the deputies effected an unreasonable seizure by deploying excessive force in opening fire after entering the shack (the “excessive force claim”). After a bench trial, the District Court ruled largely in favor of respondents. App. to Pet. for Cert. 135a–136a. The court found Deputy Conley liable on the warrantless entry claim, and the court also found both deputies liable on the knock-and-announce claim. But the court awarded nominal damages for these violations because “the act of pointing the BB gun” was a superseding cause “as far as damage [from the shooting was] concerned.” App. 238. The District Court then addressed respondents’ excessive force claim. App. to Pet. for Cert. 105a–127a. The court began by evaluating whether the deputies used excessive force under Graham v. Connor, 490 U. S. 386 (1989) . The court held that, under Graham, the deputies’ use of force was reasonable “given their belief that a man was holding a firearm rifle threatening their lives.” App. to Pet. for Cert. 108a. But the court did not end its excessive force analysis at this point. Instead, the court turned to the Ninth Circuit’s provocation rule, which holds that “an officer’s otherwise reasonable (and lawful) defensive use of force is unreasonable as a matter of law, if (1) the officer intentionally or recklessly provoked a violent response, and (2) that provocation is an independent constitutional violation.” Id., at 111a. Based on this rule, the District Court held the deputies liable for excessive force and awarded respondents around $4 million in damages. Id., at 135a–136a. The Court of Appeals affirmed in part and reversed in part. 815 F. 3d 1178 (CA9 2016). Contrary to the District Court, the Court of Appeals held that the officers were entitled to qualified immunity on the knock-and-announce claim. Id., at 1191–1193. But the court concluded that the warrantless entry of the shack violated clearly established law and was attributable to both deputies. Id., at 1191, 1195. Finally, and most important for present purposes, the court affirmed the application of the provocation rule. The Court of Appeals did not disagree with the conclusion that the shooting was reasonable under Graham; instead, like the District Court, the Court of Appeals applied the provocation rule and held the deputies liable for the use of force on the theory that they had intentionally and recklessly brought about the shooting by entering the shack without a warrant in violation of clearly established law. 815 F. 3d, at 1193. The Court of Appeals also adopted an alternative rationale for its judgment. It held that “basic notions of proximate cause” would support liability even without the provocation rule because it was “reasonably foreseeable” that the officers would meet an armed homeowner when they “barged into the shack unannounced.” Id., at 1194–1195. We granted certiorari. 580 U. S. ___ (2016). II The Ninth Circuit’s provocation rule permits an excessive force claim under the Fourth Amendment “where an officer intentionally or recklessly provokes a violent confrontation, if the provocation is an independent Fourth Amendment violation.” Billington v. Smith, 292 F. 3d 1177, 1189 (CA9 2002). The rule comes into play after a forceful seizure has been judged to be reasonable under Graham. Once a court has made that determination, the rule instructs the court to ask whether the law enforcement officer violated the Fourth Amendment in some other way in the course of events leading up to the seizure. If so, that separate Fourth Amendment violation may “render the officer’s otherwise reasonable defensive use of force unreasonable as a matter of law.” Id., at 1190–1191. The provocation rule, which has been “sharply questioned” outside the Ninth Circuit, City and County of San Francisco v. Sheehan, 575 U. S. ___, ___, n. 4 (2015) (slip op., at 14, n. 4), is incompatible with our excessive force jurisprudence. The rule’s fundamental flaw is that it uses another constitutional violation to manufacture an excessive force claim where one would not otherwise exist. The Fourth Amendment prohibits “unreasonable searches and seizures.” “[R]easonableness is always the touchstone of Fourth Amendment analysis,” Birchfield v. North Dakota, 579 U. S. ___, ___ (2016) (slip op., at 37), and reasonableness is generally assessed by carefully weighing “the nature and quality of the intrusion on the individual’s Fourth Amendment interests against the importance of the governmental interests alleged to justify the intrusion.” Tennessee v. Garner, 471 U. S. 1, 8 (1985) (internal quotation marks omitted). Our case law sets forth a settled and exclusive framework for analyzing whether the force used in making a seizure complies with the Fourth Amendment. See Graham, 490 U. S., at 395. As in other areas of our Fourth Amendment jurisprudence, “[d]etermining whether the force used to effect a particular seizure is ‘reasonable’ ” requires balancing of the individual’s Fourth Amendment interests against the relevant government interests. Id., at 396. The operative question in excessive force cases is “whether the totality of the circumstances justifie[s] a particular sort of search or seizure.” Garner, supra, at 8–9. The reasonableness of the use of force is evaluated under an “objective” inquiry that pays “careful attention to the facts and circumstances of each particular case.” Graham, supra, at 396. And “[t]he ‘reasonableness’ of a particular use of force must be judged from the perspective of a reasonable officer on the scene, rather than with the 20/20 vision of hindsight.” Ibid. “Excessive force claims . . . are evaluated for objective reasonableness based upon the information the officers had when the conduct occurred.” Saucier v. Katz, 533 U. S. 194, 207 (2001) . That inquiry is dispositive: When an officer carries out a seizure that is reasonable, taking into account all relevant circumstances, there is no valid excessive force claim. The basic problem with the provocation rule is that it fails to stop there. Instead, the rule provides a novel and unsupported path to liability in cases in which the use of force was reasonable. Specifically, it instructs courts to look back in time to see if there was a different Fourth Amendment violation that is somehow tied to the eventual use of force. That distinct violation, rather than the forceful seizure itself, may then serve as the foundation of the plaintiff’s excessive force claim. Billington, supra, at 1190 (“The basis of liability for the subsequent use of force is the initial constitutional violation . . . ”). This approach mistakenly conflates distinct Fourth Amendment claims. Contrary to this approach, the objective reasonableness analysis must be conducted separately for each search or seizure that is alleged to be unconstitutional. An excessive force claim is a claim that a law enforcement officer carried out an unreasonable seizure through a use of force that was not justified under the relevant circumstances. It is not a claim that an officer used reasonable force after committing a distinct Fourth Amendment violation such as an unreasonable entry. By conflating excessive force claims with other Fourth Amendment claims, the provocation rule permits excessive force claims that cannot succeed on their own terms. That is precisely how the rule operated in this case. The District Court found (and the Ninth Circuit did not dispute) that the use of force by the deputies was reasonable under Graham. However, respondents were still able to recover damages because the deputies committed a separate constitutional violation (the warrantless entry into the shack) that in some sense set the table for the use of force. That is wrong. The framework for analyzing excessive force claims is set out in Graham. If there is no excessive force claim under Graham, there is no excessive force claim at all. To the extent that a plaintiff has other Fourth Amendment claims, they should be analyzed separately.[1]* The Ninth Circuit’s efforts to cabin the provocation rule only undermine it further. The Ninth Circuit appears to recognize that it would be going entirely too far to suggest that any Fourth Amendment violation that is connected to a reasonable use of force should create a valid excessive force claim. See, e.g., Beier v. Lewiston, 354 F. 3d 1058, 1064 (CA9 2004) (“Because the excessive force and false arrest factual inquiries are distinct, establishing a lack of probable cause to make an arrest does not establish an excessive force claim, and vice-versa”). Instead, that court has endeavored to limit the rule to only those distinct Fourth Amendment violations that in some sense “provoked” the need to use force. The concept of provocation, in turn, has been defined using a two-prong test. First, the separate constitutional violation must “creat[e] a situation which led to” the use of force; second, the separate constitutional violation must be committed recklessly or intentionally. 815 F. 3d, at 1193 (internal quotation marks omitted). Neither of these limitations solves the fundamental problem of the provocation rule: namely, that it is an unwarranted and illogical expansion of Graham. But in addition, each of the limitations creates problems of its own. First, the rule includes a vague causal standard. It applies when a prior constitutional violation “created a situation which led to” the use of force. The rule does not incorporate the familiar proximate cause standard. Indeed, it is not clear what causal standard is being applied. Second, while the reasonableness of a search or seizure is almost always based on objective factors, see Whren v. United States, 517 U. S. 806, 814 (1996) , the provocation rule looks to the subjective intent of the officers who carried out the seizure. As noted, under the Ninth Circuit’s rule, a prior Fourth Amendment violation may be held to have provoked a later, reasonable use of force only if the prior violation was intentional or reckless. The provocation rule may be motivated by the notion that it is important to hold law enforcement officers liable for the foreseeable consequences of all of their constitutional torts. See Billington, 292 F. 3d, at 1190 (“[I]f an officer’s provocative actions are objectively unreasonable under the Fourth Amendment, . . . liability is established, and the question becomes . . . what harms the constitutional violation proximately caused”). However, there is no need to distort the excessive force inquiry in order to accomplish this objective. To the contrary, both parties accept the principle that plaintiffs can—subject to qualified immunity—generally recover damages that are proximately caused by any Fourth Amendment violation. See, e.g., Heck v. Humphrey, 512 U. S. 477, 483 (1994) (§1983 “creates a species of tort liability” informed by tort principles regarding “damages and the prerequisites for their recovery” (internal quotation marks omitted)); Memphis Community School Dist. v. Stachura, 477 U. S. 299, 306 (1986) (“[W]hen §1983 plaintiffs seek damages for violations of constitutional rights, the level of damages is ordinarily determined according to principles derived from the common law of torts”). Thus, there is no need to dress up every Fourth Amendment claim as an excessive force claim. For example, if the plaintiffs in this case cannot recover on their excessive force claim, that will not foreclose recovery for injuries proximately caused by the warrantless entry. The harm proximately caused by these two torts may overlap, but the two claims should not be confused. III The Court of Appeals also held that “even without relying on [the] provocation theory, the deputies are liable for the shooting under basic notions of proximate cause.” 815 F. 3d, at 1194. In other words, the court apparently concluded that the shooting was proximately caused by the deputies’ warrantless entry of the shack. Proper analysis of this proximate cause question required consideration of the “foreseeability or the scope of the risk created by the predicate conduct,” and required the court to conclude that there was “some direct relation between the injury asserted and the injurious conduct alleged.” Paroline v. United States, 572 U. S. ___, ___ (2014) (slip op., at 7) (internal quotation marks omitted). Unfortunately, the Court of Appeals’ proximate cause analysis appears to have been tainted by the same errors that cause us to reject the provocation rule. The court reasoned that when officers make a “startling entry” by “barg[ing] into” a home “unannounced,” it is reasonably foreseeable that violence may result. 815 F. 3d, at 1194–1195 (internal quotation marks omitted). But this appears to focus solely on the risks foreseeably associated with the failure to knock and announce, which could not serve as the basis for liability since the Court of Appeals concluded that the officers had qualified immunity on that claim. By contrast, the Court of Appeals did not identify the foreseeable risks associated with the relevant constitutional violation (the warrantless entry); nor did it explain how, on these facts, respondents’ injuries were proximately caused by the warrantless entry. In other words, the Court of Appeals’ proximate cause analysis, like the provocation rule, conflated distinct Fourth Amendment claims and required only a murky causal link between the warrantless entry and the injuries attributed to it. On remand, the court should revisit the question whether proximate cause permits respondents to recover damages for their shooting injuries based on the deputies’ failure to secure a warrant at the outset. See Bank of America Corp. v. Miami, ante, at 12 (declining to “draw the precise boundaries of proximate cause” in the first instance). The arguments made on this point by the parties and by the United States as amicus provide a useful starting point for this inquiry. See Brief for Petitioners 42–56; Brief for Respondents 20–31, 51–59; Reply Brief 17–24; Brief for United States as Amicus Curiae 26–32. * * * For these reasons, the judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 * Respondents do not attempt to defend the provocation rule. Instead, they argue that the judgment below should be affirmed under Graham itself. Graham commands that an officer’s use of force be assessed for reasonableness under the “totality of the circumstances.” 490 U. S., at 396 (internal quotation marks omitted). On respondents’ view, that means taking into account unreasonable police conduct prior to the use of force that foreseeably created the need to use it. Brief for Respondents 42–43. We did not grant certiorari on that question, and the decision below did not address it. Accordingly, we decline to address it here. See, e.g., McLane Co. v. EEOC, ante, at 11 (“[W]e are a court of review, not of first view” (internal quotation marks omitted)). All we hold today is that once a use of force is deemed reasonable under Graham, it may not be found unreasonable by reference to some separate constitutional violation. Any argument regarding the District Court’s application of Graham in this case should be addressed to the Ninth Circuit on remand.

581.US.2016_16-149

The Federal Employees Health Benefits Act of 1959 (FEHBA) authorizes the Office of Personnel Management (OPM) to contract with private carriers for federal employees’ health insurance. 5 U. S. C. §8902(a), (d). FEHBA contains an express-preemption provision, §8902(m)(1), which states that the “terms of any contract under this chapter which relate to the nature, provision, or extent of coverage or benefits (including payments with respect to benefits) shall supersede and preempt any State or local law . . . which relates to health insurance or plans.” OPM’s contracts have long required private carriers to seek subrogation and reimbursement. Accordingly, OPM’s regulations make a carrier’s “right to pursue and receive subrogation and reimbursement recoveries . . . a condition of and a limitation on the nature of benefits or benefit payments and on the provision of benefits under the plan’s coverage.” 5 CFR §890.106(b)(1). In 2015, OPM published a new rule confirming that a carrier’s subrogation and reimbursement rights and responsibilities “relate to the nature, provision, and extent of coverage or benefits (including payments with respect to benefits) within the meaning of” §8902(m)(1), and “are . . . effective notwithstanding any state or local law, or any regulation issued thereunder, which relates to health insurance or plans.” §890.106(h). Respondent Jodie Nevils was insured under a FEHBA plan offered by petitioner Coventry Health Care of Missouri. When Nevils was injured in an automobile accident, Coventry paid his medical expenses. Coventry subsequently asserted a lien against part of the settlement Nevils recovered from the driver who caused his injuries. Nevils satisfied the lien, then filed a class action in Missouri state court, alleging that, under Missouri law, which does not permit subrogation or reimbursement in this context, Coventry had unlawfully obtained reimbursement. Coventry countered that §8902(m)(1) preempted the state law. The trial court granted summary judgment in Coventry’s favor, and the Missouri Court of Appeals affirmed. The Missouri Supreme Court reversed. Finding §8902(m)(1) susceptible to diverse plausible readings, the court invoked a “presumption against preemption” to conclude that the federal statute’s preemptive scope excluded subrogation and reimbursement. On remand from this Court for further consideration in light of OPM’s 2015 rule, the Missouri Supreme Court adhered to its earlier decision. A majority of the Missouri Supreme Court also held that §8902(m)(1) violates the Supremacy Clause. Held: 1. Because contractual subrogation and reimbursement prescriptions plainly “relate to . . . payments with respect to benefits,” §8902(m)(1), they override state laws barring subrogation and reimbursement. Pp. 6–9. (a) This reading best comports with §8902(m)(1)’s text, context, and purpose. Contractual provisions for subrogation and reimbursement “relate to . . . payments with respect to benefits” because subrogation and reimbursement rights yield just such payments. When a carrier exercises its right to either reimbursement or subrogation, it receives from either the beneficiary or a third party “payment” respecting the benefits the carrier had previously paid. The carrier’s very provision of benefits triggers the right to payment. Congress’ use of the expansive phrase “relate to,” which “express[es] a broad pre-emptive purpose,” Morales v. Trans World Airlines, Inc., 504 U. S. 374 , weighs against Nevils’ effort to narrow the term “payments” to exclude payments that occur “long after” a carrier’s provision of benefits. Nevils’ argument that Congress intended to preempt only state coverage requirements, e.g., inclusion of acupuncture and chiropractic services, also miscarries. The statutory context and purpose reinforce this conclusion. FEHBA concerns “benefits from a federal health insurance plan for federal employees that arise from a federal law.” Bell v. Blue Cross & Blue Shield of Okla., 823 F. 3d 1198, 1202. Strong and “distinctly federal interests are involved,” Empire HealthChoice Assurance, Inc. v. McVeigh, 547 U. S. 677 , in uniform administration of the program, free from state interference, particularly in regard to coverage, benefits, and payments. The Federal Government also has a significant financial stake in subrogation and reimbursement. Pp. 6–8. (b) McVeigh’s suggestion that §8902(m)(1) has two “plausible” interpretations, 547 U. S., at 698, Nevils asserts, supports application of the presumption against preemption here. But the Court never chose between the two readings set out in McVeigh, because doing so was not pertinent to the discrete question whether federal courts have subject-matter jurisdiction over FEHBA reimbursement actions. Having decided in McVeigh that §8902(m)(1) is a “choice-of-law prescription,” not a “jurisdiction-conferring provision,” id., at 697, the Court had no cause to consider §8902(m)(1)’s text, context, and purpose, as it does here. Pp. 8–9. 2. The regime Congress enacted is compatible with the Supremacy Clause. The statute itself, not a contract, strips state law of its force. FEHBA contract terms have preemptive force only if they fall within §8902(m)(1)’s preemptive scope. Many other federal statutes found to preempt state law, including the Employee Retirement Income Security Act of 1974 and the Federal Arbitration Act, leave the context-specific scope of preemption to contractual terms. While §8902(m)(1)’s phrasing may differ from those other statutes’, FEHBA’s express-preemption provision manifests the same intent to preempt state law. Pp. 9–11. 492 S. W. 3d 918, reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case. Thomas, J., filed a concurring opinion.

In the Federal Employees Health Benefits Act of 1959 (FEHBA), 5 U. S. C. §8901 et seq., Congress authorized the Office of Personnel Management (OPM) to contract with private carriers for federal employees’ health insurance. §8902(a), (d). FEHBA contains a provision expressly preempting state law. §8902(m)(1). That provision reads: “The terms of any contract under this chapter which relate to the nature, provision, or extent of coverage or benefits (including payments with respect to benefits) shall supersede and preempt any State or local law, or any regulation issued thereunder, which relates to health insurance or plans.” Contracts OPM negotiates with private carriers provide for reimbursement and subrogation. Reimbursement requires an insured employee who receives payment from another source (e.g., the proceeds yielded by a tort claim) to return healthcare costs earlier paid out by the carrier. Subrogation involves transfer of the right to a third-party payment from the insured employee to the carrier, who can then pursue the claim against the third party. Several States, however, Missouri among them, bar enforcement of contractual subrogation and reimbursement provisions. The questions here presented: Does FEHBA’s express-preemption prescription, §8902(m)(1), override state law prohibiting subrogation and reimbursement; and if §8902(m)(1) has that effect, is the statutory prescription consistent with the Supremacy Clause, U. S. Const., Art. VI, cl. 2? We hold, contrary to the decision of the Missouri Supreme Court, that contractual subrogation and reimbursement prescriptions plainly “relate to . . . payments with respect to benefits,” §8902(m)(1); therefore, by statutory instruction, they override state law barring subrogation and reimbursement. We further hold, again contrary to the Missouri Supreme Court, that the regime Congress enacted is compatible with the Supremacy Clause. Section 8902(m)(1) itself, not the contracts OPM negotiates, triggers the federal preemption. As Congress directed, where FEHBA contract terms “relate to the nature, provision, or extent of coverage or benefits (including payments with respect to benefits),” §8902(m)(1) ensures that those terms will be uniformly enforceable nationwide, free from state interference. I A FEHBA “establishes a comprehensive program of health insurance for federal employees.” Empire HealthChoice Assurance, Inc. v. McVeigh, 547 U. S. 677, 682 (2006) . As just noted, supra, at 1, FEHBA contains an express-preemption provision, §8902(m)(1). FEHBA assigns to OPM broad administrative and rulemaking authority over the program. See §§8901–8913. OPM contracts with private insurance carriers to offer a range of healthcare plans. §§8902, 8903. OPM’s contracts with private carriers have long included provisions requiring those carriers to seek subrogation and reimbursement. Accordingly, OPM has issued detailed regulations governing subrogation and reimbursement clauses in FEHBA contracts. See 5 CFR §890.106 (2016). Under those regulations, a carrier’s “right to pursue and receive subrogation and reimbursement recoveries constitutes a condition of and a limitation on the nature of benefits or benefit payments and on the provision of benefits under the plan’s coverage.” §890.106(b)(1). In 2015, after notice and comment, OPM published a rule confirming that “[a] carrier’s rights and responsibilities pertaining to subrogation and reimbursement under any [FEHBA] contract relate to the nature, provision, and extent of coverage or benefits (including payments with respect to benefits) within the meaning of” §8902(m)(1). §890.106(h). Such “rights and responsibilities,” OPM’s rule provides, “are . . . effective notwithstanding any state or local law, or any regulation issued thereunder, which relates to health insurance or plans.” Ibid. Its rule, OPM explained, “comports with longstanding Federal policy and furthers Congres[s’] goals of reducing health care costs and enabling uniform, nationwide application of [FEHBA] contracts.” 80 Fed. Reg. 29203 (2015) (final rule). B Respondent Jodie Nevils is a former federal employee who enrolled in and was insured under a FEHBA plan offered by petitioner Coventry Health Care of Missouri.[1] Nevils v. Group Health Plan, Inc., 418 S. W. 3d 451, 453 (Mo. 2014) (Nevils I ). When Nevils was injured in an automobile accident, Coventry paid his medical expenses. Ibid. Nevils sued the driver who caused his injuries and recovered a settlement award. Ibid. Based on its contract with OPM, see App. to Pet. for Cert. 129a–130a, Coventry asserted a lien for $6,592.24 against part of the settlement proceeds to cover medical bills it had paid. Nevils I, 418 S. W. 3d, at 453. Nevils repaid that amount, thereby satisfying the lien. Ibid. Nevils then filed this class action against Coventry in Missouri state court, alleging that Coventry had unlawfully obtained reimbursement. Ibid. Nevils premised his claim on Missouri law, which does not permit subrogation or reimbursement in this context, see, e.g., Benton House, LLC v. Cook & Younts Ins., Inc., 249 S. W. 3d 878, 881–882 (Mo. App. 2008). Coventry countered that §8902(m)(1) makes subrogation and reimbursement clauses in FEHBA contracts enforceable notwithstanding state law. The trial court granted summary judgment in Coventry’s favor, Nevils v. Group Health Plan, Inc., No. 11SL–CC00535 (Cir. Ct., St. Louis Cty., Mo., May 21, 2012), App. to Pet. for Cert. 28a, 32a, and the Missouri Court of Appeals affirmed, Nevils v. Group Health Plan, Inc., 2012 WL 6689542, *5 (Dec. 26, 2012). The Missouri Supreme Court reversed. Nevils I, 418 S. W. 3d, at 457. That court began with “the assumption that the historic police powers of the States [are] not to be superseded by . . . Federal Act unless that [is] the clear and manifest purpose of Congress.” Id., at 454 (quoting Cipollone v. Liggett Group, Inc., 505 U. S. 504, 516 (1992) ) (alterations in original). Finding §8902(m)(1) susceptible to diverse “plausible readings,” the court invoked a “presumption against preemption” to conclude that the federal statute’s preemptive scope excluded subrogation and reimbursement. 418 S. W. 3d, at 455. Judge Wilson, joined by Judge Breckenridge, concurred in the judgment. Id., at 457. Observing that “it defies logic to insist that benefit repayment terms do not relate to the nature or extent of Nevils’ benefits,” id., at 460 (emphasis deleted), Judge Wilson concluded that “Congress plainly intended for §8902(m)(1) to apply to the benefit repayment terms in [Coventry’s] contract,” id., at 462. He nevertheless concurred, reasoning that the Supremacy Clause did not authorize preemption based on the terms of FEHBA contracts. Id., at 462–465. Coventry sought our review, and we invited the Solicitor General to file a brief expressing the views of the United States. Coventry Health Care of Mo., Inc. v. Nevils, 574 U. S. ___ (2014). While Coventry’s petition was pending, OPM finalized its rule governing subrogation and reimbursement. See supra, at 3. This Court granted certiorari, vacated the Missouri Supreme Court’s judgment, and remanded for further consideration in light of OPM’s recently adopted rule. Coventry Health Care of Mo., Inc. v. Nevils, 576 U. S. ___ (2015). On remand, the Missouri Supreme Court adhered to its earlier decision. Nevils v. Group Health Plan, Inc., 492 S. W. 3d 918, 920, 925 (2016). OPM’s rule, the court maintained, “does not overcome the presumption against preemption and demonstrate Congress’ clear and manifest intent to preempt state law.” Id., at 920. Judge Wilson again concurred, this time joined by a majority of the judges of the Missouri Supreme Court. Id., at 925.[2] In their view, Congress’ “attempt to give preemptive effect to the provisions of a contract between the federal government and a private party is not a valid application of the Supremacy Clause” and, “therefore, does not displace Missouri law here.” Ibid. We granted certiorari to resolve conflicting interpretations of §8902(m)(1). 580 U. S. ___ (2016). Compare 492 S. W. 2d, at 925 (majority opinion), with Bell v. Blue Cross & Blue Shield of Okla., 823 F. 3d 1198, 1199 (CA8 2016) (§8902(m)(1) preempts state antisubrogation law); Helfrich v. Blue Cross & Blue Shield Assn., 804 F. 3d 1090, 1092 (CA10 2015) (same). II Section 8902(m)(1) places two preconditions on federal preemption. See supra, at 1. The parties agree that Missouri’s law prohibiting subrogation and reimbursement meets one of the two limitations, i.e., the State’s law “relates to health insurance or plans.” §8902(m)(1). They dispute only whether the subrogation and reimbursement requirements in OPM’s contract with Coventry “relate to the nature, provision, or extent of coverage or benefits,” “including payments with respect to benefits.” Ibid. Coventry contends that §8902(m)(1) unambiguously covers the contractual terms at issue here. In any event, Coventry, joined by the United States as amicus curiae, urges that the rule published by OPM in 2015 leaves no room for doubt that insurance-contract terms providing for subrogation and reimbursement fall within §8902(m)(1)’s preemptive scope. See supra, at 3. Deference is due to OPM’s reading, Coventry and the United States assert, under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) . In Nevils’ view, by contrast, §8902(m)(1) does not preempt state antisubrogation and antireimbursement laws in light of the presumption against preemption. Given that presumption, Nevils maintains, OPM’s rule is not entitled to deference. Though we have called Nevils’ construction “plausible,” McVeigh, 547 U. S., at 698, the reading advanced by Coventry and the United States best comports with §8902(m)(1)’s text, context, and purpose. A Contractual provisions for subrogation and reimbursement “relate to . . . payments with respect to benefits” because subrogation and reimbursement rights yield just such payments. When a carrier exercises its right to either reimbursement or subrogation, it receives from either the beneficiary or a third party “payment” respecting the benefits the carrier had previously paid. The carrier’s very provision of benefits triggers the right to payment. See Tr. of Oral Arg. 31; Helfrich, 804 F. 3d, at 1106; Bell, 823 F. 3d, at 1204. Congress’ use of the expansive phrase “relate to” shores up that understanding. We have “repeatedly recognized” that the phrase “relate to” in a preemption clause “express[es] a broad pre-emptive purpose.” Morales v. Trans World Airlines, Inc., 504 U. S. 374, 383 (1992) ; accord Northwest, Inc. v. Ginsberg, 572 U. S. ___, ___, ___ (2014) (slip op., at 5, 9). Congress characteristically employs the phrase to reach any subject that has “a connection with, or reference to,” the topics the statute enumerates. Morales, 504 U. S., at 384. The phrase therefore weighs against Nevils’ effort to narrow the term “payments” to exclude payments that occur “long after” a carrier’s provision of benefits (Brief for Respondent 27 (quoting McVeigh, 547 U. S., at 697)). See Nevils I, 418 S. W. 3d, at 460 (Wilson, J., concurring); cf. Hillman v. Maretta, 569 U. S. ___, ___ (2013) (slip op., at 10) (in the Federal Employees’ Group Life Insurance Act context, it “makes no difference” whether state law withholds benefits in the first instance or instead takes them away after they have been paid). Given language notably “expansive [in] sweep,” Morales, 504 U. S., at 384 (internal quotation marks omitted), Nevils’ argument that Congress intended to preempt only state coverage requirements (e.g., for acupuncture and chiropractic services, see Brief for Respondent 36) also miscarries. The statutory context and purpose reinforce our conclusion. FEHBA concerns “benefits from a federal health insurance plan for federal employees that arise from a federal law” in an area with a “long history of federal involvement.” Bell, 823 F. 3d, at 1202. Strong and “distinctly federal interests are involved,” McVeigh, 547 U. S., at 696, in uniform administration of the program, free from state interference, particularly in regard to coverage, benefits, and payments. The Federal Government, more-over, has a significant financial stake. OPM estimates that, in 2014 alone, FEHBA “carriers were reimbursed by approximately $126 million in subrogation recoveries.” 80 Fed. Reg. 29203. Such “recoveries translate to premium cost savings for the federal government and [FEHBA] enrollees.” Ibid. B Invoking our suggestion in McVeigh that §8902(m)(1) has two “plausible” interpretations, 547 U. S., at 698, Nevils nonetheless urges us to apply a presumption against preemption because §8902(m)(1) does not clearly cover contractual terms pertaining to subrogation and reimbursement. This argument is blind to McVeigh’s context. In McVeigh, we considered the discrete question whether 28 U. S. C. §1331 gives federal courts subject-matter jurisdiction over FEHBA reimbursement actions. See 547 U. S., at 683. Our principal holding was that §1331 did not confer federal jurisdiction. Ibid.; see Bell, 823 F. 3d, at 1205. The carrier in McVeigh, as part of its argument in favor of federal jurisdiction, asserted that §8902(m)(1) itself conferred federal jurisdiction. See 547 U. S., at 697. In responding to that assertion, we summarized competing interpretations of §8902(m)(1) advanced in briefing, readings that map closely onto the parties’ positions here. See ibid. (carrier and United States as amicus curiae urged interpretation similar to Coventry’s; an amicus brief in support of beneficiary offered interpretation similar to Nevils’). We made no choice between the two interpretationsset out in McVeigh, however, because the answer madeno difference to the question there presented. Id., at698. “[E]ven if FEHBA’s preemption provision reaches contract-based reimbursement claims,” we explained, “that provision is not sufficiently broad to confer federal jurisdiction.” Ibid. Because §8902(m)(1) is a “choice-of-law prescription,” not a “jurisdiction-conferring provision,” id., at 697, we had no cause to consider §8902(m)(1)’s text, context, and purpose, as we do today, see supra, at 6–8.[3] III Nevils further contends that, if §8902(m)(1) covers subrogation and reimbursement clauses in OPM contracts, then the statute itself would violate the Supremacy Clause by assigning preemptive effect to the terms of a contract, not to the laws of the United States. We conclude, however, that the statute, not a contract, strips state law of its force. Without §8902(m)(1), there would be no preemption of state insurance law. FEHBA contract terms have preemptive force only as they “relate to the nature, provision, or extent of coverage or benefits (including payments with respect to benefits),” §8902(m)(1)—i.e., when the contract terms fall within the statute’s preemptive scope. It is therefore the statute that “ensures that [FEHBA contract] terms will be uniformly enforceable nationwide, notwithstanding any state law relating to health insurance or plans.” Brief for United States as Amicus Curiae 28 (internal quotation marks omitted). Many other federal statutes preempt state law in this way, leaving the context-specific scope of preemption to contractual terms. The Employee Retirement Income Security Act of 1974 (ERISA), 29 U. S. C. §1001 et seq., for example, preempts “any and all State laws insofar as they . . . relate to any employee benefit plan.” §1144(a). And the Federal Arbitration Act (FAA), 9 U. S. C. §1 et seq., limits the grounds for denying enforcement of “written provision[s] in . . . contract[s]” providing for arbitration, thereby preempting state laws that would otherwise interfere with such contracts. §2. This Court has several times held that those statutes preempt state law, see, e.g., Gobeille v. Liberty Mut. Ins. Co., 577 U. S. ___, ___–___ (2016) (slip op., at 5–12) (ERISA); Marmet Health Care Center, Inc. v. Brown, 565 U. S. 530 –534 (2012) (per curiam) (FAA), and Nevils does not contend that those measures violate the Supremacy Clause, see Brief for Respondent 22. Nevils instead attempts to distinguish those other statutes by highlighting a particular textual feature of §8902(m)(1): Section 8902(m)(1) states that the “terms of any contract” between OPM and a carrier “shall supersede and preempt” certain state or local laws. (Emphasis added.) That formulation, Nevils asserts, violates the Supremacy Clause’s mandate that only the “Laws of the United States” may reign supreme over state law. U. S. Const., Art. VI, cl. 2 (emphasis added). Nevils’ argument elevates semantics over substance. While Congress’ formulation might differ from the phrasing of other statutes, §8902(m)(1) manifests the same intent to preempt state law.[4] Because we do not require Congress to employ a particular linguistic formulation when preempting state law, Nevils’ Supremacy Clause challenge fails.[5] * * * For the reasons stated, the judgment of the Supreme Court of Missouri is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 Coventry was formerly known as Group Health Plan, Inc. Pet. for Cert. ii. We refer to both the current and former entities as “Coventry.” 2 Under Missouri law, a “concurring opinion” in which “a majority of the court concur[s]” is binding precedent. Mueller v. Burchfield, 359 Mo. 876, 880, 224 S. W. 2d 87, 89 (1949). 3 Because the statute alone resolves this dispute, we need not consider whether Chevron deference attaches to OPM’s 2015 rule. 4 Congress’ choice of language is not unique to §8902(m)(1). Several related statutes governing federal-employee and military-member benefits employ similar formulations. See §8959 (“The terms of any contract that relate to the nature, provision, or extent of coverage or benefits (including payments with respect to benefits) shall supersede and preempt any State or local law, or any regulation issued thereunder, which relates to dental benefits, insurance, plans, or contracts.”); §8989 (same for vision); §9005(a) (same for long-term care); 10 U. S. C. §1103(a) (certain state laws “shall not apply to any contract entered into pursuant to this chapter”). 5 Nevils’ speculation about the Government’s outsourcing preemption to private entities, see Brief for Respondent 24, is far afield from the matter before us. This case involves only Congress’ preemption of state insurance laws to ensure that the terms in contracts negotiated by OPM, a federal agency, operate free from state interference.

580.US.2016_15-649

There are three possible conclusions to a Chapter 11 bankruptcy. First, debtor and creditors may negotiate a plan to govern the distribution of the estate’s value. See, e.g., 11 U. S. C. §§1121, 1123, 1129, 1141. Second, the bankruptcy court may convert the case to Chapter 7 for liquidation of the business and distribution of its assets to creditors. §§1112(a), (b), 726. Finally, the bankruptcy court may dismiss the case. §1112(b). A court ordering a dismissal ordinarily attempts to restore the prepetition financial status quo. §349(b)(3). Yet if perfect restoration proves difficult or impossible, the court may, “for cause,” alter the dismissal’s normal restorative consequences, §349(b)—i.e., it may order a “structured dismissal.” The Bankruptcy Code also establishes basic priority rules for determining the order in which the court will distribute an estate’s assets. The Code makes clear that distributions in a Chapter 7 liquidation must follow this prescribed order. §§725, 726. Chapter 11 permits some flexibility, but a court still cannot confirm a plan that contains priority-violating distributions over the objection of an impaired creditor class. §§1129(a)(7), (b)(2). The Code does not explicitly state what priority rules—if any—apply to the distribution of assets in a structured dismissal. Respondent Jevic Transportation filed for Chapter 11 bankruptcy after being purchased in a leveraged buyout. The bankruptcy prompted two lawsuits. In the first, a group of former Jevic truckdrivers, petitioners here, was awarded a judgment against Jevic for Jevic’s failure to provide proper notice of termination in violation of state and federal Worker Adjustment and Retraining Notification (WARN) Acts. Part of that judgment counted as a priority wage claim under §507(a)(4), entitling the workers to payment ahead of general unsecured claims against the Jevic estate. In the second suit, a court-authorized committee representing Jevic’s unsecured creditors sued Sun Capital and CIT Group, respondents here, for fraudulent conveyance in connection with the leveraged buyout of Jevic. These parties negotiated a settlement agreement that called for a structured dismissal of Jevic’s Chapter 11 bankruptcy. Under the proposed structured dismissal, petitioners would receive nothing on their WARN claims, but lower-priority general unsecured creditors would be paid. Petitioners argued that the distribution scheme accordingly violated the Code’s priority rules by paying general unsecured claims ahead of their own. The Bankruptcy Court nevertheless approved the settlement agreement and dismissed the case, reasoning that because the proposed payouts would occur pursuant to a structured dismissal rather than an approved plan, the failure to follow ordinary priority rules did not bar approval. The District Court and the Third Circuit affirmed. Held: 1. Petitioners have Article III standing. Respondents argue that petitioners have not “suffered an injury in fact,” or at least one “likely to be redressed by a favorable judicial decision,” Spokeo, Inc. v. Robins, 578 U. S. ___, ___, because petitioners would have gotten nothing even if the Bankruptcy Court had never approved the structured dismissal and will still get nothing if the structured dismissal is undone now. That argument rests upon the assumptions that (1) without a violation of ordinary priority rules, there will be no settlement, and (2) without a settlement, the fraudulent-conveyance lawsuit has no value. The record, however, indicates both that a settlement that respects ordinary priorities remains a reasonable possibility and that the fraudulent-conveyance claim could have litigation value. Therefore, as a consequence of the Bankruptcy Court’s approval of the structured dismissal, petitioners lost a chance to obtain a settlement that respected their priorities or, if not that, the power to assert the fraudulent-conveyance claim themselves. A decision in their favor is likely to redress that loss. Pp. 9–11. 2. Bankruptcy courts may not approve structured dismissals that provide for distributions that do not follow ordinary priority rules without the consent of affected creditors. Pp. 11–18. (a) Given the importance of the priority system, this Court looks for an affirmative indication of intent before concluding that Congress means to make a major departure. See Whitman v. American Trucking Assns., Inc., 531 U. S. 457 . Nothing in the statute evinces such intent. Insofar as the dismissal sections of Chapter 11 foresee any transfer of assets, they seek a restoration of the prepetition financial status quo. Read in context, §349(b), which permits a bankruptcy judge, “for cause, [to] orde[r] otherwise,” seems designed to give courts the flexibility to protect reliance interests acquired in the bankruptcy, not to make general end-of-case distributions that would be flatly impermissible in a Chapter 11 plan or Chapter 7 liquidation. Precedent does not support a contrary position. E.g., In re Iridium Operating LLC, 478 F. 3d 452 (CA2), distinguished. Cases in which courts have approved deviations from ordinary priority rules generally involve interim distributions serving significant Code-related objectives. That is not the case here, where, e.g., the priority-violating distribution is attached to a final disposition, does not preserve the debtor as a going concern, does not make the disfavored creditors better off, does not promote the possibility of a confirmable plan, does not help to restore the status quo ante, and does not protect reliance interests. Pp. 11–16. (b) Congress did not authorize a “rare case” exception that permits courts to disregard priority in structured dismissals for “sufficient reasons.” The fact that it is difficult to give precise content to the concept of “sufficient reasons” threatens to turn the court below’s exception into a more general rule, resulting in uncertainty that has potentially serious consequences—e.g., departure from the protections granted particular classes of creditors, changes in the bargaining power of different classes of creditors even in bankruptcies that do not end in structured dismissals, risks of collusion, and increased difficulty in achieving settlements. Courts cannot deviate from the strictures of the Code, even in “rare cases.” Pp. 16–18. 787 F. 3d 173, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined.

Bankruptcy Code Chapter 11 allows debtors and their creditors to negotiate a plan for dividing an estate’s value. See 11 U. S. C. §§1123, 1129, 1141. But sometimes the parties cannot agree on a plan. If so, the bankruptcy court may decide to dismiss the case. §1112(b). The Code then ordinarily provides for what is, in effect, a restoration of the prepetition financial status quo. §349(b). In the case before us, a Bankruptcy Court dismissed a Chapter 11 bankruptcy. But the court did not simply restore the prepetition status quo. Instead, the court ordered a distribution of estate assets that gave money to high-priority secured creditors and to low-priority general unsecured creditors but which skipped certain dissenting mid-priority creditors. The skipped creditors would have been entitled to payment ahead of the general unsecured creditors in a Chapter 11 plan (or in a Chapter 7 liquidation). See §§507, 725, 726, 1129. The question before us is whether a bankruptcy court has the legal power to order this priority-skipping kind of distribution scheme in connection with a Chapter 11 dismissal. In our view, a bankruptcy court does not have such a power. A distribution scheme ordered in connection with the dismissal of a Chapter 11 case cannot, without the consent of the affected parties, deviate from the basic priority rules that apply under the primary mechanisms the Code establishes for final distributions of estate value in business bankruptcies. I A 1 We begin with a few fundamentals: A business may file for bankruptcy under either Chapter 7 or Chapter 11. In Chapter 7, a trustee liquidates the debtor’s assets and distributes them to creditors. See §701 et seq. In Chapter 11, debtor and creditors try to negotiate a plan that will govern the distribution of valuable assets from the debtor’s estate and often keep the business operating as a going concern. See, e.g., §§1121, 1123, 1129, 1141 (setting out the framework in which the parties negotiate). Filing for Chapter 11 bankruptcy has several relevant legal consequences. First, an estate is created comprising all property of the debtor. §541(a)(1). Second, a fiduciary is installed to manage the estate in the interest of the creditors. §§1106, 1107(a). This fiduciary, often the debtor’s existing management team, acts as “debtor in possession.” §§1101(1), 1104. It may operate the business, §§363(c)(1), 1108, and perform certain bankruptcy-related functions, such as seeking to recover for the estate preferential or fraudulent transfers made to other persons, §547 (transfers made before bankruptcy that unfairly preferred particular creditors); §548 (fraudulent transfers, including transfers made before bankruptcy for which the debtor did not receive fair value). Third, an “automatic stay” of all collection proceedings against the debtor takes effect. §362(a). It is important to keep in mind that Chapter 11 foresees three possible outcomes. The first is a bankruptcy-court-confirmed plan. Such a plan may keep the business operating but, at the same time, help creditors by providing for payments, perhaps over time. See §§1123, 1129, 1141. The second possible outcome is conversion of the case to a Chapter 7 proceeding for liquidation of the business and a distribution of its remaining assets. §§1112(a), (b), 726. That conversion in effect confesses an inability to find a plan. The third possible outcome is dismissal of the Chapter 11 case. §1112(b). A dismissal typically “revests the property of the estate in the entity in which such property was vested immediately before the commencement of the case”—in other words, it aims to return to the prepetition financial status quo. §349(b)(3). Nonetheless, recognizing that conditions may have changed in ways that make a perfect restoration of the status quo difficult or impossible, the Code permits the bankruptcy court, “for cause,” to alter a Chapter 11 dismissal’s ordinary restorative consequences. §349(b). A dismissal that does so (or which has other special conditions attached) is often referred to as a “structured dismissal,” defined by the American Bankruptcy Institute as a “hybrid dismissal and confirmation order . . . that . . . typically dismisses the case while, among other things, approving certain distributions to creditors, granting certain third-party releases, enjoining certain conduct by creditors, and not necessarily vacating orders or unwinding transactions undertaken during the case.” American Bankruptcy Institute Commission To Study the Reform of Chapter 11, 2012–2014 Final Report and Recommendations 270 (2014). Although the Code does not expressly mention structured dismissals, they “appear to be increasingly common.” Ibid., n. 973. 2 The Code also sets forth a basic system of priority, which ordinarily determines the order in which the bankruptcy court will distribute assets of the estate. Secured creditors are highest on the priority list, for they must receive the proceeds of the collateral that secures their debts. 11 U. S. C. §725. Special classes of creditors, such as those who hold certain claims for taxes or wages, come next in a listed order. §§507, 726(a)(1). Then come low-priority creditors, including general unsecured creditors. §726(a)(2). The Code places equity holders at the bottom of the priority list. They receive nothing until all previ-ously listed creditors have been paid in full. §726(a)(6). The Code makes clear that distributions of assets in a Chapter 7 liquidation must follow this prescribed order. §§725, 726. It provides somewhat more flexibility for distributions pursuant to Chapter 11 plans, which may impose a different ordering with the consent of the af-fected parties. But a bankruptcy court cannot confirm a plan that contains priority-violating distributions over the objection of an impaired creditor class. §§1129(a)(7), 1129(b)(2). The question here concerns the interplay between the Code’s priority rules and a Chapter 11 dismissal. Here, the Bankruptcy Court neither liquidated the debtor under Chapter 7 nor confirmed a Chapter 11 plan. But the court, instead of reverting to the prebankruptcy status quo, ordered a distribution of the estate assets to creditors by attaching conditions to the dismissal (i.e., it ordered a structured dismissal). The Code does not explicitly state what priority rules—if any—apply to a distribution in these circumstances. May a court consequently provide for distributions that deviate from the ordinary priority rules that would apply to a Chapter 7 liquidation or a Chapter 11 plan? Can it approve conditions that give estate assets to members of a lower priority class while skipping objecting members of a higher priority class? B In 2006, Sun Capital Partners, a private equity firm, acquired Jevic Transportation Corporation with money borrowed from CIT Group in a “leveraged buyout.” In a leveraged buyout, the buyer (B) typically borrows from a third party (T) a large share of the funds needed to purchase a company (C). B then pays the money to C’s shareholders. Having bought the stock, B owns C. B then pledges C’s assets to T so that T will have security for its loan. Thus, if the selling price for C is $50 million, B might use $10 million of its own money, borrow $40 million from T, pay $50 million to C’s shareholders, and then pledge C assets worth $40 million (or more) to T as secu-rity for T’s $40 million loan. If B manages C well, it might make enough money to pay T back the $40 million and earn a handsome profit on its own $10 million investment. But, if the deal sours and C descends into bankruptcy, beware of what might happen: Instead of C’s $40 million in assets being distributed to its existing creditors, the money will go to T to pay back T’s loan—the loan that allowed B to buy C. (T will receive what remains of C’s assets because T is now a secured creditor, putting it at the top of the priority list). Since C’s shareholders receive money while C’s creditors lose their claim to C’s remaining assets, unsuccessful leveraged buyouts often lead to fraudulent conveyance suits alleging that the purchaser (B) transferred the company’s assets without receiving fair value in return. See Lipson & Vandermeuse, Stern, Seriously: The Article I Judicial Power, Fraudulent Transfers, and Leveraged Buyouts, 2013 Wis. L. Rev. 1161, 1220–1221. This is precisely what happened here. Just two years after Sun’s buyout, Jevic (C in our leveraged buyout example) filed for Chapter 11 bankruptcy. At the time of filing, it owed $53 million to senior secured creditors Sun and CIT (B and T in our example), and over $20 million to tax and general unsecured creditors. The circumstances surrounding Jevic’s bankruptcy led to two lawsuits. First, petitioners, a group of former Jevic truckdrivers, filed suit in bankruptcy court against Jevic and Sun. Petitioners pointed out that, just before entering bankruptcy, Jevic had halted almost all its operations and had told petitioners that they would be fired. Petitioners claimed that Jevic and Sun had thereby violated state and federal Worker Adjustment and Retraining Notification (WARN) Acts—laws that require a company to give workers at least 60 days’ notice before their termination. See 29 U. S. C. §2102; N. J. Stat. Ann. §34:21–2 (West 2011). The Bankruptcy Court granted summary judgment for petitioners against Jevic, leaving them (and this is the point to remember) with a judgment that petitioners say is worth $12.4 million. See In re Jevic Holding Corp., 496 B.R. 151 (Bkrtcy. Ct. Del. 2013). Some $8.3 million of that judgment counts as a priority wage claim under 11 U. S. C. §507(a)(4), and is therefore entitled to payment ahead of general unsecured claims against the Jevicestate. Petitioners’ WARN suit against Sun continued throughout most of the litigation now before us. But eventually Sun prevailed on the ground that Sun was not the workers’ employer at the relevant times. See In re Jevic Holding Corp., 656 Fed. Appx. 617 (CA3 2016). Second, the Bankruptcy Court authorized a committee representing Jevic’s unsecured creditors to sue Sun and CIT. The Bankruptcy Court and the parties were aware that any proceeds from such a suit would belong not to the unsecured creditors, but to the bankruptcy estate. See §§541(a)(1), (6); Official Comm. of Unsecured Creditors of Cybergenics Corp. v. Chinery, 330 F.3d 548, 552–553 (CA3 2003) (en banc) (holding that a creditor’s committee can bring a derivative action on behalf of the estate). The committee alleged that Sun and CIT, in the course of their leveraged buyout, had “hastened Jevic’s bankruptcy by saddling it with debts that it couldn’t service.” In re Jevic Holding Corp., 787 F.3d 173, 176 (CA3 2015). In 2011, the Bankruptcy Court held that the committee had adequately pleaded claims of preferential transfer under §547 and of fraudulent transfer under §548. In re Jevic Holding Corp., 2011 WL 4345204 (Bkrtcy. Ct. Del., Sept. 15, 2011). Sun, CIT, Jevic, and the committee then tried to negotiate a settlement of this “fraudulent-conveyance” lawsuit. By that point, the depleted Jevic estate’s only remaining assets were the fraudulent-conveyance claim itself and $1.7 million in cash, which was subject to a lien held by Sun. The parties reached a settlement agreement. It pro-vided (1) that the Bankruptcy Court would dismiss the fraudulent-conveyance action with prejudice; (2) that CIT would deposit $2 million into an account earmarked to pay the committee’s legal fees and administrative expenses; (3) that Sun would assign its lien on Jevic’s remaining $1.7 million to a trust, which would pay taxes and administrative expenses and distribute the remainder on a pro rata basis to the low-priority general unsecured creditors, but which would not distribute anything to petitioners (who, by virtue of their WARN judgment, held an $8.3 million mid-level-priority wage claim against the estate); and (4) that Jevic’s Chapter 11 bankruptcy would be dismissed. Apparently Sun insisted on a distribution that would skip petitioners because petitioners’ WARN suit against Sun was still pending and Sun did not want to help finance that litigation. See 787 F. 3d, at 177–178, n. 4 (Sun’s counsel acknowledging before the Bankruptcy Court that “ ‘Sun probably does care where the money goes because you can take judicial notice that there’s a pending WARN action against Sun by the WARN plaintiffs. And if the money goes to the WARN plaintiffs, then you’re funding someone who is suing you who otherwise doesn’t have funds and is doing it on a contingent fee basis’ ”). The essential point is that, regardless of the reason, the proposed settlement called for a structured dismissal that provided for distributions that did not follow ordinary priority rules. Sun, CIT, Jevic, and the committee asked the Bankruptcy Court to approve the settlement and dismiss the case. Petitioners and the U. S. Trustee objected, arguing that the settlement’s distribution plan violated the Code’s priority scheme because it skipped petitioners—who, by virtue of their WARN judgment, had mid-level priority claims against estate assets—and distributed estate money to low-priority general unsecured creditors. The Bankruptcy Court agreed with petitioners that the settlement’s distribution scheme failed to follow ordinary priority rules. App. to Pet. for Cert. 58a. But it held that this did not bar approval. Ibid. That, in the Bankruptcy Court’s view, was because the proposed payouts would occur pursuant to a structured dismissal of a Chapter 11 petition rather than an approval of a Chapter 11 plan. Ibid. The court accordingly decided to grant the motion in light of the “dire circumstances” facing the estate and its creditors. Id., at 57a. Specifically, the court predicted that without the settlement and dismissal, there was “no realistic prospect” of a meaningful distribution for anyone other than the secured creditors. Id., at 58a. A confirm-able Chapter 11 plan was unattainable. And there would be no funds to operate, investigate, or litigate were the case converted to a proceeding in Chapter 7. Ibid. The District Court affirmed the Bankruptcy Court. It recognized that the settlement distribution violated ordinary priority rules. But those rules, it wrote, were “not a bar to the approval of the settlement as [the settlement] is not a reorganization plan.” In re Jevic Holding Corp., 2014 WL 268613, *3 (D Del., Jan. 24, 2014). The Third Circuit affirmed the District Court by a vote of 2 to 1. 787 F. 3d, at 175; id., at 186 (Scirica, J., concurring in part and dissenting in part). The majority held that structured dismissals need not always respect prior-ity. Congress, the court explained, had only “codified the absolute priority rule . . . in the specific context of plan confirmation.” Id., at 183. As a result, courts could, “in rare instances like this one, approve structured dismissals that do not strictly adhere to the Bankruptcy Code’s priority scheme.” Id., at 180. Petitioners (the workers with the WARN judgment) sought certiorari. We granted their petition. II Respondents initially argue that petitioners lack standing because they have suffered no injury, or at least no injury that will be remedied by a decision in their favor. See Spokeo, Inc. v. Robins, 578 U. S. ___, ___ (2016) (slip op., at 6) (explaining that, for Article III standing, a plaintiff must have “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision”). Respondents concede that the structured dismissal approved by the Bankruptcy Court contained distribution conditions that skipped over petitioners, ensur-ing that petitioners received nothing on their multimillion-dollar WARN claim against the Jevic estate. But respondents still assert that petitioners suffered no loss. The reason, respondents say, is that petitioners would have gotten nothing even if the Bankruptcy Court had never approved the structured dismissal in the first place, and will still get nothing if the structured dismissal is undone now. Reversal will eliminate the settlement of the committee’s fraudulent-conveyance lawsuit, which was conditioned on the Bankruptcy Court’s approval of the priority-violating structured dismissal. If the Bankruptcy Court cannot approve that dismissal, respondents contend, Sun and CIT will no longer agree to settle. Nor will petitioners ever be able to obtain a litigation recovery. Hence there will be no lawsuit money to distribute. And in the absence of lawsuit money, Jevic’s assets amount to about $1.7 million, all pledged to Sun, leaving nothing for anyone else, let alone petitioners. Thus, even if petitioners are right that the structured dismissal was impermissible, it cost them nothing. And a judicial decision in their favor will gain them nothing. No loss. No redress. This argument, however, rests upon respondents’ claims (1) that, without a violation of ordinary priority rules, there will be no settlement, and (2) that, without a settlement, the fraudulent-conveyance lawsuit has no value. In our view, the record does not support either of these propositions. As to the first, the record indicates that a settlement that respects ordinary priorities remains a reasonable possibility. It makes clear (as counsel made clear before our Court, see Tr. of Oral Arg. 58) that Sun insisted upon a settlement that gave petitioners nothing only because it did not want to help fund petitioners’ WARN lawsuit against it. See 787 F. 3d, at 177–178, n. 4. But, Sun has now won that lawsuit. See 656 Fed. Appx. 617. If Sun’s given reason for opposing distributions to petitioners has disappeared, why would Sun not settle while permitting some of the settlement money to go to petitioners? As to the second, the record indicates that the fraudulent-conveyance claim could have litigation value. CIT and Sun, after all, settled the lawsuit for $3.7 million, which would make little sense if the action truly had no chance of success. The Bankruptcy Court could convert the case to Chapter 7, allowing a Chapter 7 trustee to pursue the suit against Sun and CIT. Or the court could simply dismiss the Chapter 11 bankruptcy, thereby allowing petitioners to assert the fraudulent-conveyance claim themselves. Given these possibilities, there is no reason to believe that the claim could not be pursued with counsel obtained on a contingency basis. Of course, the lawsuit—like any lawsuit—might prove fruitless, but the mere possibility of failure does not eliminate the value of the claim or petitioners’ injury in being unable to bring it. Consequently, the Bankruptcy Court’s approval of the structured dismissal cost petitioners something. They lost a chance to obtain a settlement that respected their prior-ity. Or, if not that, they lost the power to bring their own lawsuit on a claim that had a settlement value of $3.7 million. For standing purposes, a loss of even a small amount of money is ordinarily an “injury.” See, e.g., McGowan v. Maryland, 366 U.S. 420 –431 (1961) (finding that appellants fined $5 plus costs had standing to assert an Establishment Clause challenge). And the ruling before us could well have cost petitioners considerably more. See Clinton v. City of New York, 524 U.S. 417 –431 (1998) (imposition of a “substantial contingent liability” qualifies as an injury). A decision in petitioners’ favor is likely to redress that loss. We accordingly conclude that petitioners have standing. III We turn to the basic question presented: Can a bankruptcy court approve a structured dismissal that provides for distributions that do not follow ordinary priority rules without the affected creditors’ consent? Our simple answer to this complicated question is “no.” The Code’s priority system constitutes a basic underpinning of business bankruptcy law. Distributions of estate assets at the termination of a business bankruptcy normally take place through a Chapter 7 liquidation or a Chapter 11 plan, and both are governed by priority. In Chapter 7 liquidations, priority is an absolute command—lower priority creditors cannot receive anything until higher priority creditors have been paid in full. See 11 U. S. C. §§725, 726. Chapter 11 plans provide somewhat more flexibility, but a priority-violating plan still cannot be confirmed over the objection of an impaired class of creditors. See §1129(b). The priority system applicable to those distributions has long been considered fundamental to the Bankruptcy Code’s operation. See H. R. Rep. No. 103–835, p. 33 (1994) (explaining that the Code is “designed to enforce a distribution of the debtor’s assets in an orderly manner . . . in accordance with established principles rather than on the basis of the inside influence or economic leverage of a particular creditor”); Roe & Tung, Breaking Bankruptcy Priority: How Rent-Seeking Upends The Creditors’ Bargain, 99 Va. L. Rev. 1235, 1243, 1236 (2013) (arguing that the first principle of bankruptcy is that “distribution conforms to predetermined statutory and contractual priorities,” and that priority is, “quite appropriately, bankruptcy’s most important and famous rule”); Markell, Owners, Auctions, and Absolute Priority in Bankruptcy Reorganizations, 44 Stan. L. Rev. 69, 123 (1991) (stating that a fixed priority scheme is recognized as “the cornerstone of reorganization practice and theory”). The importance of the priority system leads us to expect more than simple statutory silence if, and when, Congress were to intend a major departure. See Whitman v. American Trucking Assns., Inc., 531 U.S. 457, 468 (2001) (“Congress . . . does not, one might say, hide elephants in mouseholes”). Put somewhat more directly, we would expect to see some affirmative indication of intent if Congress actually meant to make structured dismissals a backdoor means to achieve the exact kind of nonconsen-sual priority-violating final distributions that the Code prohibits in Chapter 7 liquidations and Chapter 11 plans. We can find nothing in the statute that evinces this intent. The Code gives a bankruptcy court the power to “dismiss” a Chapter 11 case. §1112(b). But the word “dismiss” itself says nothing about the power to make nonconsensual priority-violating distributions of estate value. Neither the word “structured,” nor the word “conditions,” nor anything else about distributing estate value to creditors pursuant to a dismissal appears in any relevant part of the Code. Insofar as the dismissal sections of Chapter 11 foresee any transfer of assets, they seek a restoration of the pre-petition financial status quo. See §349(b)(1) (dismissal ordinarily reinstates a variety of avoided transfers and voided liens); §349(b)(2) (dismissal ordinarily vacates certain types of bankruptcy orders); §349(b)(3) (dismissal ordinarily “revests the property of the estate in the entity in which such property was vested immediately before the commencement of the case”); see also H. R. Rep. No. 95–595, p. 338 (1977) (dismissal’s “basic purpose . . . is to undo the bankruptcy case, as far as practicable, and to restore all property rights to the position in which they were found at the commencement of the case”). Section 349(b), we concede, also says that a bankruptcy judge may, “for cause, orde[r] otherwise.” But, read in context, this provision appears designed to give courts the flexibility to “make the appropriate orders to protect rights acquired in reliance on the bankruptcy case.” H. R. Rep. No. 95–595, at 338; cf., e.g., Wiese v. Community Bank of Central Wis., 552 F.3d 584, 590 (CA7 2009) (upholding, under §349(b), a Bankruptcy Court’s decision not to reinstate a debtor’s claim against a bank that gave up a lien in reliance on the claim being released in the debtor’s reorganization plan). Nothing else in the Code authorizes a court ordering a dismissal to make general end-of-case distributions of estate assets to creditors of the kind that normally take place in a Chapter 7 liquidation or Chapter 11 plan—let alone final distributions that do not help to restore the status quo ante or protect reliance interests acquired in the bankruptcy, and that would be flatly impermissible in a Chapter 7 liquidation or a Chapter 11 plan because they violate priority without the impaired creditors’ consent. That being so, the word “cause” is too weak a reed upon which to rest so weighty a power. See United Sav. Assn. of Tex. v. Timbers of Inwood Forest Associates, Ltd., 484 U.S. 365, 371 (1988) (noting that “[s]tatutory construction . . . is a holistic endeavor” and that a court should select a “meanin[g that] produces a substantive effect that is compatible with the rest of the law”); Kelly v. Robinson, 479 U.S. 36, 43 (1986) (in interpreting a statute, a court “must not be guided by a single sentence or member of a sentence, but look to the provisions of the whole law, and to its object and policy” (internal quotation marks omitted)); cf. In re Sadler, 935 F.2d 918, 921 (CA7 1991) (“ ‘Cause’ under §349(b) means an acceptable reason. Desire to make an end run around a statute is not an adequate reason”). We have found no contrary precedent, either from this Court, or, for that matter, from lower court decisions reflecting common bankruptcy practice. The Third Circuit referred briefly to In re Buffet Partners, L. P., 2014 WL 3735804 (Bkrtcy. Ct. ND Tex., July 28, 2014). The court in that case approved a structured dismissal. (We express no view about the legality of structured dismissals in general.) But at the same time it pointed out “that not one party with an economic stake in the case has objected to the dismissal in this manner.” Id., at *4. The Third Circuit also relied upon In re Iridium Operating LLC, 478 F.3d 452 (CA2 2007). But Iridium did not involve a structured dismissal. It addressed an interim distribution of settlement proceeds to fund a litigation trust that would press claims on the estate’s behalf. See id., at 459–460. The Iridium court observed that, when evaluating this type of preplan settlement, “[i]t is difficult to employ the rule of priorities” because “the nature and extent of the Estate and the claims against it are not yet fully resolved.” Id., at 464 (emphasis added). The decision does not state or suggest that the Code authorizes nonconsensual departures from ordinary priority rules in the context of a dismissal—which is a final distribution of estate value—and in the absence of any further unresolved bankruptcy issues. We recognize that Iridium is not the only case in which a court has approved interim distributions that violate ordinary priority rules. But in such instances one can generally find significant Code-related objectives that the priority-violating distributions serve. Courts, for example, have approved “first-day” wage orders that allow payment of employees’ prepetition wages, “critical vendor” orders that allow payment of essential suppliers’ prepetition invoices, and “roll-ups” that allow lenders who continue financing the debtor to be paid first on their prepetition claims. See Cybergenics, 330 F. 3d, at 574, n. 8; D. Baird, Elements of Bankruptcy 232–234 (6th ed. 2014); Roe, 99 Va. L. Rev., at 1250–1264. In doing so, these courts have usually found that the distributions at issue would “enable a successful reorganization and make even the disfavored creditors better off.” In re Kmart Corp., 359 F.3d 866, 872 (CA7 2004) (discussing the justifications for critical-vendor orders); see also Toibb v. Radloff, 501 U.S. 157 –164 (1991) (recognizing “permitting business debtors to reorganize and restructure their debts in order to revive the debtors’ businesses” and “maximizing the value of the bankruptcy estate” as purposes of the Code). By way of contrast, in a structured dismissal like the one ordered below, the priority-violating distribution is attached to a final disposition; it does not preserve the debtor as a going concern; it does not make the disfavored creditors better off; it does not promote the possibility of a confirmable plan; it does not help to restore the status quo ante; and it does not protect reliance interests. In short, we cannot find in the violation of ordinary priority rules that occurred here any significant offsetting bankruptcy-related justification. Rather, the distributions at issue here more closely resemble proposed transactions that lower courts have refused to allow on the ground that they circumvent the Code’s procedural safeguards. See, e.g., In re Braniff Airways, Inc., 700 F.2d 935, 940 (CA5 1983) (prohibiting an attempt to “short circuit the requirements of Chapter 11 for confirmation of a reorganization plan by establishing the terms of the plan sub rosa in connection with a sale of assets”); In re Lionel Corp., 722 F.2d 1063, 1069 (CA2 1983) (reversing a Bankruptcy Court’s approval of an asset sale after holding that §363 does not “gran[t] the bankruptcy judge carte blanche” or “swallo[w] up Chapter 11’s safeguards”); In re Biolitec, Inc., 528 B.R. 261, 269 (Bkrtcy. Ct. NJ 2014) (rejecting a structured dismissal because it “seeks to alter parties’ rights without their consent and lacks many of the Code’s most important safeguards”); cf. In re Chrysler LLC, 576 F.3d 108, 118 (CA2 2009) (approving a §363 asset sale because the bankruptcy court demonstrated “proper solicitude for the priority between creditors and deemed it essential that the [s]ale in no way upset that priority”), vacated as moot, 592 F.3d 370 (CA2 2010) (per curiam). IV We recognize that the Third Circuit did not approve nonconsensual priority-violating structured dismissals in general. To the contrary, the court held that they were permissible only in those “rare case[s]” in which courts could find “sufficient reasons” to disregard priority. 787 F. 3d, at 175, 186. Despite the “rare case” limitation, we still cannot agree. For one thing, it is difficult to give precise content to the concept “sufficient reasons.” That fact threatens to turn a “rare case” exception into a more general rule. Consider the present case. The Bankruptcy Court feared that (1) without the worker-skipping distribution, there would be no settlement, (2) without a settlement, all the unsecured creditors would receive nothing, and consequently (3) its distributions would make some creditors (high- and low-priority creditors) better off without making other (mid-priority) creditors worse off (for they would receive nothing regardless). But, as we have pointed out, the record provides equivocal support for the first two propositions. See supra, at 9–11. And, one can readily imagine other cases that turn on comparably dubious predictions. The result is uncertainty. And uncertainty will lead to similar claims being made in many, not just a few, cases. See Rudzik, A Priority Is a Priority Is a Priority—Except When It Isn’t, 34 Am. Bankr. Inst. J. 16, 79 (2015) (“[O]nce the floodgates are opened, debtors and favored creditors can be expected to make every case that ‘rare case’ ”). The consequences are potentially serious. They include departure from the protections Congress granted particular classes of creditors. See, e.g., United States v. Embassy Restaurant, Inc., 359 U.S. 29, 32 (1959) (Congress established employee wage priority “to alleviate in some degree the hardship that unemployment usually brings to workers and their families” when an employer files for bankruptcy); H. R. Rep. No. 95–595, at 187 (explaining the importance of ensuring that employees do not “abandon a failing business for fear of not being paid”). They include changes in the bargaining power of different classes of creditors even in bankruptcies that do not end in structured dismissals. See Warren, A Theory of Absolute Priority, 1991 Ann. Survey Am. L. 9, 30. They include risks of collusion, i.e., senior secured creditors and general unsecured creditors teaming up to squeeze out priority unsecured creditors. See Bank of America Nat. Trust and Sav. Assn. v. 203 North LaSalle Street Partnership, 526 U.S. 434, 444 (1999) (discussing how the absolute priority rule was developed in response to “concern with ‘the ability of a few insiders, whether representatives of management or major creditors, to use the reorganization process to gain an unfair advantage’ ” (quoting H. R. Doc. No. 93–137, pt. I, p. 255 (1973))). And they include making settlement more difficult to achieve. See Landes & Posner, Legal Precedent: A Theoretical and Empirical Analysis, 19 J. Law & Econ. 249, 271 (1976) (arguing that “the ratio of lawsuits to settlements is mainly a function of the amount of uncertainty, which leads to divergent estimates by the parties of the probable outcome”); see also RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 566 U.S. 639, 649 (2012) (noting the importance of clarity and predictability in light of the fact that the “Bankruptcy Code standardizes an expansive (and sometimes unruly) area of law”). For these reasons, as well as those set forth in Part III, we conclude that Congress did not authorize a “rare case” exception. We cannot “alter the balance struck by the statute,” Law v. Siegel, 571 U. S. ___, ___ (2014) (slip op., at 11), not even in “rare cases.” Cf. Norwest Bank Worthington v. Ahlers, 485 U.S. 197, 207 (1988) (explaining that courts cannot deviate from the procedures “specified by the Code,” even when they sincerely “believ[e] that . . . creditors would be better off”). The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

582.US.2016_16-6219

In petitioner’s state capital murder trial, the trial court overruled counsel’s objection to a proposed jury instruction and submitted the instruction to the jury, which convicted petitioner. Appellate counsel did not challenge the jury instruction, and petitioner’s conviction and sentence were affirmed. Petitioner’s state habeas counsel did not raise the instructional issue or challenge appellate counsel’s failure to raise it on appeal, and the state habeas court denied relief. Petitioner then sought federal habeas relief. Invoking Martinez v. Ryan, 566 U. S. 1 , and Trevino v. Thaler, 569 U. S. 413 , petitioner argued that his state habeas counsel’s ineffective assistance in failing to raise an ineffective-assistance-of-appellate-counsel claim provided cause to excuse the procedural default of that claim. The District Court denied relief, concluding that Martinez and Trevino apply exclusively to ineffective-assistance-of-trial-counsel claims. The Fifth Circuit denied a certificate of appealability. Held: The ineffective assistance of postconviction counsel does not provide cause to excuse the procedural default of ineffective-assistance-of-appellate-counsel claims. Pp. 4–16. (a) In Coleman v. Thompson, 501 U. S. 722 , this Court held that attorney error committed in the course of state postconviction proceedings—for which the Constitution does not guarantee the right to counsel—cannot supply cause to excuse a procedural default that occurs in those proceedings. Id., at 755. In Martinez, the Court announced an “equitable . . . qualification” of Coleman’s rule that applies where state law requires a claim of ineffective assistance of trial counsel to be raised in an “initial-review collateral proceeding,” rather than on direct appeal. 566 U. S., at 16, 17. In those situations, “a procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if” the default results from the ineffective assistance of the prisoner’s counsel in the collateral proceeding. Id., at 17. The Court clarified in Trevino that Martinez’s exception also applies where the State’s “procedural framework, by reason of its design and operation, makes it unlikely in a typical case that a defendant will have a meaningful opportunity to raise” the claim on direct appeal. 569 U. S., at ___. Pp. 4–7. (b) This Court declines to extend the Martinez exception to allow a federal court to hear a substantial, but procedurally defaulted, claim of appellate ineffectiveness when a prisoner’s state postconviction counsel provides ineffective assistance by failing to raise it. Pp. 7–16. (1) Martinez itself does not support extending this exception to new categories of procedurally defaulted claims. The Martinez Court did not purport to displace Coleman as the general rule governing procedural default. Rather, it “qualifie[d] Coleman by recognizing a narrow exception,” 566 U. S., at 9, and made clear that “[t]he rule of Coleman governs in all but th[ose] limited circumstances,” id., at 16. Applying Martinez’s highly circumscribed, equitable exception to new categories of procedurally defaulted claims would do precisely what this Court disclaimed in that case. P. 7. (2) Martinez’s underlying rationale does not support extending its exception to appellate-ineffectiveness claims. Petitioner argues that his situation is analogous to Martinez, where the Court expressed concern that trial-ineffectiveness claims might completely evade review. The Court in Martinez made clear, however, that it exercised its equitable discretion in view of the unique importance of protecting a defendant’s trial rights, particularly the right to effective assistance of trial counsel. Declining to expand Martinez to the appellate-ineffectiveness context does no more than respect that judgment. Nor is petitioner’s rule required to ensure that meritorious claims of trial error receive review by at least one state or federal court—Martinez’s chief concern. See 566 U. S., at 10, 12. A claim of trial error, preserved by trial counsel but not raised by counsel on appeal, will have been addressed by the trial court. If an unpreserved trial error was so obvious that appellate counsel was constitutionally required to raise it on appeal, then trial counsel likely provided ineffective assistance by failing to raise it at trial. In that circumstance, the prisoner likely could invoke Martinez or Coleman to obtain review of trial counsel’s failure to object. Similarly, if the underlying, defaulted claim of trial error was ineffective assistance of trial counsel premised on something other than the failure to object, then Martinez and Coleman again already provide a vehicle for obtaining review of that error in most circumstances. Pp. 7–11. (3) The equitable concerns addressed in Martinez do not apply to appellate-ineffectiveness claims. In Martinez and Trevino, the States deliberately chose to make postconviction process the only means for raising trial-ineffectiveness claims. The Court determined that it would be inequitable to refuse to hear a defaulted claim when the State had channeled that claim to a forum where the prisoner might lack the assistance of counsel in raising it. The States have not made a similar choice with respect to appellate-ineffectiveness claims—nor could they, since such claims generally cannot be presented until after the termination of direct appeal. The fact that appellate-ineffectiveness claims are considered in proceedings in which counsel is not constitutionally guaranteed is a function of the nature of the claim, not of the States’ deliberate choices. Pp. 11–12. (4) The Martinez decision was also grounded in part on the belief that its narrow exception was unlikely to impose significant systemic costs. See 566 U. S., at 15–16. But adopting petitioner’s proposed extension could flood the federal courts with defaulted appellate-ineffectiveness claims, and potentially serve as a gateway to federal review of a host of defaulted claims of trial error. It would also aggravate the harm to federalism that federal habeas review of state convictions necessarily causes. Not only would these burdens on the federal courts and federal system be severe, but the systemic benefit would be small, as claims heard in federal court solely by virtue of petitioner’s proposed rule would likely be largely meritless. Pp. 12–16. 650 Fed. Appx. 860, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Alito, and Gorsuch, JJ., joined. Breyer, J., filed a dissenting opinion, in which Ginsburg, Sotomayor, and Kagan, JJ., joined.

Federal habeas courts reviewing convictions from state courts will not consider claims that a state court refused to hear based on an adequate and independent state procedural ground. A state prisoner may be able to overcome this bar, however, if he can establish “cause” to excuse the procedural default and demonstrate that he suffered actual prejudice from the alleged error. An attorney error does not qualify as “cause” to excuse a procedural default unless the error amounted to constitutionally ineffective assistance of counsel. Because a prisoner does not have a constitutional right to counsel in state postconviction proceedings, ineffective assistance in those proceedings does not qualify as cause to excuse a procedural default. See Coleman v. Thompson, 501 U. S. 722 (1991) . In Martinez v. Ryan, 566 U. S. 1 (2012) , and Trevino v. Thaler, 569 U. S. 413 (2013) , this Court announced a narrow exception to Coleman’s general rule. That exception treats ineffective assistance by a prisoner’s state postconviction counsel as cause to overcome the default of a single claim—ineffective assistance of trial counsel—in a single context—where the State effectively requires a defendant to bring that claim in state postconviction proceedings rather than on direct appeal. The question in this case is whether we should extend that exception to allow federal courts to consider a different kind of defaulted claim—ineffective assistance of appellate counsel. We decline to do so. I A On April 6, 2008, a group of family and friends gathered at Annette Stevenson’s home to celebrate her granddaughter’s birthday. Petitioner Erick Daniel Davila, believing he had seen a member of a rival street gang at the celebration, fired a rifle at the group while they were eating cake and ice cream. He shot and killed Annette and her 5-year-old granddaughter Queshawn, and he wounded three other children and one woman. After the police arrested petitioner, he confessed to the killings. He stated that he “wasn’t aiming at the kids or the woman,” but that he was trying to kill Annette’s son (and Queshawn’s father) Jerry Stevenson and the other “guys on the porch.” App. 38. The other “guys on the porch” were, apparently, women. The State indicted petitioner for capital murder under Tex. Penal Code Ann. §19.03(a)(7)(A) (West 2016), which makes it a capital crime to “murde[r] more than one person . . . during the same criminal transaction.” In response to the jury’s request for clarification during deliberations, the trial court proposed instructing the jury on transferred intent. Under that doctrine, the jury could find petitioner guilty of murder if it determined that he intended to kill one person but instead killed a different person. Petitioner’s counsel objected to the additional instruction, arguing that the trial judge should “wait” to submit it “until the jury indicates that they can’t reach . . . a resolution.” App. 51. The trial court overruled the objection and submitted the instruction to the jury. The jury convicted petitioner of capital murder, and the trial court sentenced petitioner to death. B Petitioner appealed his conviction and sentence. Al-though his appellate counsel argued that the State presented insufficient evidence to show that he acted with the requisite intent, counsel did not challenge the instruction about transferred intent. The Texas Court of Criminal Appeals affirmed petitioner’s conviction and sentence. Davila v. State, 2011 WL 303265 (Jan. 26, 2011), cert. denied, 565 U. S. 885 (2011) . Petitioner next sought habeas relief in Texas state court. His counsel did not challenge the instruction about transferred intent, nor did he challenge the failure of his appellate counsel to raise the alleged instructional error on direct appeal. The Texas Court of Criminal Appeals denied relief. Ex parte Davila, 2013 WL 1655549 (Apr. 17, 2013), cert. denied, 571 U. S. ___ (2013). C Petitioner then sought habeas relief in Federal District Court under 28 U. S. C. §2254. As relevant here, he argued that his appellate counsel provided ineffective assistance by failing to challenge the jury instruction about transferred intent. Petitioner conceded that he had failed to raise his claim of ineffective assistance of appellate counsel in his state habeas petition, but argued that the failure was the result of his state habeas counsel’s ineffective assistance. Petitioner invoked this Court’s decisions in Martinez and Trevino to argue that his state habeas attorney’s ineffective assistance provided cause to excuse the procedural default of his claim of ineffective assistance of appellate counsel. The District Court denied petitioner’s §2254 petition. It concluded that Martinez and Trevino did not supply cause to excuse the procedural default of petitioner’s claim of ineffective assistance of appellate counsel because those decisions applied exclusively to claims of ineffective assistance of trial counsel. See Davila v. Stephens, 2015 WL 1808689, *20 (ND Tex., Apr. 21, 2015). The Court of Appeals for the Fifth Circuit denied a certificate of appealability on the same ground. 650 Fed. Appx. 860, 867–868 (2016). Petitioner then sought a writ of certiorari, asking us to reverse the Fifth Circuit on the ground that Martinez and Trevino should be extended to claims of ineffective assistance of appellate counsel. We granted certiorari, 580 U. S. ___ (2017), and now affirm. II Our decision in this case is guided by two fundamental tenets of federal review of state convictions. First, a state prisoner must exhaust available state remedies before presenting his claim to a federal habeas court. §2254(b)(1)(A). The exhaustion requirement is designed to avoid the “unseemly” result of a federal court “upset[ting] a state court conviction without” first according the state courts an “opportunity to . . . correct a constitutional violation,” Rose v. Lundy, 455 U. S. 509, 518 (1982) (internal quotation marks omitted). Second, a federal court may not review federal claims that were procedurally defaulted in state court—that is, claims that the state court denied based on an adequate and independent state procedural rule. E.g., Beard v. Kindler, 558 U. S. 53, 55 (2009) . This is an important “corollary” to the exhaustion requirement. Dretke v. Haley, 541 U. S. 386, 392 (2004) . “Just as in those cases in which a state prisoner fails to exhaust state remedies, a habeas petitioner who has failed to meet the State’s procedural requirements for presenting his federal claims has deprived the state courts of an opportunity to address” the merits of “those claims in the first instance.” Coleman, 501 U. S., at 731–732.[1] The procedural default doctrine thus advances the same comity, finality, and federalism interests advanced by the exhaustion doctrine. See McCleskey v. Zant, 499 U. S. 467, 493 (1991) . A state prisoner may overcome the prohibition on reviewing procedurally defaulted claims if he can show “cause” to excuse his failure to comply with the state procedural rule and “actual prejudice resulting from the alleged constitutional violation.” Wainwright v. Sykes, 433 U. S. 72, 84 (1977) ; Coleman, supra, at 750. To establish “cause”—the element of the doctrine relevant in this case—the prisoner must “show that some objective factor external to the defense impeded counsel’s efforts to comply with the State’s procedural rule.” Murray v. Carrier, 477 U. S. 478, 488 (1986) . A factor is external to the defense if it “cannot fairly be attributed to” the prisoner. Coleman, supra, at 753. It has long been the rule that attorney error is an objective external factor providing cause for excusing a procedural default only if that error amounted to a deprivation of the constitutional right to counsel. See Edwards v. Carpenter, 529 U. S. 446, 451 (2000) . An error amounting to constitutionally ineffective assistance is “imputed to the State” and is therefore external to the prisoner. Murray, supra, at 488. Attorney error that does not violate the Constitution, however, is attributed to the prisoner “under well-settled principles of agency law.” Coleman, supra, at 754. It follows, then, that in proceedings for which the Constitution does not guarantee the assistance of counsel at all, attorney error cannot provide cause to excuse a default. Thus, in Coleman, this Court held that attorney error committed in the course of state postconviction proceedings—for which the Constitution does not guarantee the right to counsel, see Murray v. Giarratano, 492 U. S. 1 (1989) (plurality opinion)—cannot supply cause to excuse a procedural default that occurs in those proceedings. 501 U. S., at 755. In Martinez, this Court announced a narrow, “equitable . . . qualification” of the rule in Coleman that applies where state law requires prisoners to raise claims of ineffective assistance of trial counsel “in an initial-review collateral proceeding,” rather than on direct appeal. Martinez, 566 U. S., at 16, 17. It held that, in those situations, “a procedural default will not bar a federal habeas court from hearing a substantial claim of ineffective assistance at trial if” the default results from the ineffective assistance of the prisoner’s counsel in the collateral proceeding. Id., at 17. In Trevino, the Court clarified that this exception applies both where state law explicitly prohibits prisoners from bringing claims of ineffective assistance of trial counsel on direct appeal and where the State’s “procedural framework, by reason of its design and operation, makes it unlikely in a typical case that a defendant will have a meaningful opportunity to raise” that claim on direct appeal. 569 U. S., at ___ (slip op., at 14). III Petitioner asks us to extend Martinez to allow a federal court to hear a substantial, but procedurally defaulted, claim of ineffective assistance of appellate counsel when a prisoner’s state postconviction counsel provides ineffective assistance by failing to raise that claim. We decline to do so. A On its face, Martinez provides no support for extending its narrow exception to new categories of procedurally defaulted claims. Martinez did not purport to displace Coleman as the general rule governing procedural default. Rather, it “qualifie[d] Coleman by recognizing a narrow exception” that applies only to claims of “ineffective assistance of counsel at trial” and only when, “under state law,” those claims “must be raised in an initial-review collateral proceeding.” Martinez, supra, at 9, 17. And Trevino merely clarified that the exception applies whether state law explicitly or effectively forecloses review of the claim on direct appeal. 569 U. S., at ___ (slip op. at 2, 13). In all but those “limited circumstances,” Martinez made clear that “[t]he rule of Coleman governs.” 566 U. S., at 16. Applying Martinez’s highly circumscribed, equitable exception to new categories of procedurally defaulted claims would thus do precisely what this Court disclaimed in Martinez: Replace the rule of Coleman with the exception of Martinez. B Petitioner also finds no support in the underlying rationale of Martinez. Petitioner’s primary argument is that his claim of ineffective assistance of appellate counsel might never be reviewed by any court, state or federal, without expanding the exception to the rule in Coleman. He argues that this situation is analogous to Martinez, where the Court expressed that same concern about claims of ineffective assistance of trial counsel. But the Court in Martinez was principally concerned about trial errors—in particular, claims of ineffective assistance of trial counsel. Ineffective assistance of appellate counsel is not a trial error. Nor is petitioner’s rule necessary to ensure that a meritorious trial error (of any kind) receives review. 1 Petitioner argues that allowing a claim of ineffective assistance of appellate counsel to evade review is just as concerning as allowing a claim of ineffective assistance of trial counsel to evade review. Brief for Petitioner 12; see also id., at 18–26. We do not agree. The criminal trial enjoys pride of place in our criminal justice system in a way that an appeal from that trial does not. The Constitution twice guarantees the right to a criminal trial, see Art. III, §2; Amdt. 6, but does not guarantee the right to an appeal at all, Halbert v. Michigan, 545 U. S. 605, 610 (2005) . The trial “is the main event at which a defendant’s rights are to be determined,” McFarland v. Scott, 512 U. S. 849, 859 (1994) (internal quotation marks omitted), “and not simply a tryout on the road to appellate review,” Freytag v. Commissioner, 501 U. S. 868, 895 (1991) (Scalia, J., concurring in part and concurring in judgment) (internal quotation marks omitted). And it is where the stakes for the defendant are highest, not least because it is where a presumptively innocent defendant is adjudged guilty, see Ross v. Moffitt, 417 U. S. 600, 610 (1974) ; Wainwright, 433 U. S., at 90, and where the trial judge or jury makes factual findings that nearly always receive deference on appeal and collateral review, see Jackson v. Virginia, 443 U. S. 307 –319 (1979); see also Cavazos v. Smith, 565 U. S. 1, 2 (2011) (per curiam) (under deferential standard of review, “judges will sometimes encounter convictions that they believe to be mistaken, but that they must nevertheless uphold”). The Court in Martinez made clear that it exercised its equitable discretion in view of the unique importance of protecting a defendant’s trial rights, particularly the right to effective assistance of trial counsel. As the Court explained, “the limited nature” of its holding “reflect[ed] the importance of the right to the effective assistance of trial counsel,” which is “a bedrock principle in our justice system.” 566 U. S., at 12, 16 (emphasis added). In declining to expand the Martinez exception to the distinct context of ineffective assistance of appellate counsel, we do no more than respect that judgment. 2 Petitioner’s rule also is not required to ensure that meritorious claims of trial error receive review by at least one state or federal court—the chief concern identified by this Court in Martinez. See id., at 10, 12. Martinez was concerned that a claim of trial error—specifically, ineffective assistance of trial counsel—might escape review in a State that required prisoners to bring the claim for the first time in state postconviction proceedings rather than on direct appeal. Because it is difficult to assess a trial attorney’s performance until the trial has ended, a trial court ordinarily will not have the opportunity to rule on such a claim. And when the State requires a prisoner to wait until postconviction proceedings to raise the claim, the appellate court on direct appeal also will not have the opportunity to review it. If postconviction counsel then fails to raise the claim, no state court will ever review it. Finally, because attorney error in a state postconviction proceeding does not qualify as cause to excuse procedural default under Coleman, no federal court could consider the claim either. Claims of ineffective assistance of appellate counsel, however, do not pose the same risk that a trial error—of any kind—will escape review altogether, at least in a way that could be remedied by petitioner’s proposed rule. This is true regardless of whether trial counsel preserved the alleged error at trial. If trial counsel preserved the error by properly objecting, then that claim of trial error “will have been addressed by . . . the trial court.” Martinez, 566 U. S., at 11. A claim of appellate ineffectiveness premised on a preserved trial error thus does not present the same concern that animated the Martinez exception because at least “one court” will have considered the claim on the merits. Ibid.; see also Coleman, 501 U. S., at 755–756. If trial counsel failed to preserve the error at trial, then petitioner’s proposed rule ordinarily would not give the prisoner access to federal review of the error, anyway. Effective appellate counsel should not raise every nonfrivolous argument on appeal, but rather only those arguments most likely to succeed. Smith v. Murray, 477 U. S. 527, 536 (1986) ; Jones v. Barnes, 463 U. S. 745 –753 (1983). Declining to raise a claim on appeal, therefore, is not deficient performance unless that claim was plainly stronger than those actually presented to the appellate court. See Smith v. Robbins, 528 U. S. 259, 288 (2000) . In most cases, an unpreserved trial error will not be a plainly stronger ground for appeal than preserved errors. See 2 B. Means, Postconviction Remedies §35:19, p. 627, and n. 16 (2016). Thus, in most instances in which the trial court did not rule on the alleged trial error (because it was not preserved), the prisoner could not make out a substantial claim of ineffective assistance of appellate counsel and therefore could not avail himself of petitioner’s expanded Martinez exception. Adopting petitioner’s proposed rule would be unnecessary to ensure review of a claim of trial error even when a prisoner has a legitimate claim of ineffective assistance of appellate counsel based on something other than a preserved trial error. If an unpreserved trial error was so obvious that appellate counsel was constitutionally required to raise it on appeal, then trial counsel likely provided ineffective assistance by failing to object to it in the first instance. In that circumstance, the prisoner likely could invoke Martinez or Coleman to obtain review of trial counsel’s failure to object. Similarly, if the underlying, defaulted claim of trial error was ineffective assistance of trial counsel premised on something other than the failure to object, then Martinez and Coleman again already provide a vehicle for obtaining review of that error in most circumstances. Petitioner’s proposed rule is thus unnecessary for ensuring that trial errors are reviewed by at least one court. C The Court in Martinez also was responding to an equitable consideration that is unique to claims of ineffective assistance of trial counsel and accordingly inapplicable to claims of ineffective assistance of appellate counsel. In Martinez, the State “deliberately cho[se] to move trial-ineffectiveness claims outside of the direct-appeal process, where counsel is constitutionally guaranteed,” into the postconviction review process, where we have never held that the Constitution guarantees a right to counsel. 566 U. S., at 13; id., at 9. By doing so, “the State significantly diminishe[d] prisoners’ ability to file such claims.” Id., at 13. Similarly, in Trevino, the State had chosen a procedural framework pursuant to which collateral review was, “as a practical matter, the onl[y] method for raising an ineffective-assistance-of-trial-counsel claim.” 569 U. S., at ___ (slip op., at 13). Although this Court acknowledged in Martinez that there was nothing inappropriate about the State’s choice, it explained that the choice was “not without consequences for the State’s ability to assert a procedural default” in subsequent federal habeas proceedings. 566 U. S., at 13. Specifically, the Court concluded that it would be inequitable to refuse to hear a defaulted claim of ineffective assistance of trial counsel when the State had channeled that claim to a forum where the prisoner might lack the assistance of counsel in raising it. The States have not made a similar choice with respect to claims of ineffective assistance of appellate counsel—nor could they. By their very nature, such claims gener-ally cannot be presented until after the termination of direct appeal. Put another way, they necessarily must be heard in collateral proceedings, where counsel is not constitutionally guaranteed. The fact that claims of appellate ineffectiveness are considered in proceedings in which counsel is not constitutionally guaranteed is a function of the nature of the claim, not of the State’s “deliberat[e] cho[ice] to move . . . claims outside of the direct-appeal process.” Ibid. The equitable concerns raised in Martinez therefore do not apply. D Finally, the Court in Martinez grounded its decision in part on the belief that its narrow exception was unlikely to impose significant systemic costs. See id., at 15–16. The same cannot be said of petitioner’s proposed extension. 1 Adopting petitioner’s argument could flood the federal courts with defaulted claims of appellate ineffectiveness. For one thing, every prisoner in the country could bring these claims. Martinez currently applies only to States that deliberately choose to channel claims of ineffective assistance of trial counsel into collateral proceedings. See, e.g., Lee v. Corsini, 777 F. 3d 46, 60–61 (CA1 2015) (Martinez and Trevino do not apply to Massachusetts); Henness v. Bagley, 766 F. 3d 550, 557 (CA6 2014) (Martinez does not apply to Ohio). If we applied Martinez to claims of appellate ineffectiveness, however, we would bring every State within Martinez’s ambit, because claims of appellate ineffectiveness necessarily must be heard in collateral proceedings. See supra, at 12. Extending Martinez to defaulted claims of ineffective assistance of appellate counsel would be especially troublesome because those claims could serve as the gatewayto federal review of a host of trial errors, while Martinez covers only one trial error (ineffective assistance of trial counsel). If a prisoner can establish ineffective assistance of trial counsel under Martinez, he ordinarily is entitled to a new trial. See United States v. Morrison, 449 U. S. 361 –365 (1981); see also Hagens v. State, 979 S. W. 2d 788, 792 (Tex. App. 1998). But if he cannot, Martinez provides no avenue for litigating other defaulted trial errors.[2] An expanded Martinez exception, however, would mean that any defaulted trial error could result in a new trial. In Carpenter, this Court held that, when a prisoner can show cause to excuse a defaulted claim of ineffective assistance of appellate counsel, he can in turn rely on that claim as cause to litigate an underlying claim of trial error that was defaulted due to appellate counsel’s ineffectiveness. 529 U. S., at 453. Expanding Martinez as petitioner suggests would thus produce a domino effect: Prisoners could assert their postconviction counsel’s inadequacy as cause to excuse the default of their appellate ineffectiveness claims, and use those newly reviewable appellate ineffectiveness claims as cause to excuse the default of their underlying claims of trial error. Petitioner’s rule thus could ultimately knock down the procedural barriers to federal habeas review of nearly any defaulted claim of trial error. The scope of that review would exceed anything the Martinez Court envisioned when it established its narrow exception to Coleman. Petitioner insists that these concerns are overstated because many of the newly raised claims will be meritless. See Brief for Petitioner 28. But even if that were true, courts would still have to undertake the task of separating the wheat from the chaff. And we are not reassured by petitioner’s suggestion that extending Martinez would increase only the number of claims in each petition rather than the number of federal habeas petitions themselves. Reply Brief 14. Each additional claim would require the district court to review the prisoner’s trial record, appellate briefing, and state postconviction record to determine the claim’s viability. This effort could be repeated at each level of federal review. We cannot “assume that these costs would be negligible,” Murray, 477 U. S., at 487, and we are loath to further “burden . . . scarce federal judicial resources” in this way, McCleskey, 499 U. S., at 491. 2 Expanding Martinez would not only impose significant costs on the federal courts, but would also aggravate the harm to federalism that federal habeas review necessarily causes. Federal habeas review of state convictions “entails significant costs,” Engle v. Isaac, 456 U. S. 107, 126 (1982) , “ ‘and intrudes on state sovereignty to a degree matched by few exercises of federal judicial authority,’ ” Harrington v. Richter, 562 U. S. 86, 103 (2011) (quoting Harris v. Reed, 489 U. S. 255, 282 (1989) (Kennedy, J., dissenting)). It “frustrates both the States’ sovereign power to punish offenders and their good-faith attempts to honor constitutional rights.” Calderon v. Thompson, 523 U. S. 538 –556 (1998) (internal quotation marks omitted). It “degrades the prominence of the [State] trial,” Engle, supra, at 127, and it “disturbs the State’s significant interest in repose for concluded litigation [and] denies society the right to punish some admitted offenders,” Harrington, supra, at 103 (internal quotation marksomitted). Apart from increasing the sheer frequency of federal intrusion into state criminal affairs, petitioner’s proposed rule would also undermine the doctrine of procedural default and the values it serves. That doctrine, like the federal habeas statute generally, is designed to ameliorate the injuries to state sovereignty that federal habeas review necessarily inflicts by giving state courts the first opportunity to address challenges to convictions in state court, thereby “promoting comity, finality, and federalism.” Cullen v. Pinholster, 563 U. S. 170, 185 (2011) ; McCleskey, supra, at 493. Expanding the narrow exception announced in Martinez would unduly aggravate the “special costs on our federal system” that federal habeas review already imposes. Engle, supra, at 128. 3 Not only would these burdens on the federal courts and our federal system be severe, but the benefit would—as a systemic matter—be small. To be sure, permitting a state prisoner to bring a meritorious constitutional claim that could not otherwise be heard is beneficial to that prisoner. Petitioner’s counsel concedes, however, that relief is granted in, “[i]f any, a very minute number” of “post-conviction ineffective assistance of appellate counselcases.” Tr. of Oral Arg. 14. Indeed, he concedes that the number of meritorious cases is “infinitesimally small.” Ibid. We think it is likely that the claims heard in federal court because of petitioner’s proposed rule would also be largely meritless, given that the proposed rule would generally affect only those cases in which the trial court already adjudicated, and rejected, the prisoner’s argument regarding the alleged underlying trial error. See supra, at 11. Given that petitioner’s proposed rule would likely generate high systemic costs and low systemic benefits, and that the unique concerns of Martinez are not implicated in cases like his, we do not think equity requires anexpansion of Martinez. * * * For the foregoing reasons, we affirm the judgment of the Court of Appeals. It is so ordered.Notes 1 The Fifth Circuit treats unexhausted claims as procedurally defaulted if “the court to which the petitioner would be required to present his claims in order to meet the exhaustion requirement would now find the claims procedurally barred.” Bagwell v. Dretke, 372 F. 3d 748, 755 (2004) (internal quotation marks omitted); cf. Coleman, 501 U. S., at 735, n. Relying on this doctrine, the District Court concluded that petitioner’s federal claim was procedurally defaulted (even though a state court had never actually found it procedurally barred) because Texas law would likely bar a Texas court from deciding the claim on the merits if petitioner were to present it in a successive habeas petition. Davila v. Stephens, 2015 WL 1808689, *19–*20 (ND Tex., Apr. 21, 2015) (citing Davila v. Stephens, 2014 WL 5879879, *2 (ND Tex., Nov. 10, 2014)); see also Davila v. Stephens, 2014 WL 6057907, *2 (ND Tex., Nov. 10, 2014). Petitioner did not seek a certificate of appealability regarding that holding, and neither petitioner nor the State disputes in this Court that the claim was procedurally defaulted. Accordingly, we assume that it was procedurally defaulted for purposes of this case. 2 The dissent argues that Martinez already provides a gateway to the review of underlying trial errors no differently than would petitioner’s proposed rule. See post, at 7 (opinion of Breyer, J.). That is not so. If a prisoner succeeds on his claim of ineffective assistance of trial counsel under Martinez, the federal habeas court would not need to consider any other claim of trial error since the successful claim of trial ineffectiveness—unlike a successful claim of ineffective assistance of appellate counsel—entitles the prisoner to a new trial. See 7 W. LaFave, J. Israel, N. King, & O. Kerr, Criminal Procedure §28.4(d), p. 258, n. 75 (4th ed. 2015).

581.US.2016_15-9260

Petitioner Dean and his brother committed two robberies of drug dealers. During each robbery, Dean’s brother threatened and assaulted the victim with a gun, while Dean searched the premises for valuables. Dean was convicted of multiple robbery and firearms counts, as well as two counts of possessing a firearm in furtherance of a crime of violence, in violation of 18 U. S. C. §924(c). Section 924(c) criminalizes using or carrying a firearm during and in relation to a crime of violence or drug trafficking crime, or possessing a firearm in furtherance of such an underlying crime. That provision mandates a distinct penalty to be imposed “in addition to the punishment provided for [the predicate] crime,” §924(c)(1)(A). Further, §924(c) says that any sentence mandated by that provision must run consecutively to “any other term of imprisonment imposed on the person,” including any sentence for the predicate crime, §924(c)(1)(D)(ii). A first conviction under §924(c) carries a five-year mandatory minimum penalty, §924(c)(1)(A)(i), while a second conviction carries an additional 25-year mandatory minimum, §924(c)(1)(C)(i). For Dean, that meant a 30-year mandatory minimum, to be served after and in addition to any sentence he received for his other counts of conviction. At sentencing, Dean urged the District Court to consider his lengthy mandatory minimum sentences when calculating the sentences for his other counts and to impose concurrent one-day sentences for those counts. The judge said he would have agreed to Dean’s request but understood §924(c) to preclude a sentence of 30 years plus one day. On appeal, Dean argued that the District Court had erred in concluding that it could not vary from the Guidelines range based on the mandatory minimum sentences he would receive under §924(c). The Court of Appeals ruled that Dean’s argument was foreclosed by Circuit precedent and that his sentence was otherwise substantively reasonable. Held: Section 924(c) does not prevent a sentencing court from considering a mandatory minimum imposed under that provision when calculating an appropriate sentence for the predicate offense. Pp. 3–8. (a) Sentencing courts have long enjoyed discretion in the sort of information they may consider when setting an appropriate sentence, and they continue to do so even as federal laws have required them to evaluate certain factors when exercising their discretion. Pepper v. United States, 562 U. S. 476 –489. Section 3553(a) specifies the factors courts are to consider when imposing a sentence. They include “the nature and circumstances of the offense and the history and characteristics of the defendant,” as well as “the need for the sentence imposed” to serve the four overarching aims of sentencing: just punishment, deterrence, protection of the public, and rehabilitation. The §3553(a) factors are used to set both the length of separate prison terms, §3582(a), and an aggregate prison term comprising separate sentences for multiple counts of conviction, §3584(b). As a general matter, these sentencing provisions permit a court imposing a sentence on one count of conviction to consider sentences imposed on other counts. The Government argues that district courts should calculate the appropriate term of imprisonment for each individual offense, disregarding whatever sentences a defendant may face on other counts. Only when determining an aggregate prison sentence, the Government maintains, should a district court consider the effect of those other sentences. Nothing in the law requires such an approach. There is no reason that the §3553(a) factors may not also be considered when determining a prison sentence for each individual offense in a multicount case. The Government’s interpretation is at odds not only with the text of those provisions but also with the Government’s own practice in “sentencing package cases.” Greenlaw v. United States, 554 U. S. 237 . Pp. 3–6. (b) The Government points to two limitations in §924(c) that, in its view, restrict the authority of sentencing courts to consider a sentence imposed under §924(c) when calculating a just sentence for the predicate count. Neither limitation supports the Government’s position. First, that a mandatory sentence under §924(c) must be imposed “in addition to the punishment provided” for the predicate crime says nothing about the length of a non-§924(c) sentence, much less about what information a court may consider in determining that sentence. Second, nothing in the requirement of consecutive sentences prevents a district court from imposing a 30-year mandatory minimum sentence under §924(c) and a one-day sentence for the predicate crime, provided those terms run one after the other. The Government would, in effect, have this Court read into §924(c) the limitation explicitly made in §1028A(b)(3), which provides that in determining the appropriate length of imprisonment for a predicate felony, “a court shall not in any way reduce the term to be imposed for such crime so as to compensate for, or otherwise take into account, any separate term of imprisonment imposed or to be imposed for a violation of this section.” But “[d]rawing meaning from silence is particularly inappropriate” where, as demonstrated in §1028A, “Congress has shown that it knows how to direct sentencing practices in express terms.” Kimbrough v. United States, 552 U. S. 85 . Pp. 6–8. 810 F. 3d 521, reversed and remanded. Roberts, C. J., delivered the opinion for a unanimous Court.

Congress has made it a separate offense to use or possess a firearm in connection with a violent or drug trafficking crime. 18 U. S. C. §924(c). That separate firearm offense carries a mandatory minimum sentence of five years for the first conviction and 25 years for a second. Those sentences must be in addition to and consecutive to the sentence for the underlying predicate offense. The question presented is whether, in calculating the sentence for the predicate offense, a judge must ignore the fact that the defendant will serve the mandatory minimums imposed under §924(c). I Levon Dean, Jr., and his brother robbed a methamphetamine dealer in a Sioux City motel room. Less than two weeks later, they robbed another drug dealer at his home. During each robbery, Dean’s brother threatened the victim with a modified semiautomatic rifle, later using that rifle to club the victim on the head. Dean, meanwhile, ransacked the area for drugs, money, and other valuables. A federal grand jury returned a multicount indictment charging Dean and his brother with a host of crimes related to the two robberies. Following a joint trial, a jury convicted Dean of one count of conspiracy to commit robbery, two counts of robbery, and one count of possessing a firearm as a convicted felon. He was also convicted of two counts of possessing and aiding and abetting the possession of a firearm in furtherance of a crime of violence, in violation of 18 U. S. C. §§2 and 924(c). Section 924(c) criminalizes using or carrying a firearm during and in relation to a crime of violence or drug trafficking crime, or possessing a firearm in furtherance of such an underlying crime. There is no dispute that Dean’s two robbery convictions qualified as predicate crimes of violence for purposes of §924(c). Section 924(c) does more than create a distinct offense. It also mandates a distinct penalty, one that must be imposed “in addition to the punishment provided for [the predicate] crime of violence or drug trafficking crime.” §924(c)(1)(A) (emphasis added). A first-time offender under §924(c) receives a five-year mandatory minimum. A “second or subsequent conviction” under §924(c) carries an additional 25-year mandatory minimum. §§924(c)(1)(A)(i), (C)(i). A sentence imposed under §924(c) must run consecutively to “any other term of imprisonment imposed on the person,” including any sentence for the predicate crime “during which the firearm was used, carried, or possessed.” §924(c)(1)(D)(ii). For Dean, this meant a 30-year mandatory minimum, to be served after and in addition to any sentence he received for his other counts of conviction. At sentencing Dean did not dispute that each of his four other counts resulted in a sentencing range of 84–105 months under the Sentencing Guidelines. He argued, however, that the court should consider his lengthy mandatory minimum sentences when calculating the sentences for his other counts, and impose concurrent one-day sentences for those counts. Finding that Dean was “clearly the follower” and that he lacked “any significant history of any violence,” the District Judge agreed that 30 years plus one day was “more than sufficient for a sentence in this case.” App. 26. Yet the judge understood §924(c) to preclude such a sentence. In his view, he was required to disregard Dean’s 30-year mandatory minimum when determining the appropriate sentences for Dean’s other counts of conviction. Viewed on their own—and not as part of a combined package—those counts plainly warranted sentences longer than one day. In the end, the judge still granted a significant downward variance from the 84–105 month Guidelines range. Dean received concurrent sentences of 40 months for each non-§924(c) conviction, which, when added to his 360-month mandatory minimum, yielded a total sentence of 400 months. Dean appealed. Before the Eighth Circuit, Dean argued that the District Court had erred in concluding that it could not vary from the Guidelines range based on the mandatory minimum sentences he would receive under §924(c). The Court of Appeals disagreed, ruling that Dean’s argument was foreclosed by Circuit precedent and that his sentence was otherwise substantively reasonable. 810 F. 3d 521 (2015). We granted certiorari. 580 U. S. ___ (2016). II Sentencing courts have long enjoyed discretion in the sort of information they may consider when setting an appropriate sentence. Pepper v. United States, 562 U. S. 476 –489 (2011). This durable tradition remains, even as federal laws have required sentencing courts to evaluate certain factors when exercising their discretion. Ibid. A Section 3553(a) of Title 18 specifies the factors courts are to consider in imposing a sentence. The list of factors is preceded by what is known as the parsimony principle, a broad command that instructs courts to “impose a sentence sufficient, but not greater than necessary, to comply with” the four identified purposes of sentencing: just punishment, deterrence, protection of the public, and rehabilitation. Ibid. A sentencing court is then directed to take into account “the nature and circumstances of the offense and the history and characteristics of the defendant,” as well as “the need for the sentence imposed” to serve the four overarching aims of sentencing. §§3553(a)(1), (2)(A)–(D); see Gall v. United States, 552 U. S. 38 , n. 6 (2007). The court must also consider the pertinent guidelines and policies adopted by the Sentencing Commission. §§3553(a)(4), (5); see id., at 50, n. 6. The §3553(a) factors are used to set both the length of separate prison terms and an aggregate prison term comprising separate sentences for multiple counts of conviction. Under §3582 a court, “in determining whether to impose a term of imprisonment, and, if a term of imprisonment is to be imposed, in determining the length of the term, shall consider the factors set forth in section 3553(a).” §3582(a). And §3584 provides: “[I]n determining whether the terms imposed are to be ordered to run concurrently or consecutively, [the court] shall consider, as to each offense for which a term of imprisonment is being imposed, the factors set forth in section 3553(a).” §3584(b). As a general matter, the foregoing provisions permit a court imposing a sentence on one count of conviction to consider sentences imposed on other counts. Take the directive that a court assess “the need for the sentence imposed . . . to protect the public from further crimes of the defendant.” §3553(a)(2)(C). Dean committed the two robberies at issue here when he was 23 years old. That he will not be released from prison until well after his fiftieth birthday because of the §924(c) convictions surely bears on whether—in connection with his predicate crimes—still more incarceration is necessary to protect the public. Likewise, in considering “the need for the sentence imposed . . . to afford adequate deterrence,” §3553(a)(2)(B), the District Court could not reasonably ignore the deterrent effect of Dean’s 30-year mandatory minimum. According to the Government, this is not how sentencing is meant to work. Rather, district courts should calculate the appropriate term of imprisonment for each individual offense. That determination, insists the Government, disregards whatever sentences the defendant may also face on other counts. Not until deciding whether to run sentences consecutively or concurrently—i.e., not until applying §3584—should a district court consider the effect of those other sentences. Brief for United States 21–26. Nothing in the law requires such an approach. The Government states that the §3553(a) factors are “normally relevant in determining the total length of imprisonment” under §3584. Id., at 28. No doubt they are. But there is no reason they may not also be considered at the front end, when determining a prison sentence for each individual offense in a multicount case. At odds with the text, the Government’s interpretation is also at odds with its own practice in “sentencing package cases.” Greenlaw v. United States, 554 U. S. 237, 253 (2008) . “Those cases typically involve multicount indictments and a successful attack by a defendant on some but not all of the counts of conviction.” Ibid. In those cases—including ones where §924(c) convictions are invalidated—the Government routinely argues that an appellate court should vacate the entire sentence so that the district court may increase the sentences for any remaining counts up to the limit set by the original aggregate sentence. See United States v. Smith, 756 F. 3d 1179, 1188–1189, and n. 5 (CA10 2014) (collecting cases). And appellate courts routinely agree. Id., at 1189, and n. 6 (same). As we understand it, the Government’s theory in those cases is that the district court may have relied on a now-vacated conviction when imposing sentences for the other counts. But that theory of course directly contradicts the position the Government now advances—that district courts must determine sentences independently of one another, accounting for multiple sentences only when deciding whether to stack them or run them concurrently. B Nothing in §924(c) restricts the authority conferred on sentencing courts by §3553(a) and the related provisions to consider a sentence imposed under §924(c) when calculating a just sentence for the predicate count. The Government points to two limitations in §924(c). First, the Government notes, a mandatory sentence under §924(c) must be imposed “in addition to the punishment provided” for the predicate crime. §924(c)(1)(A) (emphasis added). This limitation says nothing about the length of a non-§924(c) sentence, much less about what information a court may consider in determining that sentence. Whether the sentence for the predicate offense is one day or one decade, a district court does not violate the terms of §924(c) so long as it imposes the mandatory minimum “in addition to” the sentence for the violent or drug trafficking crime. Second, §924(c) states that “no term of imprisonment imposed on a person under this subsection shall run concurrently with any other term of imprisonment imposed on the person, including any term of imprisonment imposed for the [predicate] crime during which the firearm was used, carried, or possessed.” §924(c)(1)(D)(ii). Nothing in that language prevents a district court from imposing a 30-year mandatory minimum sentence under §924(c) and a one-day sentence for the predicate violent or drug trafficking crime, provided those terms run one after the other. The Government emphasizes that the requirement of consecutive sentences removes the discretion to run sentences concurrently that district courts exercise under §3584. We agree. So does Dean, for that matter. But we fail to see the significance of the point. The bar on imposing concurrent sentences does not affect a court’s discretion to consider a mandatory minimum when calculating each individual sentence. The Government would, in effect, have us read an additional limitation into §924(c): Where §924(c) says “in addition to the punishment provided for [the predicate] crime of violence,” what the statute really means is “in addition to the punishment provided for [the predicate] crime of violence in the absence of a Section 924(c) conviction.” See Reply Brief 2. We have said that “[d]rawing meaning from silence is particularly inappropriate” where “Congress has shown that it knows how to direct sentencing practices in express terms.” Kimbrough v. United States, 552 U. S. 85, 103 (2007) . Congress has shown just that in another statute, 18 U. S. C. §1028A. That section, which criminalizes the commission of identity theft “during and in relation to” certain predicate felonies, imposes a mandatory minimum sentence “in addition to the punishment provided for” the underlying offense. §1028A(a)(1). It also says that the mandatory minimum must be consecutive to the sentence for the underlying offense. §1028A(b)(2). So far, §1028A tracks §924(c) in relevant respects. But §1028A goes further: It provides that in determining the appropriate length of imprisonment for the predicate felony “a court shall not in any way reduce the term to be imposed for such crime so as to compensate for, or otherwise take into account, any separate term of imprisonment imposed or to be imposed for a violation of this section.” §1028A(b)(3). Section 1028A says just what the Government reads §924(c) to say—of course, without actually saying it. The Government responds that §1028A was passed in 2004, long after Congress enacted the 1984 amendments creating the current sentencing regime in §924(c). Brief for United States 46. True. But §1028A confirms that it would have been easy enough to make explicit what the Government argues is implicit in §924(c). It also underscores that for over a decade Congress has been aware of a clear way to bar consideration of a mandatory minimum, but never during that time changed the language of §924(c) to mirror that of §1028A, even as it has amended other aspects of §924(c). * * * The Government speaks of Congress’s intent to prevent district courts from bottoming out sentences for predicate §924(c) offenses whenever they think a mandatory minimum under §924(c) is already punishment enough. But no such intent finds expression in the language of §924(c). That language simply requires any mandatory minimum under §924(c) to be imposed “in addition to” the sentence for the predicate offense, and to run consecutively to that sentence. Nothing in those requirements prevents a sentencing court from considering a mandatory minimum under §924(c) when calculating an appropriate sentence for the predicate offense. The judgment of the United States Court of Appeals for the Eighth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

580.US.2016_15-827

The Individuals with Disabilities Education Act (IDEA) offers States federal funds to assist in educating children with disabilities. The Act conditions that funding on compliance with certain statutory requirements, including the requirement that States provide every eligible child a “free appropriate public education,” or FAPE, by means of a uniquely tailored “individualized education program,” or IEP. 20 U. S. C. §§1401(9)(D), 1412(a)(1). This Court first addressed the FAPE requirement in Board of Ed. of Hendrick Hudson Central School Dist., Westchester Cty. v. Rowley, 458 U. S. 176 . The Court held that the Act guarantees a substantively adequate program of education to all eligible children, and that this requirement is satisfied if the child’s IEP sets out an educational program that is “reasonably calculated to enable the child to receive educational benefits.” Id., at 207. For children fully integrated in the regular classroom, this would typically require an IEP “reasonably calculated to enable the child to achieve passing marks and advance from grade to grade.” Id., at 204. Because the IEP challenged in Rowley plainly met this standard, the Court declined “to establish any one test for determining the adequacy of educational benefits conferred upon all children covered by the Act,” instead “confin[ing] its analysis” to the facts of the case before it. Id., at 202. Petitioner Endrew F., a child with autism, received annual IEPs in respondent Douglas County School District from preschool through fourth grade. By fourth grade, Endrew’s parents believed his academic and functional progress had stalled. When the school district proposed a fifth grade IEP that resembled those from past years, Endrew’s parents removed him from public school and enrolled him in a specialized private school, where he made significant progress. School district representatives later presented Endrew’s parents with a new fifth grade IEP, but they considered it no more adequate than the original plan. They then sought reimbursement for Endrew’s private school tuition by filing a complaint under the IDEA with the Colorado Department of Education. Their claim was denied, and a Federal District Court affirmed that determination. The Tenth Circuit also affirmed. That court interpreted Rowley to establish a rule that a child’s IEP is adequate as long as it is calculated to confer an “ educational benefit [that is] merely . . . more than de minimis,” 798 F. 3d 1329, 1338 (internal quotation marks omitted), and concluded that Endrew’s IEP had been “ reasonably calculated to enable [him] to make some progress, ” id., at 1342 (internal quotation marks omitted). The court accordingly held that Endrew had received a FAPE. Held: To meet its substantive obligation under the IDEA, a school must offer an IEP reasonably calculated to enable a child to make progress appropriate in light of the child’s circumstances. Pp. 9–16. (a) Rowley and the language of the IDEA point to the approach adopted here. The “reasonably calculated” qualification reflects a recognition that crafting an appropriate program of education requires a prospective judgment by school officials, informed by their own expertise and the views of a child’s parents or guardians; any review of an IEP must appreciate that the question is whether the IEP is reasonable, not whether the court regards it as ideal. An IEP must aim to enable the child to make progress; the essential function of an IEP is to set out a plan for pursuing academic and functional advancement. And the degree of progress contemplated by the IEP must be appropriate in light of the child’s circumstances, which should come as no surprise. This reflects the focus on the particular child that is at the core of the IDEA, and the directive that States offer instruction “specially designed” to meet a child’s “unique needs” through an “[i]ndividualized education program.” §§1401(29), (14) (emphasis added). Rowley sheds light on what appropriate progress will look like in many cases: For a child fully integrated in the regular classroom, an IEP typically should be “reasonably calculated to enable the child to achieve passing marks and advance from grade to grade.” 458 U. S., at 204. This guidance is grounded in the statutory definition of a FAPE. One component of a FAPE is “special education,” defined as “specially designed instruction . . . to meet the unique needs of a child with a disability.” §§1401(9), (29). In determining what it means to “meet the unique needs” of a child with a disability, the provisions of the IDEA governing the IEP development process provide guidance. These provisions reflect what the Court said in Rowley by focusing on “progress in the general education curriculum.” §§1414(d)(1)(A)(i) (I)(aa), (II)(aa), (IV)(bb). Rowley did not provide concrete guidance with respect to a child who is not fully integrated in the regular classroom and not able to achieve on grade level. A child’s IEP need not aim for grade-level advancement if that is not a reasonable prospect. But that child’s educational program must be appropriately ambitious in light of his circumstances, just as advancement from grade to grade is appropriately ambitious for most children in the regular classroom. The goals may differ, but every child should have the chance to meet challenging objectives. This standard is more demanding than the “merely more than de minimis” test applied by the Tenth Circuit. It cannot be right that the IDEA generally contemplates grade-level advancement for children with disabilities who are fully integrated in the regular classroom, but is satisfied with barely more than de minimis progress for children who are not. Pp. 9–15. (b) Endrew’s parents argue that the Act goes even further and requires States to provide children with disabilities educational opportunities that are “substantially equal to the opportunities afforded children without disabilities.” Brief for Petitioner 40. But the lower courts in Rowley adopted a strikingly similar standard, and this Court rejected it in clear terms. Mindful that Congress has not materially changed the statutory definition of a FAPE since Rowley was decided, this Court declines to interpret the FAPE provision in a manner so plainly at odds with the Court’s analysis in that case. P. 15. (c) The adequacy of a given IEP turns on the unique circumstances of the child for whom it was created. This absence of a bright-line rule should not be mistaken for “an invitation to the courts to substitute their own notions of sound educational policy for those of the school authorities which they review.” Rowley, 458 U. S., at 206. At the same time, deference is based on the application of expertise and the exercise of judgment by school authorities. The nature of the IEP process ensures that parents and school representatives will fully air their respective opinions on the degree of progress a child’s IEP should pursue; thus, by the time any dispute reaches court, school authorities will have had the chance to bring their expertise and judgment to bear on areas of disagreement. See §§1414, 1415; Rowley, 458 U. S., at 208–209. At that point, a reviewing court may fairly expect those authorities to be able to offer a cogent and responsive explanation for their decisions that shows the IEP is reasonably calculated to enable the child to make progress appropriate in light of his circumstances. Pp. 15–16. 798 F. 3d 1329, vacated and remanded. Roberts, C. J., delivered the opinion for a unanimous Court.

Thirty-five years ago, this Court held that the Individuals with Disabilities Education Act establishes a substantive right to a “free appropriate public education” for certain children with disabilities. Board of Ed. of Hendrick Hudson Central School Dist., Westchester Cty. v. Rowley, 458 U. S. 176 (1982) . We declined, however, to endorse any one standard for determining “when handicapped children are receiving sufficient educational benefits to satisfy the requirements of the Act.” Id., at 202. That “more difficult problem” is before us today. Ibid. I A The Individuals with Disabilities Education Act (IDEA or Act) offers States federal funds to assist in educating children with disabilities. 84Stat. 175, as amended, 20 U. S. C. §1400 et seq.; see Arlington Central School Dist. Bd. of Ed. v. Murphy, 548 U. S. 291, 295 (2006) . In exchange for the funds, a State pledges to comply with a number of statutory conditions. Among them, the State must provide a free appropriate public education—a FAPE, for short—to all eligible children. §1412(a)(1). A FAPE, as the Act defines it, includes both “special education” and “related services.” §1401(9). “Special education” is “specially designed instruction . . . to meet the unique needs of a child with a disability”; “related services” are the support services “required to assist a child . . . to benefit from” that instruction. §§1401(26), (29). A State covered by the IDEA must provide a disabled child with such special education and related services “in conformity with the [child’s] individualized education program,” or IEP. §1401(9)(D). The IEP is “the centerpiece of the statute’s education delivery system for disabled children.” Honig v. Doe, 484 U. S. 305, 311 (1988) . A comprehensive plan prepared by a child’s “IEP Team” (which includes teachers, school officials, and the child’s parents), an IEP must be drafted in compliance with a detailed set of procedures. §1414(d)(1)(B) (internal quotation marks omitted). These procedures emphasize collaboration among parents and educators and require careful consideration of the child’s individual circumstances. §1414. The IEP is the means by which special education and related services are “tailored to the unique needs” of a particular child. Rowley, 458 U. S., at 181. The IDEA requires that every IEP include “a statement of the child’s present levels of academic achievement and functional performance,” describe “how the child’s disability affects the child’s involvement and progress in the general education curriculum,” and set out “measurable annual goals, including academic and functional goals,” along with a “description of how the child’s progresstoward meeting” those goals will be gauged. §§1414(d)(1)(A)(i)(I)–(III). The IEP must also describe the “special education and related services . . . that will be provided” so that the child may “advance appropriately toward attaining the annual goals” and, when possible, “be involved in and make progress in the general education curriculum.” §1414(d)(1)(A)(i)(IV). Parents and educators often agree about what a child’s IEP should contain. But not always. When disagreement arises, parents may turn to dispute resolution procedures established by the IDEA. The parties may resolve their differences informally, through a “[p]reliminary meeting,” or, somewhat more formally, through mediation. §§1415(e), (f )(1)(B)(i). If these measures fail to produce accord, the parties may proceed to what the Act calls a “due process hearing” before a state or local educational agency. §§1415(f)(1)(A), (g). And at the conclusion of the administrative process, the losing party may seek redress in state or federal court. §1415(i)(2)(A). B This Court first addressed the FAPE requirement in Rowley.[1] Plaintiff Amy Rowley was a first grader with impaired hearing. Her school district offered an IEP under which Amy would receive instruction in the regular classroom and spend time each week with a special tutor and a speech therapist. The district proposed that Amy’s classroom teacher speak into a wireless transmitter and that Amy use an FM hearing aid designed to amplify her teacher’s words; the district offered to supply both components of this system. But Amy’s parents argued that the IEP should go further and provide a sign-language interpreter in all of her classes. Contending that the school district’s refusal to furnish an interpreter denied Amy a FAPE, Amy’s parents initiated administrative proceedings, then filed a lawsuit under the Act. Rowley, 458 U. S., at 184–185. The District Court agreed that Amy had been denied a FAPE. The court acknowledged that Amy was making excellent progress in school: She was “perform[ing] better than the average child in her class” and “advancing easily from grade to grade.” Id., at 185 (internal quotation marks omitted). At the same time, Amy “under[stood] considerably less of what goes on in class than she could if she were not deaf.” Ibid. (internal quotation marks omitted). Concluding that “it has been left entirely to the courts and the hearings officers to give content to the requirement of an ‘appropriate education,’ ” 483 F. Supp. 528, 533 (SDNY 1980), the District Court ruled that Amy’s education was not “appropriate” unless it provided her “an opportunity to achieve [her] full potential commensurate with the opportunity provided to other children.” Rowley, 458 U. S., at 185–186 (internal quotation marks omitted). The Second Circuit agreed with this analysis andaffirmed. In this Court, the parties advanced starkly different understandings of the FAPE requirement. Amy’s parents defended the approach of the lower courts, arguing that the school district was required to provide instruction and services that would provide Amy an “equal educational opportunity” relative to children without disabilities. Id., at 198 (internal quotation marks omitted). The school district, for its part, contended that the IDEA “did not create substantive individual rights”; the FAPE provision was instead merely aspirational. Brief for Petitioners in Rowley, O. T. 1981, No. 80–1002, pp. 28, 41. Neither position carried the day. On the one hand, this Court rejected the view that the IDEA gives “courts carte blanche to impose upon the States whatever burden their various judgments indicate should be imposed.” Rowley, 458 U. S., at 190, n. 11. After all, the statutory phrase “free appropriate public education” was expressly defined in the Act, even if the definition “tend[ed] toward the cryptic rather than the comprehensive.” Id., at 188. This Court went on to reject the “equal opportunity” standard adopted by the lower courts, concluding that “free appropriate public education” was a phrase “too complex to be captured by the word ‘equal’ whether one is speaking of opportunities or services.” Id., at 199. The Court also viewed the standard as “entirely unworkable,” apt to require “impossible measurements and comparisons” that courts were ill suited to make. Id., at 198. On the other hand, the Court also rejected the school district’s argument that the FAPE requirement was actually no requirement at all. Id., at 200. Instead, the Court carefully charted a middle path. Even though “Congress was rather sketchy in establishing substantive requirements” under the Act, id., at 206, the Court nonetheless made clear that the Act guarantees a substantively adequate program of education to all eligible children, id., at 200–202, 207; see id., at 193, n. 15 (describing the “substantive standard . . . implicit in the Act”). We explained that this requirement is satisfied, and a child has received a FAPE, if the child’s IEP sets out an educational program that is “reasonably calculated to enable the child to receive educational benefits.” Id., at 207. For children receiving instruction in the regular classroom, this would generally require an IEP “reasonably calculated to enable the child to achieve passing marks and advance from grade to grade.” Id., at 204; see also id., at 203, n. 25. In view of Amy Rowley’s excellent progress and the “substantial” suite of specialized instruction and services offered in her IEP, we concluded that her program satisfied the FAPE requirement. Id., at 202. But we went no further. Instead, we expressly “confine[d] our analysis” to the facts of the case before us. Ibid. Observing that the Act requires States to “educate a wide spectrum” of children with disabilities and that “the benefits obtainable by children at one end of the spectrum will differ dramatically from those obtainable by children at the other end,” we declined “to establish any one test for determining the adequacy of educational benefits conferred upon all children covered by the Act.” Ibid. C Petitioner Endrew F. was diagnosed with autism at age two. Autism is a neurodevelopmental disorder generally marked by impaired social and communicative skills, “engagement in repetitive activities and stereotyped movements, resistance to environmental change or change in daily routines, and unusual responses to sensory experiences.” 34 CFR §300.8(c)(1)(i) (2016); see Brief for Petitioner 8. A child with autism qualifies as a “[c]hild with a disability” under the IDEA, and Colorado (where Endrew resides) accepts IDEA funding. §1401(3)(A). Endrew is therefore entitled to the benefits of the Act, including a FAPE provided by the State. Endrew attended school in respondent Douglas County School District from preschool through fourth grade. Each year, his IEP Team drafted an IEP addressed to his educational and functional needs. By Endrew’s fourth grade year, however, his parents had become dissatisfied with his progress. Although Endrew displayed a number of strengths—his teachers described him as a humorous child with a “sweet disposition” who “show[ed] concern[ ] for friends”—he still “exhibited multiple behaviors that inhibited his ability to access learning in the classroom.” Supp. App. 182a; 798 F. 3d 1329, 1336 (CA10 2015). Endrew would scream in class, climb over furniture and other students, and occasionally run away from school. Id., at 1336. He was afflicted by severe fears of commonplace things like flies, spills, and public restrooms. As Endrew’s parents saw it, his academic and functional progress had essentially stalled: Endrew’s IEPs largely carried over the same basic goals and objectives from one year to the next, indicating that he was failing to make meaningful progress toward his aims. His parents believed that only a thorough overhaul of the school district’s approach to Endrew’s behavioral problems could reverse the trend. But in April 2010, the school district presented Endrew’s parents with a proposed fifth grade IEP that was, in their view, pretty much the same as his past ones. So his parents removed Endrew from public school and enrolled him at Firefly Autism House, a private school that specializes in educating children with autism. Endrew did much better at Firefly. The school developed a “behavioral intervention plan” that identified Endrew’s most problematic behaviors and set out particular strategies for addressing them. See Supp. App. 198a–201a. Firefly also added heft to Endrew’s academic goals. Within months, Endrew’s behavior improved significantly, permitting him to make a degree of academic progress that had eluded him in public school. In November 2010, some six months after Endrew started classes at Firefly, his parents again met with representatives of the Douglas County School District. The district presented a new IEP. Endrew’s parents considered the IEP no more adequate than the one proposed in April, and rejected it. They were particularly concerned that the stated plan for addressing Endrew’s behavior did not differ meaningfully from the plan in his fourth grade IEP, despite the fact that his experience at Firefly suggested that he would benefit from a different approach. In February 2012, Endrew’s parents filed a complaint with the Colorado Department of Education seeking reimbursement for Endrew’s tuition at Firefly. To qualify for such relief, they were required to show that the school district had not provided Endrew a FAPE in a timely manner prior to his enrollment at the private school. See §1412(a)(10)(C)(ii). Endrew’s parents contended that the final IEP proposed by the school district was not “reason-ably calculated to enable [Endrew] to receive educational benefits” and that Endrew had therefore been denied a FAPE. Rowley, 458 U. S., at 207. An Administrative Law Judge (ALJ) disagreed and denied relief. Endrew’s parents sought review in Federal District Court. Giving “due weight” to the decision of the ALJ, the District Court affirmed. 2014 WL 4548439, *5 (D Colo., Sept. 15, 2014) (quoting Rowley, 458 U. S., at 206). The court acknowledged that Endrew’s performance under past IEPs “did not reveal immense educational growth.” 2014 WL 4548439, at *9. But it concluded that annual modifications to Endrew’s IEP objectives were “sufficient to show a pattern of, at the least, minimal progress.” Ibid. Because Endrew’s previous IEPs had enabled him to make this sort of progress, the court reasoned, his latest, similar IEP was reasonably calculated to do the same thing. In the court’s view, that was all Rowley demanded. 2014 WL 4548439, at *9. The Tenth Circuit affirmed. The Court of Appeals recited language from Rowley stating that the instruction and services furnished to children with disabilities must be calculated to confer “some educational benefit.” 798 F. 3d, at 1338 (quoting Rowley, 458 U. S., at 200; emphasis added by Tenth Circuit). The court noted that it had long interpreted this language to mean that a child’s IEP is adequate as long as it is calculated to confer an “educational benefit [that is] merely . . . more than de minimis.” 798 F. 3d, at 1338 (internal quotation marks omitted). Applying this standard, the Tenth Circuit held that Endrew’s IEP had been “reasonably calculated to enable [him] to make some progress.” Id., at 1342 (internal quotation marks omitted). Accordingly, he had not been denied a FAPE. We granted certiorari. 579 U. S. ___ (2016). II A The Court in Rowley declined “to establish any one test for determining the adequacy of educational benefits conferred upon all children covered by the Act.” 458 U. S., at 202. The school district, however, contends that Rowley nonetheless established that “an IEP need not promise any particular level of benefit,” so long as it is “ ‘ reasonably calculated’ to provide some benefit, as opposed to none.” Brief for Respondent 15. The district relies on several passages from Rowley to make its case. It points to our observation that “any substantive standard prescribing the level of education to be accorded” children with disabilities was “[n]oticeably absent from the language of the statute.” 458 U. S., at 189; see Brief for Respondent 14. The district also emphasizes the Court’s statement that the Act requires States to provide access to instruction “sufficient to confer some educational benefit,” reasoning that any benefit, however minimal, satisfies this mandate. Brief for Respondent 15 (quoting Rowley, 458 U. S., at 200). Finally, the district urges that the Court conclusively adopted a “some educational benefit” standard when it wrote that “the intent of the Act was more to open the door of public education to handicapped children . . . than to guarantee any particular level of education.” Id., at 192; see Brief for Respond-ent 14. These statements in isolation do support the school district’s argument. But the district makes too much of them. Our statement that the face of the IDEA imposed no explicit substantive standard must be evaluated alongside our statement that a substantive standard was “implicit in the Act.” Rowley, 458 U. S., at 193, n. 15. Similarly, we find little significance in the Court’s language concerning the requirement that States provide instruction calculated to “confer some educational benefit.” Id., at 200. The Court had no need to say anything more particular, since the case before it involved a child whose progress plainly demonstrated that her IEP was designed to deliver more than adequate educational benefits. See id., at 202, 209–210. The Court’s principal concern was to correct what it viewed as the surprising rulings below: that the IDEA effectively empowers judges to elaborate a federal common law of public education, and that a child performing better than most in her class had been denied a FAPE. The Court was not concerned with precisely articulating a governing standard for closer cases. See id., at 202. And the statement that the Act did not “guarantee any particular level of education” simply reflects the unobjectionable proposition that the IDEA cannot and does not promise “any particular [educational] outcome.” Id., at 192 (internal quotation marks omitted). No law could do that—for any child. More important, the school district’s reading of these isolated statements runs headlong into several points on which Rowley is crystal clear. For instance—just after saying that the Act requires instruction that is “sufficient to confer some educational benefit”—we noted that “[t]he determination of when handicapped children are receiving sufficient educational benefits . . . presents a . . . difficult problem.” Id., at 200, 202 (emphasis added). And then we expressly declined “to establish any one test for determining the adequacy of educational benefits” under the Act. Id., at 202 (emphasis added). It would not have been “difficult” for us to say when educational benefits are sufficient if we had just said that any educational benefit was enough. And it would have been strange to refuse to set out a test for the adequacy of educational benefits if we had just done exactly that. We cannot accept the school district’s reading of Rowley. B While Rowley declined to articulate an overarching standard to evaluate the adequacy of the education provided under the Act, the decision and the statutory language point to a general approach: To meet its substantive obligation under the IDEA, a school must offer an IEP reasonably calculated to enable a child to make progress appropriate in light of the child’s circumstances. The “reasonably calculated” qualification reflects a recognition that crafting an appropriate program of education requires a prospective judgment by school officials. Id., at 207. The Act contemplates that this fact-intensive exercise will be informed not only by the expertise of school officials, but also by the input of the child’s parents or guardians. Id., at 208–209. Any review of an IEP must appreciate that the question is whether the IEP is reasonable, not whether the court regards it as ideal. Id., at 206–207. The IEP must aim to enable the child to make progress. After all, the essential function of an IEP is to set out a plan for pursuing academic and functional advancement. See §§1414(d)(1)(A)(i)(I)–(IV). This reflects the broad purpose of the IDEA, an “ambitious” piece of legislation enacted “in response to Congress’ perception that a majority of handicapped children in the United States ‘were either totally excluded from schools or [were] sitting idly in regular classrooms awaiting the time when they were old enough to “drop out.” ’ ” Rowley, 458 U. S., at 179 (quoting H. R. Rep. No. 94–332, p. 2 (1975)). A substantive standard not focused on student progress would do little to remedy the pervasive and tragic academic stagnation that prompted Congress to act. That the progress contemplated by the IEP must be appropriate in light of the child’s circumstances should come as no surprise. A focus on the particular child is at the core of the IDEA. The instruction offered must be “specially designed” to meet a child’s “unique needs” through an “[i]ndividualized education program.” §§1401(29), (14) (emphasis added). An IEP is not a form document. It is constructed only after careful consideration of the child’s present levels of achievement, disability, and potential for growth. §§1414(d)(1)(A)(i)(I)–(IV), (d)(3)(A)(i)–(iv). As we observed in Rowley, the IDEA “requires participating States to educate a wide spectrum of handicapped children,” and “the benefits obtainable by children at one end of the spectrum will differ dramatically from those obtainable by children at the other end, with infinite variations in between.” 458 U. S., at 202. Rowley sheds light on what appropriate progress will look like in many cases. There, the Court recognized that the IDEA requires that children with disabilities receive education in the regular classroom “whenever possible.” Ibid. (citing §1412(a)(5)). When this preference is met, “the system itself monitors the educational progress of the child.” Id., at 202–203. “Regular examinations are administered, grades are awarded, and yearly advancement to higher grade levels is permitted for those children who attain an adequate knowledge of the course material.” Id., at 203. Progress through this system is what our society generally means by an “education.” And access to an “education” is what the IDEA promises. Ibid. Accordingly, for a child fully integrated in the regular classroom, an IEP typically should, as Rowley put it, be “reasonably calculated to enable the child to achieve passing marks and advance from grade to grade.” Id., at 203–204. This guidance is grounded in the statutory definition of a FAPE. One of the components of a FAPE is “special education,” defined as “specially designed instruction . . . to meet the unique needs of a child with a disability.” §§1401(9), (29). In determining what it means to “meet the unique needs” of a child with a disability, the provisions governing the IEP development process are a natural source of guidance: It is through the IEP that “[t]he ‘free appropriate public education’ required by the Act is tailored to the unique needs of” a particular child. Id.,at 181. The IEP provisions reflect Rowley’s expectation that, for most children, a FAPE will involve integration in the reg-ular classroom and individualized special education calculated to achieve advancement from grade to grade. Every IEP begins by describing a child’s present level of achieve-ment, including explaining “how the child’s disability affects the child’s involvement and progress in the general education curriculum.” §1414(d)(1)(A)(i)(I)(aa). It then sets out “a statement of measurable annual goals . . . designed to . . . enable the child to be involved in and make progress in the general education curriculum,” along with a description of specialized instruction and services that the child will receive. §§1414(d)(1)(A)(i)(II), (IV). The instruction and services must likewise be provided with an eye toward “progress in the general education curriculum.” §1414(d)(1)(A)(i)(IV)(bb). Similar IEP requirements have been in place since the time the States began accepting funding under the IDEA. The school district protests that these provisions impose only procedural requirements—a checklist of items the IEP must address—not a substantive standard enforce-able in court. Tr. of Oral Arg. 50–51. But the procedures are there for a reason, and their focus provides insight into what it means, for purposes of the FAPE definition, to “meet the unique needs” of a child with a disability. §§1401(9), (29). When a child is fully integrated in the regular classroom, as the Act prefers, what that typically means is providing a level of instruction reasonablycalculated to permit advancement through the general curriculum.[2] Rowley had no need to provide concrete guidance with respect to a child who is not fully integrated in the regular classroom and not able to achieve on grade level. That case concerned a young girl who was progressing smoothly through the regular curriculum. If that is not a reason-able prospect for a child, his IEP need not aim for grade-level advancement. But his educational program must be appropriately ambitious in light of his circumstances, just as advancement from grade to grade is appropriately ambitious for most children in the regular classroom. The goals may differ, but every child should have the chance to meet challenging objectives. Of course this describes a general standard, not a formula. But whatever else can be said about it, this standard is markedly more demanding than the “merely more than de minimis” test applied by the Tenth Circuit. It cannot be the case that the Act typically aims for grade-level advancement for children with disabilities who can be educated in the regular classroom, but is satisfiedwith barely more than de minimis progress for those who cannot. When all is said and done, a student offered an educational program providing “merely more than de minimis” progress from year to year can hardly be said to have been offered an education at all. For children with disabilities, receiving instruction that aims so low would be tantamount to “sitting idly . . . awaiting the time when they were old enough to ‘drop out.’ ” Rowley, 458 U. S., at 179 (some internal quotation marks omitted). The IDEA demands more. It requires an educational program reasonably calculated to enable a child to make progress appropriate in light of the child’s circumstances. C Endrew’s parents argue that the Act goes even further. In their view, a FAPE is “an education that aims to provide a child with a disability opportunities to achieve academic success, attain self-sufficiency, and contribute to society that are substantially equal to the opportunities afforded children without disabilities.” Brief for Petitioner 40. This standard is strikingly similar to the one the lower courts adopted in Rowley, and it is virtually identical to the formulation advanced by Justice Blackmun in his separate writing in that case. See 458 U. S., at 185–186; id., at 211 (opinion concurring in judgment) (“[T]he question is whether Amy’s program . . . offered her an opportunity to understand and participate in the classroom that was substantially equal to that given her non-handicapped classmates”). But the majority rejected any such standard in clear terms. Id., at 198 (“The requirement that States provide ‘equal’ educational opportunities would . . . seem to present an entirely unworkable standard requiring impossible measurements and comparisons”). Mindful that Congress (despite several intervening amendments to the IDEA) has not materially changed the statutory definition of a FAPE since Rowley was decided, we decline to interpret the FAPE provision in a manner so plainly at odds with the Court’s analysis in that case. Compare §1401(18) (1976 ed.) with §1401(9) (2012 ed.). D We will not attempt to elaborate on what “appropriate” progress will look like from case to case. It is in the nature of the Act and the standard we adopt to resist such an effort: The adequacy of a given IEP turns on the unique circumstances of the child for whom it was created. This absence of a bright-line rule, however, should not be mistaken for “an invitation to the courts to substitute their own notions of sound educational policy for those of the school authorities which they review.” Rowley, 458 U. S., at 206. At the same time, deference is based on the application of expertise and the exercise of judgment by school authorities. The Act vests these officials with responsibility for decisions of critical importance to the life of a disabled child. The nature of the IEP process, from the initial consultation through state administrative proceedings, ensures that parents and school representatives will fully air their respective opinions on the degree of progress a child’s IEP should pursue. See §§1414, 1415; id., at 208–209. By the time any dispute reaches court, school authorities will have had a complete opportunity to bring their expertise and judgment to bear on areas of disagreement. A reviewing court may fairly expect those authorities to be able to offer a cogent and responsive explanation for their decisions that shows the IEP is reasonably calculated to enable the child to make progress appropriate in light of his circumstances. The judgment of the United States Court of Appeals for the Tenth Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The requirement was initially set out in the Education of the Handicapped Act, which was later amended and renamed the IDEA. See Pub. L. 101–476, §901(a), 104Stat. 1141. For simplicity’s sake—and to avoid “acronym overload”—we use the latter title throughout this opinion. Fry v. Napoleon Community Schools, 580 U. S. ___, ___, n. 1 (2017) (slip op., at 4, n. 1). 2 This guidance should not be interpreted as an inflexible rule. We declined to hold in Rowley, and do not hold today, that “every handicapped child who is advancing from grade to grade . . . is automatically receiving a [FAPE].” Board of Ed. of Hendrick Hudson Central School Dist., Westchester Cty. v. Rowley, 458 U. S. 176, 203, n. 25 (1982) .

581.US.2016_16-54

Petitioner, a citizen of Mexico and lawful permanent resident of the United States, pleaded no contest in a California court to a statutory rape offense criminalizing “unlawful sexual intercourse with a minor who is more than three years younger than the perpetrator.” Cal. Penal Code Ann. §261.5(c). For purposes of that offense, California defines “minor” as “a person under the age of 18.” §261.5(a). Based on this conviction, the Department of Homeland Security initiated removal proceedings under the Immigration and Nationality Act (INA), which makes removable “[a]ny alien who is convicted of an aggravated felony,” 8 U. S. C. §1227(a)(2)(A)(iii), including “sexual abuse of a minor,” §1101(a)(43)(A). An Immigration Judge ordered petitioner removed to Mexico. The Board of Immigration Appeals agreed that petitioner’s crime constituted sexual abuse of a minor and dismissed his appeal. A divided Court of Appeals denied his petition for review. Held: In the context of statutory rape offenses that criminalize sexual intercourse based solely on the ages of the participants, the generic federal definition of “sexual abuse of a minor” requires the age of the victim to be less than 16. Pp. 2–12. (a) Under the categorical approach employed to determine whether an alien’s conviction qualifies as an aggravated felony, the Court asks whether “ ‘the state statute defining the crime of conviction’ categorically fits within the ‘generic’ federal definition of a corresponding aggravated felony.” Moncrieffe v. Holder, 569 U. S. 184 . Petitioner’s state conviction is thus an “aggravated felony” only if the least of the acts criminalized by the state statute falls within the generic federal definition of sexual abuse of a minor. Johnson v. United States, 559 U. S. 133 . Pp. 2–3. (b) The least of the acts criminalized by Cal. Penal Code §261.5(c) would be consensual sexual intercourse between a victim who is almost 18 and a perpetrator who just turned 21. Regardless of the actual facts of the case, this Court presumes that petitioner’s conviction was based on those acts. Pp. 3–4. (c) In the context of statutory rape offenses that criminalize sexual intercourse based solely on the ages of the participants, the generic federal definition of “sexual abuse of a minor” requires that the victim be younger than 16. The Court begins, as always, with the text. Pp. 4–7. (1) Congress added sexual abuse of a minor to the INA in 1996. At that time, the ordinary meaning of “sexual abuse” included “the engaging in sexual contact with a person who is below a specified age or who is incapable of giving consent because of age or mental or physical incapacity.” Merriam-Webster’s Dictionary of Law 454. By providing that the abuse must be “of a minor,” the INA focuses on age, rather than mental or physical incapacity. Accordingly, to qualify as sexual abuse of a minor, the statute of conviction must prohibit certain sexual acts based at least in part on the age of the victim. Statutory rape laws, which are one example of this category of crimes, generally provide that an older person may not engage in sexual intercourse with a younger person under the “age of consent.” Reliable dictionaries indicate that the “generic” age of consent in 1996 was 16, and it remains so today. Pp. 4–6. (2) The Government argues that sexual abuse of a minor includes any conduct that is illegal, involves sexual activity, and is directed at a person younger than 18. For support, it points to the 1990 Black’s Law Dictionary, which defined sexual abuse of a minor as “[i]llegal sex acts performed against a minor by a parent, guardian, relative, or acquaintance” and defined “[m]inor” as “[a]n infant or person who is under the age of legal competence,” which in “most states” was “18.” But the generic federal offense does not correspond to the Government’s definition, for three reasons. First, the Government’s definition is inconsistent with its own dictionary’s requirement that a special relationship of trust exist between the victim and offender. Second, in the statutory rape context, “of a minor” refers to the age of consent, not the age of legal competence. Third, the Government’s definition turns the categorical approach on its head by defining the generic federal offense as whatever is illegal under the law of the State of conviction. Pp. 6–7. (d) The structure of the INA, a related federal statute, and evidence from state criminal codes confirm that, for a statutory rape offense based solely on the age of the participants to qualify as sexual abuse of a minor under the INA, the victim must be younger than 16. The INA lists sexual abuse of a minor as an “aggravated” felony, §1227(a)(2)(A)(iii), and lists it in the same subparagraph as “murder” and “rape,” §1101(a)(43)(A), suggesting that it encompasses only especially egregious felonies. A different statute, 18 U. S. C. §2243, criminalizes “[s]exual abuse of a minor or ward.” Section 2243 was amended to protect anyone under age 16 in the same omnibus law that added sexual abuse of a minor to the INA, suggesting that Congress understood that phrase to cover victims under (but not over) age 16. Finally, a significant majority of state criminal codes set the age of consent at 16 for statutory rape offenses predicated exclusively on the age of the participants. Pp. 7–11. (e) This Court does not decide whether the generic crime of sexual abuse of a minor requires a particular age differential between the victim and the perpetrator or whether it encompasses sexual intercourse involving victims over 16 that is abusive because of the nature of the relationship between the participants. P. 11. (f) Because the statute, read in context, unambiguously forecloses the Board’s interpretation of sexual abuse of a minor, neither the rule of lenity nor Chevron deference applies. Pp. 11–12. 810 F. 3d 1019, reversed. Thomas, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case.

The Immigration and Nationality Act (INA), 66Stat. 163, as amended, provides that “[a]ny alien who is convicted of an aggravated felony after admission” to the United States may be removed from the country by the Attorney General. 8 U. S. C. §1227(a)(2)(A)(iii). One of the many crimes that constitutes an aggravated felony under the INA is “sexual abuse of a minor.” §1101(a)(43)(A). A conviction for sexual abuse of a minor is an aggravated felony regardless of whether it is for a “violation of Federal or State law.” §1101(a)(43). The INA does not expressly define sexual abuse of a minor. We must decide whether a conviction under a state statute criminalizing consensual sexual intercourse between a 21-year-old and a 17-year-old qualifies as sexual abuse of a minor under the INA. We hold that it does not. I Petitioner Juan Esquivel-Quintana is a native and citizen of Mexico. He was admitted to the United States as a lawful permanent resident in 2000. In 2009, he pleaded no contest in the Superior Court of California to a statutory rape offense: “unlawful sexual intercourse with a minor who is more than three years younger than the perpetrator,” Cal. Penal Code Ann. §261.5(c) (West 2014); see also §261.5(a) (“Unlawful sexual intercourse is an act of sexual intercourse accomplished with a person who is not the spouse of the perpetrator, if the person is a minor”). For purposes of that offense, California defines “minor” as “a person under the age of 18 years.” Ibid. The Department of Homeland Security initiated removal proceedings against petitioner based on that conviction. An Immigration Judge concluded that the conviction qualified as “sexual abuse of a minor,” 8 U. S. C. §1101(a)(43)(A), and ordered petitioner removed to Mexico. The Board of Immigration Appeals (Board) dismissed his appeal. 26 I. & N. Dec. 469 (2015). “[F]or a statutory rape offense involving a 16- or 17-year-old victim” to qualify as “ ‘sexual abuse of a minor,’ ” it reasoned, “the statute must require a meaningful age difference between the victim and the perpetrator.” Id., at 477. In its view, the 3-year age difference required by Cal. Penal Code §261.5(c) was meaningful. Id., at 477. Accordingly, the Board concluded that petitioner’s crime of conviction was an aggravated felony, making him removable under the INA. Ibid. A divided Court of Appeals denied Esquivel-Quintana’s petition for review, deferring to the Board’s interpretation of sexual abuse of a minor under Chevron U. S. A. Inc. v. Natural Resources Defense Council, Inc., 467 U. S. 837 (1984) . 810 F. 3d 1019 (CA6 2016); see also id., at 1027 (Sutton, J., concurring in part and dissenting in part). We granted certiorari, 580 U. S. ___ (2016), and now reverse. II Section 1227(a)(2)(A)(iii) makes aliens removable based on the nature of their convictions, not based on their actual conduct. See Mellouli v. Lynch, 575 U. S. ___, ___ (2015) (slip op., at 7). Accordingly, to determine whether an alien’s conviction qualifies as an aggravated felony under that section, we “employ a categorical approach by looking to the statute . . . of conviction, rather than to the specific facts underlying the crime.” Kawashima v. Holder, 565 U. S. 478, 483 (2012) ; see, e.g., Gonzales v. Duenas-Alvarez, 549 U. S. 183, 186 (2007) (applying the categorical approach set forth in Taylor v. United States, 495 U. S. 575 (1990) , to the INA). Under that approach, we ask whether “ ‘the state statute defining the crime of conviction’ categorically fits within the ‘generic’ federal definition of a corresponding aggravated felony.” Moncrieffe v. Holder, 569 U. S. 184, 190 (2013) (quoting Duenas-Alvarez, supra, at 186). In other words, we presume that the state conviction “rested upon . . . the least of th[e] acts” criminalized by the statute, and then we determine whether that conduct would fall within the federal definition of the crime. Johnson v. United States, 559 U. S. 133, 137 (2010) ; see also Moncrieffe, supra, at 191 (focusing “on the minimum conduct criminalized by the state statute”).[1] Petitioner’s state conviction is thus an “aggravated felony” under the INA only if the least of the acts criminalized by the state statute falls within the generic federal definition of sexual abuse of a minor. A Because Cal. Penal Code §261.5(c) criminalizes “unlawful sexual intercourse with a minor who is more than three years younger than the perpetrator” and defines a minor as someone under age 18, the conduct criminalized under this provision would be, at a minimum, consensual sexual intercourse between a victim who is almost 18 and a perpetrator who just turned 21. Regardless of the actual facts of petitioner’s crime, we must presume that his conviction was based on acts that were no more criminal than that. If those acts do not constitute sexual abuse of a minor under the INA, then petitioner was not convicted of an aggravated felony and is not, on that basis, removable. Petitioner concedes that sexual abuse of a minor under the INA includes some statutory rape offenses. But he argues that a statutory rape offense based solely on the partners’ ages (like the one here) is “ ‘abuse’ ” “only when the younger partner is under 16.” Reply Brief 2. Because the California statute criminalizes sexual intercourse when the victim is up to 17 years old, petitioner contends that it does not categorically qualify as sexual abuse of a minor. B We agree with petitioner that, in the context of statutory rape offenses that criminalize sexual intercourse based solely on the age of the participants, the generic federal definition of sexual abuse of a minor requires that the victim be younger than 16. Because the California statute at issue in this case does not categorically fall within that definition, a conviction pursuant to it is not an aggravated felony under §1101(a)(43)(A). We begin, as always, with the text. 1 Section 1101(a)(43)(A) does not expressly define sexual abuse of a minor, so we interpret that phrase using the normal tools of statutory interpretation. “Our analysis begins with the language of the statute.” Leocal v. Ashcroft, 543 U. S. 1, 8 (2004) ; see also Lopez v. Gonzales, 549 U. S. 47, 53 (2006) (“The everyday understanding of” the term used in §1101 “should count for a lot here, for the statutes in play do not define the term, and so remit us to regular usage to see what Congress probably meant”). Congress added sexual abuse of a minor to the INA in 1996, as part of a comprehensive immigration reform act. See Illegal Immigration Reform and Immigrant Responsibility Act of 1996, §321(a)(i), 110Stat. 3009–627. At that time, the ordinary meaning of “sexual abuse” included “the engaging in sexual contact with a person who is below a specified age or who is incapable of giving consent because of age or mental or physical incapacity.” Merriam-Webster’s Dictionary of Law 454 (1996). By providing that the abuse must be “of a minor,” the INA focuses on age, rather than mental or physical incapacity. Accordingly, to qualify as sexual abuse of a minor, the statute of conviction must prohibit certain sexual acts based at least in part on the age of the victim. Statutory rape laws are one example of this category of crimes. Those laws generally provide that an older person may not engage in sexual intercourse with a younger person under a specified age, known as the “age of consent.” See id., at 20 (defining “age of consent” as “the age at which a person is deemed competent by law to give consent esp. to sexual intercourse” and cross-referencing “statutory rape”). Many laws also require an age differential between the two partners. Although the age of consent for statutory rape purposes varies by jurisdiction, see infra, at 9, reliable dictionaries provide evidence that the “generic” age—in 1996 and today—is 16. See B. Garner, A Dictionary of Modern Legal Usage 38 (2d ed. 1995) (“Age of consent, usu[ally] 16, denotes the age when one is legally capable of agreeing . . . to sexual intercourse” and cross-referencing “statutory rape”); Black’s Law Dictionary 73 (10th ed. 2014) (noting that the age of consent is “usu[ally] defined by statute as 16 years”). 2 Relying on a different dictionary (and “sparse” legislative history), the Government suggests an alternative “ ‘everyday understanding’ ” of “sexual abuse of a minor.” Brief for Respondent 16–17 (citing Black’s Law Dictionary 1375 (6th ed. 1990)). Around the time sexual abuse of a minor was added to the INA’s list of aggravated felonies, that dictionary defined “[s]exual abuse” as “[i]llegal sex acts performed against a minor by a parent, guardian, relative, or acquaintance,” and defined “[m]inor” as “[a]n infant or person who is under the age of legal competence,” which in “most states” was “18.” Id., at 997, 1375. “ ‘Sexual abuse of a minor,’ ” the Government accordingly contends, “most naturally connotes conduct that (1) is illegal, (2) involves sexual activity, and (3) is directed at a person younger than 18 years old.” Brief for Respondent 17. We are not persuaded that the generic federal offense corresponds to the Government’s definition. First, the Government’s proposed definition is flatly inconsistent with the definition of sexual abuse contained in the very dictionary on which it relies; the Government’s proposed definition does not require that the act be performed “by a parent, guardian, relative, or acquaintance.” Black’s Law Dictionary 1375 (6th ed. 1990) (emphasis added). In any event, as we explain below, offenses predicated on a special relationship of trust between the victim and offender are not at issue here and frequently have a different age requirement than the general age of consent. Second, in the context of statutory rape, the prepositional phrase “of a minor” naturally refers not to the age of legal competence (when a person is legally capable of agreeing to a contract, for example), but to the age of consent (when a person is legally capable of agreeing to sexual intercourse). Third, the Government’s definition turns the categorical approach on its head by defining the generic federal offense of sexual abuse of a minor as whatever is illegal under the particular law of the State where the defendant was convicted. Under the Government’s preferred approach, there is no “generic” definition at all. See Taylor, 495 U. S., at 591 (requiring “a clear indication that . . . Congress intended to abandon its general approach of using uniform categorical definitions to identify predicate offenses”); id., at 592 (“We think that ‘burglary’ in §924(e) must have some uniform definition independent of the labels employed by the various States’ criminal codes”). C The structure of the INA, a related federal statute, and evidence from state criminal codes confirm that, for a statutory rape offense to qualify as sexual abuse of a minor under the INA based solely on the age of the participants, the victim must be younger than 16. 1 Surrounding provisions of the INA guide our interpretation of sexual abuse of a minor. See A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 167 (2012). This offense is listed in the INA as an “aggravated felony.” 8 U. S. C. §1227(a)(2)(A)(iii) (emphasis added). “An ‘aggravated’ offense is one ‘made worse or more serious by circumstances such as violence, the presence of a deadly weapon, or the intent to commit another crime.’ ” Carachuri-Rosendo v. Holder, 560 U. S. 563, 574 (2010) (quoting Black’s Law Dictionary 75 (9th ed. 2009)). Moreover, the INA lists sexual abuse of a minor in the samesubparagraph as “murder” and “rape,” §1101(a)(43)(A)—among the most heinous crimes it defines as aggravated felonies. §1227(a)(2)(A)(iii). The structure of the INA therefore suggests that sexual abuse of a minor encompasses only especially egregious felonies. A closely related federal statute, 18 U. S. C. §2243, provides further evidence that the generic federal definition of sexual abuse of a minor incorporates an age of consent of 16, at least in the context of statutory rape offenses predicated solely on the age of the participants. Cf. Leocal, 543 U. S., at 12–13, n. 9 (concluding that Congress’ treatment of 18 U. S. C. §16 in an Act passed “just nine months earlier” provided “stron[g] suppor[t]” for our interpretation of §16 as incorporated into the INA); Powerex Corp. v. Reliant Energy Services, Inc., 551 U. S. 224, 232 (2007) . Section 2243, which criminalizes “[s]exual abuse of a minor or ward,” contains the only definition of that phrase in the United States Code. As originally enacted in 1986, §2243 proscribed engaging in a “sexual act” with a person between the ages of 12 and 16 if the perpetrator was at least four years older than the victim. In 1996, Congress expanded §2243 to include victims who were younger than 12, thereby protecting anyone under the age of 16. §2243(a); see also §2241(c). Congress did this in the same omnibus law that added sexual abuse of a minor to the INA, which suggests that Congress understood that phrase to cover victims under age 16.[2] See Omnibus Consolidated Appropriations Act, 1997, §§121(7), 321, 110Stat. 3009–31, 3009–627. Petitioner does not contend that the definition in §2243(a) must be imported wholesale into the INA, Brief for Petitioner 17, and we do not do so. One reason is that the INA does not cross-reference §2243(a), whereas many other aggravated felonies in the INA are defined by cross-reference to other provisions of the United States Code, see, e.g., §1101(a)(43)(H) (“an offense described in section 875, 876, 877, or 1202 of Title 18 (relating to the demand for or receipt of ransom)”). Another is that §2243(a) requires a 4-year age difference between the perpetrator and the victim. Combining that element with a 16-year age of consent would categorically exclude the statutory rape laws of most States. See Brief for Respondent 34–35; cf. Taylor, 495 U. S., at 594 (declining to “constru[e] ‘burglary’ to mean common-law burglary,” because that “would come close to nullifying that term’s effect in the statute,” since “few of the crimes now generally recognized as burglaries would fall within the common-law definition”). Accordingly, we rely on §2243(a) for evidence of the meaning of sexual abuse of a minor, but not as providing the complete or exclusive definition. 2 As in other cases where we have applied the categorical approach, we look to state criminal codes for additional evidence about the generic meaning of sexual abuse of a minor. See Taylor, 495 U. S., at 598 (interpreting “ ‘bur-glary’ ” under the Armed Career Criminal Act of 1984 according to “the generic sense in which the term is now used in the criminal codes of most States”); Duenas-Alvarez, 549 U. S., at 190 (interpreting “theft” in the INA in the same manner). When “sexual abuse of a minor” was added to the INA in 1996, thirty-one States and the District of Columbia set the age of consent at 16 for statutory rape offenses that hinged solely on the age of the participants. As for the other States, one set the age of consent at 14; two set the age of consent at 15; six set the age of consent at 17; and the remaining ten, including California, set the age of consent at 18. See Appendix, infra; cf. ALI, Model Penal Code §213.3(1)(a) (1980) (in the absence of a special relationship, setting the default age of consent at 16 for the crime of “[c]orruption of [m]inors”).[3] A significant majority of jurisdictions thus set the age of consent at 16 for statutory rape offenses predicated exclusively on the age of the participants. Many jurisdictions set a different age of consent for offenses that include an element apart from the age of the participants, such as offenses that focus on whether the perpetrator is in some special relationship of trust with the victim. That was true in the two States that had offenses labeled “sexual abuse of a minor” in 1996. See Alaska Stat. §11.41.438 (1996) (age of consent for third-degree “sexual abuse of a minor” was 16 generally but 18 where “the offender occupie[d] a position of authority in relation to the victim”); Me. Rev. Stat. Ann., Tit. 17–A, §254(1) (1983), as amended by 1995 Me. Laws p. 123 (age of consent for “[s]exual abuse of minors” was 16 generally but 18 where the victim was “a student” and the offender was “a teacher, employee or other official in the . . . school . . . in which the student [was] enrolled”). And that is true in four of the five jurisdictions that have offenses titled “sexual abuse of a minor” today. Compare, e.g., D. C. Code §§22–3001 (2012), 22–3008 (2016 Cum. Supp.) (age of consent is 16 in the absence of a significant relationship) with §22–3009.01 (age of consent is 18 where the offender “is in a significant relationship” with the victim); see also Brief for Respondent 31 (listing statutes with that title). Accordingly, the generic crime of sexual abuse of a minor may include a different age of consent where the perpetrator and victim are in a significant relationship of trust. As relevant to this case, however, the general consensus from state criminal codes points to the same generic definition as dictionaries and federal law: Where sexual intercourse is abusive solely because of the ages of the participants, the victim must be younger than 16. D The laws of many States and of the Federal Government include a minimum age differential (in addition to an age of consent) in defining statutory rape. We need not and do not decide whether the generic crime of sexual abuse of a minor under 8 U. S. C. §1101(a)(43)(A) includes an additional element of that kind. Petitioner has “show[n] something special about California’s version of the doctrine”—that the age of consent is 18, rather than 16—and needs no more to prevail. Duenas-Alvarez, supra, at 191. Absent some special relationship of trust, consensual sexual conduct involving a younger partner who is at least 16 years of age does not qualify as sexual abuse of a minor under the INA, regardless of the age differential between the two participants. We leave for another day whether the generic offense requires a particular age differential between the victim and the perpetrator, and whether the generic offense encompasses sexual intercourse involving victims over the age of 16 that is abusive because of the nature of the relationship between the participants. III Finally, petitioner and the Government debate whether the Board’s interpretation of sexual abuse of a minor is entitled to deference under Chevron, 467 U. S. 837 . Petitioner argues that any ambiguity in the meaning of this phrase must be resolved in favor of the alien under the rule of lenity. See Brief for Petitioner 41–45. The Government responds that ambiguities should be resolved by deferring to the Board’s interpretation. See Brief for Respondent 45–53. We have no need to resolve whether the rule of lenity or Chevron receives priority in this case because the statute, read in context, unambiguously forecloses the Board’s interpretation. Therefore, neither the rule of lenity nor Chevron applies. * * * We hold that in the context of statutory rape offenses focused solely on the age of the participants, the generic federal definition of “sexual abuse of a minor” under §1101(a)(43)(A) requires the age of the victim to be less than 16. The judgment of the Court of Appeals, accordingly, is reversed. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case. APPENDIX These tables list offenses criminalizing sexual intercourse solely because of the age of the participants. The tables are organized according to the statutory age of consent as of September 30, 1996—the date “sexual abuse of a minor” was added to the INA. 14 Years 15 Years 16 Years 17 Years 18 Years Notes 1 Where a state statute contains several different crimes that are described separately, we employ what is known as the “modified categorical approach.” See Gonzales v. Duenas-Alvarez, 549 U. S. 183, 187 (2007) (internal quotation marks omitted). Under that approach, which is not at issue here, the court may review the charging documents, jury instructions, plea agreement, plea colloquy, and similar sources to determine the actual crime of which the alien was convicted. See ibid. 2 To eliminate a redundancy, Congress later amended §2243(a) to revert to the pre-1996 language. See Protection of Children From Sexual Predators Act of 1998, §301(b), 112Stat. 2979. That amendment does not change Congress’ understanding in 1996, when it added sexual abuse of a minor to the INA. 3 The Government notes that this sort of multijurisdictional analysis can “be useful insofar as it helps shed light on the ‘common understanding and meaning’ of the federal provision being interpreted,” but that it is not required by the categorical approach. Brief for Respondent 23–25 (quoting Perrin v. United States, 444 U. S. 37, 45 (1979) ). We agree. In this case, state criminal codes aid our interpretation of “sexual abuse of a minor” by offering useful context.

581.US.2016_15-1391

New York General Business Law §518 provides that “[n]o seller in any sales transaction may impose a surcharge on a holder who elects to use a credit card in lieu of payment by cash, check, or similar means.” Petitioners, five New York businesses and their owners who wish to impose surcharges for credit card use, filed suit against state officials, arguing that the law violates the First Amendment by regulating how they communicate their prices, and that it is unconstitutionally vague. The District Court ruled in favor of the merchants, but the Court of Appeals vacated the judgment with instructions to dismiss. The Court of Appeals concluded that in the context of single-sticker pricing—where merchants post one price and would like to charge more to customers who pay by credit card—the law required that the sticker price be the same as the price charged to credit card users. In that context, the law regulated a relationship between two prices. Relying on this Court’s precedent holding that price regulation alone regulates conduct, not speech, the Court of Appeals concluded that §518 did not violate the First Amendment. The Court of Appeals abstained from reaching the merits of the constitutional challenge to pricing practices outside the single-sticker context. Held: 1. This Court’s review is limited to whether §518 is unconstitutional as applied to the particular pricing scheme that, before this Court, petitioners have argued they seek to employ: a single-sticker regime, in which merchants post a cash price and an additional credit card surcharge. Pp. 5–6. 2. Section 518 prohibits the pricing regime petitioners wish to employ. Section 518 does not define “surcharge.” Relying on the term’s ordinary meaning, the Court of Appeals concluded that a merchant imposes a surcharge when he posts a single sticker price and charges a credit card user more than that sticker price. This Court “generally accord[s] great deference to the interpretation and application of state law by the courts of appeals.” Pembaur v. Cincinnati, 475 U. S. 469 , n. 13. Because the interpretation of the Court of Appeals is not “clearly wrong,” Brockett v. Spokane Arcades, Inc., 472 U. S. 491 , n. 9, this Court follows that interpretation. Pp. 6–8. 3. Section 518 regulates speech. The Court of Appeals concluded that §518 posed no First Amendment problem because price controls regulate conduct, not speech. Section 518, however, is not like a typical price regulation, which simply regulates the amount a store can collect. The law tells merchants nothing about the amount they are allowed to collect from a cash or credit card payer. Instead, it regulates how sellers may communicate their prices. In regulating the communication of prices rather than prices themselves, §518 regulates speech. Because the Court of Appeals concluded otherwise, it did not determine whether §518 survives First Amendment scrutiny. On remand the Court of Appeals should analyze §518 as a speech regulation. Pp. 8–10. 4. Section 518 is not vague as applied to petitioners. As explained, §518 bans the single-sticker pricing petitioners argue they wish to employ, and “a plaintiff whose speech is clearly proscribed cannot raise a successful vagueness claim,” Holder v. Humanitarian Law Project, 561 U. S. 1 . Pp. 10–11. 808 F. 3d 118, vacated and remanded. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Thomas, Ginsburg, and Kagan, JJ., joined. Breyer, J., filed an opinion concurring in the judgment. Sotomayor, J., filed an opinion concurring in the judgment, in which Alito, J., joined.

Each time a customer pays for an item with a credit card, the merchant selling that item must pay a transaction fee to the credit card issuer. Some merchants balk at paying the fees and want to discourage the use of credit cards, or at least pass on the fees to customers who use them. One method of achieving those ends is through differential pricing—charging credit card users more than customers using cash. Merchants who wish to employ differential pricing may do so in two ways relevant here: impose a surcharge for the use of a credit card, or offer a discount for the use of cash. In N. Y. Gen. Bus. Law §518, New York has banned the former practice. The question presented is whether §518 regulates merchants’ speech and—if so—whether the statute violates the First Amendment. We conclude that §518 does regulate speech and remand for the Court of Appeals to determine in the first instance whether that regulation is unconstitutional. I A When credit cards were first introduced, contracts between card issuers and merchants barred merchants from charging credit card users higher prices than cash customers. Congress put a partial stop to this practice in the 1974 amendments to the Truth in Lending Act (TILA). The amendments prohibited card issuers from contractually preventing merchants from giving discounts to customers who paid in cash. See §306, 88Stat. 1515. The law, however, said nothing about surcharges for the use of credit. Two years later, Congress refined its dissimilar treatment of discounts and surcharges. First, the 1976 version of TILA barred merchants from imposing surcharges on customers who use credit cards. Act of Feb. 27, 1976, §3(c)(1), 90Stat. 197. Second, Congress added definitions of the two terms. A discount was “a reduction made from the regular price,” while a surcharge was “any means of increasing the regular price to a cardholder which is not imposed upon customers paying by cash, check, or similar means.” §3(a), ibid. In 1981, Congress further delineated the distinction between discounts and surcharges by defining “regular price.” Where a merchant “tagged or posted” a single price, the regular price was that single price. Cash Discount Act, §102(a), 95Stat. 144. If no price was tagged or posted, or if a merchant employed a two-tag approach—posting one price for credit and another for cash—the regular price was whatever was charged to credit card users. Ibid. Because a surcharge was defined as an increase from the regular price, there could be no credit card surcharge where the regular price was the same as the amount charged to customers using credit cards. The effect of all this was that a merchant could violate the surcharge ban only by posting a single price and charging credit card users more than that posted price. The federal surcharge ban was short lived. Congress allowed it to expire in 1984 and has not renewed the ban since. See §201, ibid. The provision preventing credit card issuers from contractually barring discounts for cash, however, remained in place. With the lapse of the federal surcharge ban, several States, New York among them, immediately enacted their own surcharge bans. Passed in 1984, N. Y. Gen. Bus. Law §518 adopted the operative language of the federal ban verbatim, providing that “[n]o seller in any sales transaction may impose a surcharge on a holder who elects to use a credit card in lieu of payment by cash, check, or similar means.” N. Y. Gen. Bus. Law Ann. §518 (West 2012); see also 15 U. S. C. §1666f(a)(2) (1982 ed.). Unlike the federal ban, the New York legislation included no definition of “surcharge.” In addition to these state legislative bans, credit card companies—though barred from prohibiting discounts for cash—included provisions in their contracts prohibiting merchants from imposing surcharges for credit card use. For most of its history, the New York law was essentially coextensive with these contractual prohibitions. In recent years, however, merchants have brought antitrust challenges to contractual no-surcharge provisions. Those suits have created uncertainty about the legal validity of such contractual surcharge bans. The result is that otherwise redundant legislative surcharge bans like §518 have increasingly gained importance, and increasingly come under scrutiny. B Petitioners, five New York businesses and their owners, wish to impose surcharges on customers who use credit cards. Each time one of their customers pays with a credit card, these merchants must pay some transaction fee to the company that issued the credit card. The fee is generally two to three percent of the purchase price. Those fees add up, and the merchants allege that they pay tens of thousands of dollars every year to credit card companies. Rather than increase prices across the board to absorb those costs, the merchants want to pass the fees along only to their customers who choose to use credit cards. They also want to make clear that they are not the bad guys—that the credit card companies, not the merchants, are responsible for the higher prices. The merchants believe that surcharges for credit are more effective than discounts for cash in accomplishing these goals. In 2013, after several major credit card issuers agreed to drop their contractual surcharge prohibitions, the merchants filed suit against the New York Attorney General and three New York District Attorneys to challenge §518—the only remaining obstacle to their charging surcharges for credit card use. As relevant here, they argued that the law violated the First Amendment by regulating how they communicated their prices, and that it was unconstitutionally vague because liability under the law “turn[ed] on the blurry difference” between surcharges and discounts. App. 39, Complaint ¶51. The District Court ruled in favor of the merchants. It read the statute as “draw[ing a] line between prohibited ‘surcharges’ and permissible ‘discounts’ based on words and labels, rather than economic realities.” 975 F. Supp. 2d 430, 444 (SDNY 2013). The court concluded that the law therefore regulated speech, and violated the First Amendment under this Court’s commercial speech doctrine. In addition, because the law turned on the “virtually incomprehensible distinction between what a vendor can and cannot tell its customers,” the District Court found that the law was unconstitutionally vague. Id., at 436. The Court of Appeals for the Second Circuit vacated the judgment of the District Court with instructions to dismiss the merchants’ claims. It began by considering single-sticker pricing, where merchants post one price and would like to charge more to customers who pay by credit card. All the law did in this context, the Court of Appeals explained, was regulate a relationship between two prices—the sticker price and the price charged to a credit card user—by requiring that the two prices be equal. Relying on our precedent holding that price regulation alone regulates conduct, not speech, the Court of Appeals concluded that §518 did not violate the First Amendment. The court also considered other types of pricing regimes—for example, posting separate cash and credit prices. The Court of Appeals thought it “far from clear” that §518 prohibited such pricing schemes. 808 F. 3d 118, 137 (CA2 2015). The federal surcharge ban on which §518 was modeled did not apply outside the single-sticker context, and the merchants had not clearly shown that §518 had a “broader reach” than the federal law. Ibid. Deciding that petitioners’ challenge in this regard “turn[ed] on an unsettled question of state law,” the Court of Appeals abstained from reaching the merits of the constitutional question beyond the single-sticker context. Id., at 135 (citing Railroad Comm’n of Tex. v. Pullman Co., 312 U. S. 496 (1941) ). We granted certiorari. 579 U. S. ___ (2016). II As a preliminary matter, we note that petitioners present us with a limited challenge. Observing that the merchants were not always particularly clear about the scope of their suit, the Court of Appeals deemed them to be bringing a facial attack on §518 as well as a challenge to the application of the statute to two particular pricing regimes: single-sticker pricing and two-sticker pricing. Before us, however, the merchants have disclaimed a facial challenge, assuring us that theirs is an as-applied challenge only. See Tr. of Oral Arg. 4–5, 18. There remains the question of what precise application of the law they seek to challenge. Although the merchants have presented a wide array of hypothetical pricing regimes, they have expressly identified only one pricing scheme that they seek to employ: posting a cash price and an additional credit card surcharge, expressed either as a percentage surcharge or a “dollars-and-cents” additional amount. See, e.g., App. 101–102, 104; Tr. of Oral Arg. 4–5, 18. Under this pricing approach, petitioner Expressions Hair Design might, for example, post a sign outside its salon reading “Haircuts $10 (we add a 3% surcharge if you pay by credit card).” Or, petitioner Brooklyn Farmacy & Soda Fountain might list one of the sundaes on its menu as costing “$10 (with a $0.30 surcharge for credit card users).” We take petitioners at their word and limit our review to the question whether §518 is unconstitutional as applied to this particular pricing practice.[1] III The next question is whether §518 prohibits the pricing regime petitioners wish to employ. The Court of Appeals concluded that it does. The court read “surcharge” in §518 to mean “an additional amount above the seller’s regular price,” and found it “basically self-evident” how §518 applies to sellers who post a single sticker price: “the sticker price is the ‘regular’ price, so sellers may not charge credit-card customers an additional amount above the sticker price that is not also charged to cash customers.” 808 F. 3d, at 128. Under this interpretation, signs of the kind that the merchants wish to post—“$10, with a $0.30 surcharge for credit card users”—violate §518 because they identify one sticker price—$10—and indicate that credit card users are charged more than that amount. “We generally accord great deference to the interpretation and application of state law by the courts of appeals.” Pembaur v. Cincinnati, 475 U. S. 469 , n. 13 (1986). This deference is warranted to “render unnecessary review of their decisions in this respect” and because lower fed-eral courts “are better schooled in and more able to interpret the laws of their respective States.” Brockett v. Spokane Arcades, Inc., 472 U. S. 491, 500 (1985) (quoting Cort v. Ash, 422 U. S. 66 , n. 6 (1975); internal quotation marks omitted). “[W]e surely have the authority to differ with the lower federal courts as to the meaning of a state statute,” and have done so in instances where the lower court’s construction was “clearly wrong” or “plain error.” 472 U. S., at 500, and n. 9 (internal quotation marks omitted). But that is not the case here. Section 518 does not define “surcharge,” but the Court of Appeals looked to the ordinary meaning of the term: “a charge in excess of the usual or normal amount.” 808 F. 3d, at 127 (quoting Webster’s Third New International Dictionary 2299 (2002); internal quotation marks omitted). Where a seller posts a single sticker price, it is reasonable to treat that sticker price as the “usual or normal amount” and conclude, as the court below did, that a merchant imposes a surcharge when he charges a credit card user more than that sticker price. In short, we cannot dismiss the Court of Appeals’ interpretation of §518 as “clearly wrong.” Accordingly, consistent with our customary practice, we follow that interpretation. IV Having concluded that §518 bars the pricing regime petitioners wish to employ, we turn to their constitutional arguments: that the law unconstitutionally regulates speech and is impermissibly vague. A The Court of Appeals concluded that §518 posed no First Amendment problem because the law regulated conduct, not speech.[2] In reaching this conclusion, the Court of Appeals began with the premise that price controls regulate conduct alone. See 44 Liquormart, Inc. v. Rhode Island, 517 U. S. 484, 507 (1996) (plurality opinion); id., at 524 (Thomas, J., concurring in part and concurring in judgment); id., at 530 (O’Connor, J., concurring in judgment). Section 518 regulates the relationship between“(1) the seller’s sticker price and (2) the price the seller charges to credit card customers,” requiring that these two amounts be equal. 808 F. 3d, at 131. A law regulating the relationship between two prices regulates speech nomore than a law regulating a single price. The Court of Ap-peals concluded that §518 was therefore simply a conduct regulation. But §518 is not like a typical price regulation. Such a regulation—for example, a law requiring all New York delis to charge $10 for their sandwiches—would simply regulate the amount that a store could collect. In other words, it would regulate the sandwich seller’s conduct. To be sure, in order to actually collect that money, a store would likely have to put “$10” on its menus or have its employees tell customers that price. Those written or oral communications would be speech, and the law—by determining the amount charged—would indirectly dictate the content of that speech. But the law’s effect on speech would be only incidental to its primary effect on conduct, and “it has never been deemed an abridgment of freedom of speech or press to make a course of conduct illegal merely because the conduct was in part initiated, evidenced, or carried out by means of language, either spoken, written, or printed.” Rumsfeld v. Forum for Aca-demic and Institutional Rights, Inc., 547 U. S. 47, 62 (2006) (quoting Giboney v. Empire Storage & Ice Co., 336 U. S. 490, 502 (1949) ; internal quotation marks omitted); see also Sorrell v. IMS Health Inc., 564 U. S. 552, 567 (2011) . Section 518 is different. The law tells merchants nothing about the amount they are allowed to collect from a cash or credit card payer. Sellers are free to charge $10 for cash and $9.70, $10, $10.30, or any other amount for credit. What the law does regulate is how sellers may communicate their prices. A merchant who wants to charge $10 for cash and $10.30 for credit may not convey that price any way he pleases. He is not free to say “$10, with a 3% credit card surcharge” or “$10, plus $0.30 for credit” because both of those displays identify a single sticker price—$10—that is less than the amount credit card users will be charged. Instead, if the merchant wishes to post a single sticker price, he must display $10.30 ashis sticker price. Accordingly, while we agree with the Court of Appeals that §518 regulates a relationship between a sticker price and the price charged to credit card users, we cannot accept its conclusion that §518 is nothing more than a mine-run price regulation. In regulating the communication of prices rather than prices themselves, §518 regulates speech. Because it concluded otherwise, the Court of Appeals had no occasion to conduct a further inquiry into whether §518, as a speech regulation, survived First Amendment scrutiny. On that question, the parties dispute whether §518 is a valid commercial speech regulation under Central Hudson Gas & Elec. Corp. v. Public Serv. Comm’n of N. Y., 447 U. S. 557 (1980) , and whether the law can be upheld as a valid disclosure requirement under Zauderer v. Office of Disciplinary Counsel of Supreme Court of Ohio, 471 U. S. 626 (1985) . “[W]e are a court of review, not of first view.” Nautilus, Inc. v. Biosig Instruments, Inc., 572 U. S. ___, ___ (2014) (slip op., at 14) (internal quotation marks omitted). Accordingly, we decline to consider those questions in the first instance. Instead, we remand for the Court of Appeals to analyze §518 as a speech regulation.[3] B Given the way the merchants have presented their case, their vagueness challenge gives us little pause. Before this Court, the only pricing practice they express an interest in employing is a single-sticker regime, listing one price and a separate surcharge amount. As we have explained, §518 bars them from doing so. “[A] plaintiff whose speech is clearly proscribed cannot raise a successful vagueness claim.” Holder v. Humanitarian Law Project, 561 U. S. 1, 20 (2010) . Although the merchants argue that “no one can seem to put a finger on just how far the law sweeps,” Brief for Petitioners 51, it is at least clear that §518 proscribes their intended speech. Accordingly, the law is not vague as applied to them.[4] C The judgment of the Court of Appeals for the Second Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Petitioner Expressions Hair Design currently posts separate dollars-and-cents prices for cash and credit—that is, it posts something like “$10 cash, $10.30 credit.” It displays its prices in this way, however, only because it considers itself compelled to do so by the challenged law if it wants to charge different prices. Prior to becoming aware of the law, Expressions posted single prices along with a notice informing customers that a three percent surcharge would be added to their bill if they paid by credit card. Expressions has indicated that it would prefer to return to its prior practice. See App. 19, Complaint ¶3; id., at 103–104. Given petitioners’ representations about the narrow scope of their as-applied challenge, we limit our consideration to the single-sticker pricing regime for present purposes. Petitioners’ affidavits and briefing reference other potential pricing schemes, which may be considered by the Court of Appeals to the extent it deems appropriate. See, e.g., id., at 56; Brief for Petitioners 50. 2 Relying fully on their claim that §518 regulated speech, petitioners did not advance any argument before the Court of Appeals that §518 was constitutionally problematic even if deemed a regulation of conduct. See 808 F. 3d 118, 135 (CA2 2015) (noting that petitioners had not challenged §518 under United States v. O’Brien, 391 U. S. 367 (1968) ). 3 To assess the statute’s constitutionality, the Court of Appeals may need to consider a question we need not answer here: whether the statute permits two-sticker pricing schemes like the one petitioner Expressions currently uses, see n. 1, supra. Respondents’ argument that §518 is a constitutionally valid disclosure requirement rests on an interpretation of the statute that allows such two-sticker schemes. 4 For similar reasons, petitioners’ related argument regarding abstention is no longer at issue. The Court of Appeals abstained from deciding whether §518 was constitutional outside of the single-sticker context, but the merchants have disavowed any intent to challenge the law outside of this context.

580.US.2016_15-497

The Individuals with Disabilities Education Act (IDEA) offers federal funds to States in exchange for a commitment to furnish a “free appropriate public education” (FAPE) to children with certain disabilities, 20 U. S. C. §1412(a)(1)(A), and establishes formal administrative procedures for resolving disputes between parents and schools concerning the provision of a FAPE. Other federal statutes also protect the interests of children with disabilities, including Title II of the Americans with Disabilities Act (ADA) and §504 of the Rehabilitation Act. In Smith v. Robinson, 468 U. S. 992 , this Court considered the interaction between those other laws and the IDEA, holding that the IDEA was “the exclusive avenue” through which a child with a disability could challenge the adequacy of his education. Id., at 1009. Congress responded by passing the Handicapped Children’s Protection Act of 1986, overturning Smith’s preclusion of non-IDEA claims and adding a carefully defined exhaustion provision. Under that provision, a plaintiff bringing suit under the ADA, the Rehabilitation Act, or similar laws “seeking relief that is also available under [the IDEA]” must first exhaust the IDEA’s administrative procedures. §1415(l). Petitioner E. F. is a child with a severe form of cerebral palsy; a trained service dog named Wonder assists her with various daily life activities. When E. F.’s parents, petitioners Stacy and Brent Fry, sought permission for Wonder to join E. F. in kindergarten, officials at Ezra Eby Elementary School refused. The officials reasoned that the human aide provided as part of E. F.’s individualized education program rendered the dog superfluous. In response, the Frys removed E. F. from Ezra Eby and began homeschooling her. They also filed a complaint with the Department of Education’s Office for Civil Rights (OCR), claiming that the exclusion of E. F.’s service animal violated her rights under Title II and §504. OCR agreed, and school officials invited E. F. to return to Ezra Eby with Wonder. But the Frys, concerned about resentment from school officials, instead enrolled E. F. in a different school that welcomed the service dog. The Frys then filed this suit in federal court against Ezra Eby’s local and regional school districts and principal (collectively, the school districts), alleging that they violated Title II and §504 and seeking declaratory and monetary relief. The District Court granted the school districts’ motion to dismiss the suit, holding that §1415(l) required the Frys to first exhaust the IDEA’s administrative procedures. The Sixth Circuit affirmed, reasoning that §1415(l) applies whenever a plaintiff’s alleged harms are “educational” in nature. Held: 1. Exhaustion of the IDEA’s administrative procedures is unnecessary where the gravamen of the plaintiff’s suit is something other than the denial of the IDEA’s core guarantee of a FAPE. Pp. 9–18. (a) The language of §1415(l) compels exhaustion when a plaintiff seeks “relief” that is “available” under the IDEA. Establishing the scope of §1415(l), then, requires identifying the circumstances in which the IDEA enables a person to obtain redress or access a benefit. That inquiry immediately reveals the primacy of a FAPE in the statutory scheme. The IDEA’s stated purpose and specific commands center on ensuring a FAPE for children with disabilities. And the IDEA’s administrative procedures test whether a school has met this obligation: Any decision by a hearing officer on a request for substantive relief “shall” be “based on a determination of whether the child received a free appropriate public education.” §1415(f)(3)(E)(i). Accordingly, §1415(l)’s exhaustion rule hinges on whether a lawsuit seeks relief for the denial of a FAPE. If a lawsuit charges such a denial, the plaintiff cannot escape §1415(l) merely by bringing the suit under a statute other than the IDEA. But if the remedy sought in a suit brought under a different statute is not for the denial of a FAPE, then exhaustion of the IDEA’s procedures is not required. Pp. 9–13. (b) In determining whether a plaintiff seeks relief for the denial of a FAPE, what matters is the gravamen of the plaintiff’s complaint, setting aside any attempts at artful pleading. That inquiry makes central the plaintiff’s own claims, as §1415(l) explicitly requires in asking whether a lawsuit in fact “seeks” relief available under the IDEA. But examination of a plaintiff’s complaint should consider substance, not surface: §1415(l) requires exhaustion when the gravamen of a complaint seeks redress for a school’s failure to provide a FAPE, even if not phrased or framed in precisely that way. In addressing whether a complaint fits that description, a court should attend to the diverse means and ends of the statutes covering persons with disabilities. The IDEA guarantees individually tailored educational services for children with disabilities, while Title II and §504 promise nondiscriminatory access to public institutions for people with disabilities of all ages. That is not to deny some overlap in coverage: The same conduct might violate all three statutes. But still, these statutory differences mean that a complaint brought under Title II and §504 might instead seek relief for simple discrimination, irrespective of the IDEA’s FAPE obligation. One clue to the gravamen of a complaint can come from asking a pair of hypothetical questions. First, could the plaintiff have brought essentially the same claim if the alleged conduct had occurred at a public facility that was not a school? Second, could an adult at the school have pressed essentially the same grievance? When the answer to those questions is yes, a complaint that does not expressly allege the denial of a FAPE is also unlikely to be truly about that subject. But when the answer is no, then the complaint probably does concern a FAPE. A further sign of the gravamen of a suit can emerge from the history of the proceedings. Prior pursuit of the IDEA’s administrative remedies may provide strong evidence that the substance of a plaintiff’s claim concerns the denial of a FAPE, even if the complaint never explicitly uses that term. Pp. 13–18. 2. This case is remanded to the Court of Appeals for a proper analysis of whether the gravamen of E. F.’s complaint charges, and seeks relief for, the denial of a FAPE. The Frys’ complaint alleges only disability-based discrimination, without making any reference to the adequacy of the special education services E. F.’s school provided. Instead, the Frys have maintained that the school districts infringed E. F.’s right to equal access—even if their actions complied in full with the IDEA’s requirements. But the possibility remains that the history of these proceedings might suggest something different. The parties have not addressed whether the Frys initially pursued the IDEA’s administrative remedies, and the record is cloudy as to the relevant facts. On remand, the court below should establish whether (or to what extent) the Frys invoked the IDEA’s dispute resolution process before filing suit. And if the Frys started down that road, the court should decide whether their actions reveal that the gravamen of their complaint is indeed the denial of a FAPE, thus necessitating further exhaustion. Pp. 18–20. 788 F. 3d 622, vacated and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined. Alito, J., filed an opinion concurring in part and concurring in the judgment, in which Thomas, J., joined.

The Individuals with Disabilities Education Act (IDEA or Act), 84Stat. 175, as amended, 20 U. S. C. §1400 et seq., ensures that children with disabilities receive needed special education services. One of its provisions, §1415(l), addresses the Act’s relationship with other laws protecting those children. Section 1415(l) makes clear that nothing in the IDEA “restrict[s] or limit[s] the rights [or] remedies” that other federal laws, including antidiscrimination statutes, confer on children with disabilities. At the same time, the section states that if a suit brought under such a law “seek[s] relief that is also available under” the IDEA, the plaintiff must first exhaust the IDEA’s administrative procedures. In this case, we consider the scope of that exhaustion requirement. We hold that exhaustion is not necessary when the gravamen of the plaintiff’s suit is something other than the denial of the IDEA’s core guarantee—what the Act calls a “free appropriate public education.” §1412(a)(1)(A). I A The IDEA offers federal funds to States in exchange for a commitment: to furnish a “free appropriate public education”—more concisely known as a FAPE—to all children with certain physical or intellectual disabilities. Ibid.; see §1401(3)(A)(i) (listing covered disabilities). As defined in the Act, a FAPE comprises “special education and related services”—both “instruction” tailored to meet a child’s “unique needs” and sufficient “supportive services” to permit the child to benefit from that instruction. §§1401(9), (26), (29); see Board of Ed. of Hendrick Hudson Central School Dist., Westchester Cty. v. Rowley, 458 U. S. 176, 203 (1982) . An eligible child, as this Court has explained, acquires a “substantive right” to such an education once a State accepts the IDEA’s financial assistance. Smith v. Robinson, 468 U. S. 992, 1010 (1984) . Under the IDEA, an “individualized education program,” called an IEP for short, serves as the “primary vehicle” for providing each child with the promised FAPE. Honig v. Doe, 484 U. S. 305, 311 (1988) ; see §1414(d). (Welcome to—and apologies for—the acronymic world of federal legislation.) Crafted by a child’s “IEP Team”—a group of school officials, teachers, and parents—the IEP spells out a personalized plan to meet all of the child’s “educational needs.” §§1414(d)(1)(A)(i)(II)(bb), (d)(1)(B). Most notably, the IEP documents the child’s current “levels of academic achievement,” specifies “measurable an-nual goals” for how she can “make progress in the general education curriculum,” and lists the “special education and related services” to be provided so that she can “advance appropriately toward [those] goals.” §§1414(d)(1)(A)(i)(I), (II), (IV)(aa). Because parents and school representatives sometimes cannot agree on such issues, the IDEA establishes formal procedures for resolving disputes. To begin, a dissatisfied parent may file a complaint as to any matter concerning the provision of a FAPE with the local or state educational agency (as state law provides). See §1415(b)(6). That pleading generally triggers a “[p]reliminary meeting” involving the contending parties, §1415(f )(1)(B)(i); at their option, the parties may instead (or also) pursue a full-fledged mediation process, see §1415(e). Assuming their impasse continues, the matter proceeds to a “due pro-cess hearing” before an impartial hearing officer. §1415(f )(1)(A); see §1415(f )(3)(A)(i). Any decision of the officer granting substantive relief must be “based on a determination of whether the child received a [FAPE].” §1415(f )(3)(E)(i). If the hearing is initially conducted at the local level, the ruling is appealable to the state agency. See §1415(g). Finally, a parent unhappy with the outcome of the administrative process may seek judicial review by filing a civil action in state or federal court. See §1415(i)(2)(A). Important as the IDEA is for children with disabilities, it is not the only federal statute protecting their interests. Of particular relevance to this case are two antidiscrimination laws—Title II of the Americans with Disabilities Act (ADA), 42 U. S. C. §12131 et seq., and §504 of the Rehabilitation Act, 29 U. S. C. §794—which cover both adults and children with disabilities, in both public schools and other settings. Title II forbids any “public entity” from discriminating based on disability; Section 504 applies the same prohibition to any federally funded “program or activity.” 42 U. S. C. §§12131–12132; 29 U. S. C. §794(a). A regulation implementing Title II requires a public entity to make “reasonable modifications” to its “policies, practices, or procedures” when necessary to avoid such discrimination. 28 CFR §35.130(b)(7) (2016); see, e.g., Alboniga v. School Bd. of Broward Cty., 87 F. Supp. 3d 1319, 1345 (SD Fla. 2015) (requiring an accommodation to permit use of a service animal under Title II). In similar vein, courts have interpreted §504 as demanding certain “reasonable” modifications to existing practices in order to “accommodate” persons with disabilities. Alexander v. Choate, 469 U. S. 287 –300 (1985); see, e.g., Sullivan v. Vallejo City Unified School Dist., 731 F. Supp. 947, 961–962 (ED Cal. 1990) (requiring an accommodation to permit use of a service animal under §504). And both statutes authorize individuals to seek redress for violations of their substantive guarantees by bringing suits for injunctive relief or money damages. See 29 U. S. C. §794a(a)(2); 42 U. S. C. §12133. This Court first considered the interaction between such laws and the IDEA in Smith v. Robinson, 468 U. S. 992 .[1] The plaintiffs there sought “to secure a ‘free appropriate public education’ for [their] handicapped child.” Id., at 994. But instead of bringing suit under the IDEA alone, they appended “virtually identical” claims (again alleging the denial of a “free appropriate public education”) under §504 of the Rehabilitation Act and the Fourteenth Amendment’s Equal Protection Clause. Id., at 1009; see id., at 1016. The Court held that the IDEA altogether foreclosed those additional claims: With its “comprehensive” and “carefully tailored” provisions, the Act was “the exclusive avenue” through which a child with a disability (or his parents) could challenge the adequacy of his education. Id., at 1009; see id., at 1013, 1016, 1021. Congress was quick to respond. In the Handicapped Children’s Protection Act of 1986, 100Stat. 796, it overturned Smith’s preclusion of non-IDEA claims while also adding a carefully defined exhaustion requirement. Now codified at 20 U. S. C. §1415(l), the relevant provision of that statute reads: “Nothing in [the IDEA] shall be construed to restrict or limit the rights, procedures, and remedies available under the Constitution, the [ADA], title V of the Rehabilitation Act [including §504], or other Federal laws protecting the rights of children with disabilities, except that before the filing of a civil action under such laws seeking relief that is also available under [the IDEA], the [IDEA’s administrative procedures] shall be exhausted to the same extent as would be required had the action been brought under [the IDEA].” The first half of §1415(l) (up until “except that”) “reaffirm[s] the viability” of federal statutes like the ADA or Rehabilitation Act “as separate vehicles,” no less integral than the IDEA, “for ensuring the rights of handicapped children.” H. R. Rep. No. 99–296, p. 4 (1985); see id., at 6. According to that opening phrase, the IDEA does not prevent a plaintiff from asserting claims under such laws even if, as in Smith itself, those claims allege the denial of an appropriate public education (much as an IDEA claim would). But the second half of §1415(l) (from “except that” onward) imposes a limit on that “anything goes” regime, in the form of an exhaustion provision. According to that closing phrase, a plaintiff bringing suit under the ADA, the Rehabilitation Act, or similar laws must in certain circumstances—that is, when “seeking relief that is also available under” the IDEA—first exhaust the IDEA’s administrative procedures. The reach of that requirement is the issue in this case. B Petitioner E. F. is a child with a severe form of cerebral palsy, which “significantly limits her motor skills and mobility.” App. to Brief in Opposition 6, Complaint ¶19.[2] When E. F. was five years old, her parents—petitioners Stacy and Brent Fry—obtained a trained service dog for her, as recommended by her pediatrician. The dog, a goldendoodle named Wonder, “help[s E. F.] to live as independently as possible” by assisting her with various life activities. Id., at 2, ¶3. In particular, Wonder aids E. F. by “retrieving dropped items, helping her balance when she uses her walker, opening and closing doors, turning on and off lights, helping her take off her coat, [and] helping her transfer to and from the toilet.” Id., at 7, ¶27. But when the Frys sought permission for Wonder to join E. F. in kindergarten, officials at Ezra Eby Elementary School refused the request. Under E. F.’s existing IEP, a human aide provided E. F. with one-on-one support throughout the day; that two-legged assistance, the school officials thought, rendered Wonder superfluous. In the words of one administrator, Wonder should be barred from Ezra Eby because all of E. F.’s “physical and academic needs [were] being met through the services/programs/accommodations” that the school had already agreed to. Id., at 8, ¶33. Later that year, the school officials briefly allowed Wonder to accompany E. F. to school on a trial basis; but even then, “the dog was required to remain in the back of the room during classes, and was forbidden from assisting [E. F.] with many tasks he had been specifically trained to do.” Ibid., ¶35. And when the trial period concluded, the administrators again informed the Frys that Wonder was not welcome. As a result, the Frys removed E. F. from Ezra Eby and began homeschooling her. In addition, the Frys filed a complaint with the U. S. Department of Education’s Office for Civil Rights (OCR), charging that Ezra Eby’s exclusion of E. F.’s service animal violated her rights under Title II of the ADA and §504 of the Rehabilitation Act. Following an investigation, OCR agreed. The office explained in its decision letter that a school’s obligations under those statutes go beyond providing educational services: A school could offer a FAPE to a child with a disability but still run afoul of the laws’ ban on discrimination. See App. 30–32. And here, OCR found, Ezra Eby had indeed violated that ban, even if its use of a human aide satisfied the FAPE standard. See id., at 35–36. OCR analogized the school’s conduct to “requir[ing] a student who uses a wheelchair to be carried” by an aide or “requir[ing] a blind student to be led [around by a] teacher” instead of permitting him to use a guide dog or cane. Id., at 35. Regardless whether those—or Ezra Eby’s—policies denied a FAPE, they violated Title II and §504 by discriminating against children with disabilities. See id., at 35–36. In response to OCR’s decision, school officials at last agreed that E. F. could come to school with Wonder. But after meeting with Ezra Eby’s principal, the Frys became concerned that the school administration “would resent [E. F.] and make her return to school difficult.” App. to Brief in Opposition 10, ¶48. Accordingly, the Frys found a different public school, in a different district, where administrators and teachers enthusiastically received both E. F. and Wonder. C The Frys then filed this suit in federal court against the local and regional school districts in which Ezra Eby is located, along with the school’s principal (collectively, the school districts). The complaint alleged that the school districts violated Title II of the ADA and §504 of the Rehabilitation Act by “denying [E. F.] equal access” to Ezra Eby and its programs, “refus[ing] to reasonably accommodate” E. F.’s use of a service animal, and otherwise “discriminat[ing] against [E. F.] as a person with disabilities.” Id., at 15, ¶68, 17–18, ¶¶82–83. According to the complaint, E. F. suffered harm as a result of that discrimination, including “emotional distress and pain, embarrassment, [and] mental anguish.” Id., at 11–12, ¶51. In their prayer for relief, the Frys sought a declaration that the school districts had violated Title II and §504, along with money damages to compensate for E. F.’s injuries. The District Court granted the school districts’ motion to dismiss the suit, holding that §1415(l) required the Frys to first exhaust the IDEA’s administrative procedures. See App. to Pet. for Cert. 50. A divided panel of the Court of Appeals for the Sixth Circuit affirmed on the same ground. In that court’s view, §1415(l) applies if “the injuries [alleged in a suit] relate to the specific substantive protections of the IDEA.” 788 F. 3d 622, 625 (2015). And that means, the court continued, that exhaustion is necessary whenever “the genesis and the manifestations” of the complained-of harms were “educational” in nature. Id., at 627 (quoting Charlie F. v. Board of Ed. of Skokie School Dist. 68, 98 F. 3d 989, 993 (CA7 1996)). On that understanding of §1415(l), the Sixth Circuit held, the Frys’ suit could not proceed: Because the harms to E. F. were generally “educational”—most notably, the court reasoned, because “Wonder’s absence hurt her sense of independence and social confidence at school”—the Frys had to exhaust the IDEA’s procedures. 788 F. 3d, at 627. Judge Daugh-trey dissented, emphasizing that in bringing their Title II and §504 claims, the Frys “did not allege the denial of a FAPE” or “seek to modify [E. F.’s] IEP in any way.” Id., at 634. We granted certiorari to address confusion in the courts of appeals as to the scope of §1415(l)’s exhaustion requirement. 579 U. S. ___ (2016).[3] We now vacate the Sixth Circuit’s decision. II Section 1415(l) requires that a plaintiff exhaust the IDEA’s procedures before filing an action under the ADA, the Rehabilitation Act, or similar laws when (but only when) her suit “seek[s] relief that is also available” under the IDEA. We first hold that to meet that statutory standard, a suit must seek relief for the denial of a FAPE, because that is the only “relief” the IDEA makes “avail-able.” We next conclude that in determining whether a suit indeed “seeks” relief for such a denial, a court should look to the substance, or gravamen, of the plaintiff’s complaint.[4] A In this Court, the parties have reached substantial agreement about what “relief” the IDEA makes “avail-able” for children with disabilities—and about how the Sixth Circuit went wrong in addressing that question. The Frys maintain that such a child can obtain remedies under the IDEA for decisions that deprive her of a FAPE, but none for those that do not. So in the Frys’ view, §1415(l)’s exhaustion requirement can come into play only when a suit concerns the denial of a FAPE—and not, as the Sixth Circuit held, when it merely has some articulable connection to the education of a child with a disability. See Reply Brief 13–15. The school districts, for their part, also believe that the Sixth Circuit’s exhaustion standard “goes too far” because it could mandate exhaustion when a plaintiff is “seeking relief that is not in substance avail-able” under the IDEA. Brief for Respondents 30. And in particular, the school districts acknowledge that the IDEA makes remedies available only in suits that “directly implicate[ ]” a FAPE—so that only in those suits can §1415(l) apply. Tr. of Oral Arg. 46. For the reasons that follow, we agree with the parties’ shared view: The only relief that an IDEA officer can give—hence the thing a plaintiff must seek in order to trigger §1415(l)’s exhaustion rule—is relief for the denial of a FAPE. We begin, as always, with the statutory language at issue, which (at risk of repetition) compels exhaustion when a plaintiff seeks “relief” that is “available” under the IDEA. The ordinary meaning of “relief” in the context of a lawsuit is the “redress[ ] or benefit” that attends a favor-able judgment. Black’s Law Dictionary 1161 (5th ed. 1979). And such relief is “available,” as we recently explained, when it is “accessible or may be obtained.” Ross v. Blake, 578 U. S. ___, ___ (2016) (slip op., at 8) (quoting Webster’s Third New International Dictionary 150 (1993)). So to establish the scope of §1415(l), we must identify the circumstances in which the IDEA enables a person to obtain redress (or, similarly, to access a benefit). That inquiry immediately reveals the primacy of a FAPE in the statutory scheme. In its first section, the IDEA declares as its first purpose “to ensure that all children with disabilities have available to them a free appropriate public education.” §1400(d)(1)(A). That principal purpose then becomes the Act’s principal command: A State receiving federal funding under the IDEA must make such an education “available to all children with disabilities.” §1412(a)(1)(A). The guarantee of a FAPE to those children gives rise to the bulk of the statute’s more specific provisions. For example, the IEP—“the centerpiece of the statute’s education delivery system”—serves as the “vehicle” or “means” of providing a FAPE. Honig, 484 U. S., at 311; Rowley, 458 U. S., at 181; see supra, at 2. And finally, as all the above suggests, the FAPE requirement provides the yardstick for measuring the adequacy of the education that a school offers to a child with a disability: Under that standard, this Court has held, a child is entitled to “meaningful” access to education based on her individual needs. Rowley, 458 U. S., at 192.[5] The IDEA’s administrative procedures test whether a school has met that obligation—and so center on the Act’s FAPE requirement. As noted earlier, any decision by a hearing officer on a request for substantive relief “shall” be “based on a determination of whether the child received a free appropriate public education.” §1415(f )(3)(E)(i); see supra, at 3.[6] Or said in Latin: In the IDEA’s administrative process, a FAPE denial is the sine qua non. Suppose that a parent’s complaint protests a school’s failure to provide some accommodation for a child with a disability. If that accommodation is needed to fulfill the IDEA’s FAPE requirement, the hearing officer must order relief. But if it is not, he cannot—even though the dispute is between a child with a disability and the school she attends. There might be good reasons, unrelated to a FAPE, for the school to make the requested accommodation. Indeed, another federal law (like the ADA or Rehabilitation Act) might require the accommodation on one of those alternative grounds. See infra, at 15. But still, the hearing officer cannot provide the requested relief. His role, under the IDEA, is to enforce the child’s “substantive right” to a FAPE. Smith, 468 U. S., at 1010. And that is all.[7] For that reason, §1415(l)’s exhaustion rule hinges on whether a lawsuit seeks relief for the denial of a free appropriate public education. If a lawsuit charges such a denial, the plaintiff cannot escape §1415(l) merely by bringing her suit under a statute other than the IDEA—as when, for example, the plaintiffs in Smith claimed that a school’s failure to provide a FAPE also violated the Rehabilitation Act.[8] Rather, that plaintiff must first submit her case to an IDEA hearing officer, experienced in addressing exactly the issues she raises. But if, in a suit brought under a different statute, the remedy sought is not for the denial of a FAPE, then exhaustion of the IDEA’s procedures is not required. After all, the plaintiff could not get any relief from those procedures: A hearing officer, as just explained, would have to send her away empty-handed. And that is true even when the suit arises directly from a school’s treatment of a child with a disability—and so could be said to relate in some way to her education. A school’s conduct toward such a child—say, some refusal to make an accommodation—might injure her in ways unrelated to a FAPE, which are addressed in statutes other than the IDEA. A complaint seeking redress for those other harms, independent of any FAPE denial, is not subject to §1415(l)’s exhaustion rule because, once again, the only “relief” the IDEA makes “available” is relief for the denial of a FAPE. B Still, an important question remains: How is a court to tell when a plaintiff “seeks” relief for the denial of a FAPE and when she does not? Here, too, the parties have found some common ground: By looking, they both say, to the “substance” of, rather than the labels used in, the plaintiff’s complaint. Brief for Respondents 20; Reply Brief 7–8. And here, too, we agree with that view: What matters is the crux—or, in legal-speak, the gravamen—of the plaintiff’s complaint, setting aside any attempts at artful pleading. That inquiry makes central the plaintiff’s own claims, as §1415(l) explicitly requires. The statutory language asks whether a lawsuit in fact “seeks” relief available under the IDEA—not, as a stricter exhaustion statute might, whether the suit “could have sought” relief avail-able under the IDEA (or, what is much the same, whether any remedies “are” available under that law). See Brief for United States as Amicus Curiae 20 (contrasting §1415(l) with the exhaustion provision in the Prison Litigation Reform Act, 42 U. S. C. §1997e(a)). In effect, §1415(l) treats the plaintiff as “the master of the claim”: She identifies its remedial basis—and is subject to exhaustion or not based on that choice. Caterpillar Inc. v. Williams, 482 U. S. 386 , and n. 7 (1987). A court deciding whether §1415(l) applies must therefore examine whether a plaintiff’s complaint—the principal instrument by which she describes her case—seeks relief for the de-nial of an appropriate education. But that examination should consider substance, not surface. The use (or non-use) of particular labels and terms is not what matters. The inquiry, for example, does not ride on whether a complaint includes (or, alternatively, omits) the precise words(?) “FAPE” or “IEP.” After all, §1415(l)’s premise is that the plaintiff is suing under a statute other than the IDEA, like the Rehabilitation Act; in such a suit, the plaintiff might see no need to use the IDEA’s distinctive language—even if she is in essence contesting the adequacy of a special education program. And still more critically, a “magic words” approach would make §1415(l)’s exhaustion rule too easy to bypass. Just last Term, a similar worry led us to hold that a court’s jurisdiction under the Foreign Sovereign Immunities Act turns on the “gravamen,” or “essentials,” of the plaintiff’s suit. OBB Personenverkehr AG v. Sachs, 577 U. S. ___, ___, ___, ___ (2015) (slip op., at 6, 8, 9). “[A]ny other approach,” we explained, “would allow plaintiffs to evade the Act’s restrictions through artful pleading.” Id., at ___ (slip op., at 8). So too here. Section 1415(l) is not merely a pleading hurdle. It requires exhaustion when the gravamen of a complaint seeks redress for a school’s failure to provide a FAPE, even if not phrased or framed in precisely that way. In addressing whether a complaint fits that description, a court should attend to the diverse means and ends of the statutes covering persons with disabilities—the IDEA on the one hand, the ADA and Rehabilitation Act (most notably) on the other. The IDEA, of course, protects only “children” (well, really, adolescents too) and concerns only their schooling. §1412(a)(1)(A). And as earlier noted, the statute’s goal is to provide each child with meaningful access to education by offering individualized instruction and related services appropriate to her “unique needs.” §1401(29); see Rowley, 458 U. S., at 192, 198; supra, at 11. By contrast, Title II of the ADA and §504 of the Rehabilitation Act cover people with disabilities of all ages, and do so both inside and outside schools. And those statutes aim to root out disability-based discrimination, enabling each covered person (sometimes by means of reasonable accommodations) to participate equally to all others in public facilities and federally funded programs. See supra, at 3–4. In short, the IDEA guarantees individually tailored educational services, while Title II and §504 promise non-discriminatory access to public institutions. That is not to deny some overlap in coverage: The same conduct might violate all three statutes—which is why, as in Smith, a plaintiff might seek relief for the denial of a FAPE under Title II and §504 as well as the IDEA. But still, the statutory differences just discussed mean that a complaint brought under Title II and §504 might instead seek relief for simple discrimination, irrespective of the IDEA’s FAPE obligation. One clue to whether the gravamen of a complaint against a school concerns the denial of a FAPE, or instead addresses disability-based discrimination, can come from asking a pair of hypothetical questions. First, could the plaintiff have brought essentially the same claim if the alleged conduct had occurred at a public facility that was not a school—say, a public theater or library? And second, could an adult at the school—say, an employee or visitor—have pressed essentially the same grievance? When the answer to those questions is yes, a complaint that does not expressly allege the denial of a FAPE is also unlikely to be truly about that subject; after all, in those other situations there is no FAPE obligation and yet the same basic suit could go forward. But when the answer is no, then the complaint probably does concern a FAPE, even if it does not explicitly say so; for the FAPE requirement is all that explains why only a child in the school setting (not an adult in that setting or a child in some other) has a viable claim. Take two contrasting examples. Suppose first that a wheelchair-bound child sues his school for discrimination under Title II (again, without mentioning the denial of a FAPE) because the building lacks access ramps. In some sense, that architectural feature has educational consequences, and a different lawsuit might have alleged that it violates the IDEA: After all, if the child cannot get inside the school, he cannot receive instruction there; and if he must be carried inside, he may not achieve the sense of independence conducive to academic (or later to real-world) success. But is the denial of a FAPE really the gravamen of the plaintiff’s Title II complaint? Consider that the child could file the same basic complaint if a municipal library or theater had no ramps. And similarly, an employee or visitor could bring a mostly identical complaint against the school. That the claim can stay the same in those alternative scenarios suggests that its essence is equality of access to public facilities, not adequacy of special education. See supra, at 7 (describing OCR’s use of a similar example). And so §1415(l) does not require exhaustion.[9] But suppose next that a student with a learning disability sues his school under Title II for failing to provide remedial tutoring in mathematics. That suit, too, might be cast as one for disability-based discrimination, grounded on the school’s refusal to make a reasonable accommo-dation; the complaint might make no reference at all to a FAPE or an IEP. But can anyone imagine the student making the same claim against a public theater or library? Or, similarly, imagine an adult visitor or employee suing the school to obtain a math tutorial? The difficulty of transplanting the complaint to those other contexts suggests that its essence—even though not its wording—is the provision of a FAPE, thus bringing §1415(l ) into play.[10] A further sign that the gravamen of a suit is the denial of a FAPE can emerge from the history of the proceedings. In particular, a court may consider that a plaintiff has previously invoked the IDEA’s formal procedures to handle the dispute—thus starting to exhaust the Act’s remedies before switching midstream. Recall that a parent dissatisfied with her child’s education initiates those administrative procedures by filing a complaint, which triggers a preliminary meeting (or possibly mediation) and then a due process hearing. See supra, at 2–3. A plain-tiff’s initial choice to pursue that process may suggest that she is indeed seeking relief for the denial of a FAPE—with the shift to judicial proceedings prior to full exhaustion reflecting only strategic calculations about how to maximize the prospects of such a remedy. Whether that is so depends on the facts; a court may conclude, for example, that the move to a courtroom came from a late-acquired awareness that the school had fulfilled its FAPE obligation and that the grievance involves something else entirely. But prior pursuit of the IDEA’s administrative reme-dies will often provide strong evidence that the substance of a plaintiff’s claim concerns the denial of a FAPE, even if the complaint never explicitly uses that term.[11] III The Court of Appeals did not undertake the analysis we have just set forward. As noted above, it asked whether E. F.’s injuries were, broadly speaking, “educational” in nature. See supra, at 8; 788 F. 3d, at 627 (reasoning that the “value of allowing Wonder to attend [school] with E. F. was educational” because it would foster “her sense of independence and social confidence,” which is “the sort of interest the IDEA protects”). That is not the same as asking whether the gravamen of E. F.’s complaint charges, and seeks relief for, the denial of a FAPE. And that difference in standard may have led to a difference in result in this case. Understood correctly, §1415(l) might not require exhaustion of the Frys’ claim. We lack some important information on that score, however, and so we remand the issue to the court below. The Frys’ complaint alleges only disability-based discrimination, without making any reference to the ade-quacy of the special education services E. F.’s school provided. The school districts’ “refusal to allow Wonder to act as a service dog,” the complaint states, “discriminated against [E. F.] as a person with disabilities . . . by denying her equal access” to public facilities. App. to Brief in Opposition 15, Complaint ¶68. The complaint contains no allegation about the denial of a FAPE or about any deficiency in E. F.’s IEP. More, it does not accuse the school even in general terms of refusing to provide the educational instruction and services that E. F. needs. See 788 F. 3d, at 631 (acknowledging that the Frys do not “state that Wonder enhances E. F.’s educational opportunities”). As the Frys explained in this Court: The school districts “have said all along that because they gave [E. F.] a one-on-one [human] aide, that all of her . . . educational needs were satisfied. And we have not challenged that, and it would be difficult for us to challenge that.” Tr. of Oral Arg. 16. The Frys instead maintained, just as OCR had earlier found, that the school districts infringed E. F.’s right to equal access—even if their actions complied in full with the IDEA’s requirements. See App. to Brief in Opposition 15, 18–19, Complaint ¶¶ 69, 85, 87; App. 34–37; supra, at 7–8. And nothing in the nature of the Frys’ suit suggests any implicit focus on the adequacy of E. F.’s education. Consider, as suggested above, that the Frys could have filed essentially the same complaint if a public library or theater had refused admittance to Wonder. See supra, at 16. Or similarly, consider that an adult visitor to the school could have leveled much the same charges if prevented from entering with his service dog. See ibid. In each case, the plaintiff would challenge a public facility’s policy of precluding service dogs (just as a blind person might challenge a policy of barring guide dogs, see supra, at 7) as violating Title II’s and §504’s equal access requirements. The suit would have nothing to do with the provision of educational services. From all that we know now, that is exactly the kind of action the Frys have brought. But we do not foreclose the possibility that the history of these proceedings might suggest something different. As earlier discussed, a plaintiff’s initial pursuit of the IDEA’s administrative remedies can serve as evidence that the gravamen of her later suit is the denial of a FAPE, even though that does not appear on the face of her complaint. See supra, at 17–18. The Frys may or may not have sought those remedies before filing this case: None of the parties here have addressed that issue, and the record is cloudy as to the relevant facts. Accordingly, on remand, the court below should establish whether (or to what extent) the Frys invoked the IDEA’s dispute resolution process before bringing this suit. And if the Frys started down that road, the court should decide whether their actions reveal that the gravamen of their complaint is indeed the denial of a FAPE, thus necessitating further exhaustion. With these instructions and for the reasons stated,we vacate the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 At the time (and until 1990), the IDEA was called the Education of the Handicapped Act, or EHA. See §901(a), 104Stat. 1141–1142 (renaming the statute). To avoid confusion—and acronym overload—we refer throughout this opinion only to the IDEA. 2 Because this case comes to us on review of a motion to dismiss E. F.’s suit, we accept as true all facts pleaded in her complaint. See Leatherman v. Tarrant County Narcotics Intelligence and Coordination Unit, 507 U. S. 163, 164 (1993) . 3 See Payne v. Peninsula School Dist., 653 F. 3d 863, 874 (CA9 2011) (en banc) (cataloguing different Circuits’ understandings of §1415(l)). In particular, the Ninth Circuit has criticized an approach similar to the Sixth Circuit’s for “treat[ing] §1415(l) as a quasi-preemption provision, requiring administrative exhaustion for any case that falls within the general ‘field’ of educating disabled students.” Id., at 875. 4 In reaching these conclusions, we leave for another day a further question about the meaning of §1415(l): Is exhaustion required when the plaintiff complains of the denial of a FAPE, but the specific remedy she requests—here, money damages for emotional distress—is not one that an IDEA hearing officer may award? The Frys, along with the Solicitor General, say the answer is no. See Reply Brief 2–3; Brief for United States as Amicus Curiae 16. But resolution of that question might not be needed in this case because the Frys also say that their complaint is not about the denial of a FAPE, see Reply Brief 17—and, as later explained, we must remand that distinct issue to the Sixth Circuit, see infra, at 18–20. Only if that court rejects the Frys’ view of their lawsuit, using the analysis we set out below, will the question about the effect of their request for money damages arise. 5 A case now before this Court, Endrew F. v. Douglas County School Dist. RE–1, No. 15–827, presents unresolved questions about the precise content of the FAPE standard. 6 Without finding the denial of a FAPE, a hearing officer may do nothing more than order a school district to comply with the Act’s various procedural requirements, see §1415(f )(3)(E)(iii)—for example, by allowing parents to “examine all records” relating to their child, §1415(b)(1). 7 Similarly, a court in IDEA litigation may provide a substantive remedy only when it determines that a school has denied a FAPE. See School Comm. of Burlington v. Department of Ed. of Mass., 471 U. S. 359, 369 (1985) . Without such a finding, that kind of relief is (once again) unavailable under the Act. 8 Once again, we do not address here (or anywhere else in this opinion) a case in which a plaintiff, although charging the denial of a FAPE, seeks a form of remedy that an IDEA officer cannot give—for example, as in the Frys’ complaint, money damages for resulting emotional injury. See n. 4, supra. 9 The school districts offer another example illustrating the point. They suppose that a teacher, acting out of animus or frustration, strikes a student with a disability, who then sues the school under a statute other than the IDEA. See Brief for Respondents 36–37. Here too, the suit could be said to relate, in both genesis and effect, to the child’s education. But the school districts opine, we think correctly, that the substance of the plaintiff’s claim is unlikely to involve the adequacy of special education—and thus is unlikely to require exhaustion. See ibid. A telling indicator of that conclusion is that a child could file the same kind of suit against an official at another public facility for inflicting such physical abuse—as could an adult subject to similar treatment by a school official. To be sure, the particular circumstances of such a suit (school or theater? student or employee?) might be pertinent in assessing the reasonableness of the challenged conduct. But even if that is so, the plausibility of bringing other variants of the suit indicates that the gravamen of the plaintiff’s complaint does not concern the appropriateness of an educational program. 10 According to Justice Alito, the hypothetical inquiries described above are useful only if the IDEA and other federal laws are mutually exclusive in scope. See post, at 1 (opinion concurring in part and concurring in judgment). That is incorrect. The point of the questions is not to show that a plaintiff faced with a particular set of circumstances could only have proceeded under Title II or §504—or, alternatively, could only have proceeded under the IDEA. (Depending on the circumstances, she might well have been able to proceed under both.) Rather, these questions help determine whether a plaintiff who has chosen to bring a claim under Title II or §504 instead of the IDEA—and whose complaint makes no mention of a FAPE—nevertheless raises a claim whose substance is the denial of an appropriate education. 11 The point here is limited to commencement of the IDEA’s formal administrative procedures; it does not apply to more informal requests to IEP Team members or other school administrators for accommodations or changes to a special education program. After all, parents of a child with a disability are likely to bring all grievances first to those familiar officials, whether or not they involve the denial of a FAPE.

581.US.2016_15-1406

Respondents Leroy, Donna, Barry, and Suzanne Haeger sued petitioner Goodyear Tire & Rubber Company, alleging that the failure of a Goodyear G159 tire caused the family’s motorhome to swerve off the road and flip over. After several years of contentious discovery, marked by Goodyear’s slow response to repeated requests for internal G159 test results, the parties settled the case. Some months later, the Haegers’ lawyer learned that, in another lawsuit involving the G159, Goodyear had disclosed test results indicating that the tire got unusually hot at highway speeds. In subsequent correspondence, Goodyear conceded withholding the information from the Haegers, even though they had requested all testing data. The Haegers then sought sanctions for discovery fraud, urging that Goodyear’s misconduct entitled them to attorney’s fees and costs expended in the litigation. The District Court found that Goodyear had engaged in an extended course of misconduct. Exercising its inherent power to sanction bad-faith behavior, the court awarded the Haegers $2.7 million—the entire sum they had spent in legal fees and costs since the moment, early in the litigation, when Goodyear made its first dishonest discovery response. The court said that in the usual case, sanctions ordered pursuant to a court’s inherent power to sanction litigation misconduct must be limited to the amount of legal fees caused by that misconduct. But it determined that in cases of particularly egregious behavior, a court can award a party all of the attorney’s fees incurred in a case, without any need to find a “causal link between [the expenses and] the sanctionable conduct.” 906 F. Supp. 2d 938, 975. As further support for its award, the District Court concluded that full and timely disclosure of the test results would likely have led Goodyear to settle the case much earlier. Acknowledging that the Ninth Circuit might require a link between the misconduct and the harm caused, however, the court also made a contingent award of $2 million. That smaller amount, designed to take effect if the Ninth Circuit reversed the larger award, deducted $700,000 in fees the Haegers incurred in developing claims against other defendants and proving their own medical damages. The Ninth Circuit affirmed the full $2.7 million award, concluding that the District Court had properly awarded the Haegers all the fees they incurred during the time when Goodyear was acting in bad faith. Held: When a federal court exercises its inherent authority to sanction bad-faith conduct by ordering a litigant to pay the other side’s legal fees, the award is limited to the fees the innocent party incurred solely because of the misconduct—or put another way, to the fees that party would not have incurred but for the bad faith. Pp. 5–13. (a) Federal courts possess certain inherent powers, including “the ability to fashion an appropriate sanction for conduct which abuses the judicial process.” Chambers v. NASCO, Inc., 501 U. S. 32 –45. One permissible sanction is an assessment of attorney’s fees against a party that acts in bad faith. Such a sanction must be compensatory, rather than punitive, when imposed pursuant to civil procedures. See Mine Workers v. Bagwell, 512 U. S. 821 –830. A sanction counts as compensatory only if it is “calibrate[d] to [the] damages caused by” the bad-faith acts on which it is based. Id., at 834. Hence the need for a court to establish a causal link between the litigant’s misbehavior and legal fees paid by the opposing party. That kind of causal connection is appropriately framed as a but-for test, meaning a court may award only those fees that the innocent party would not have incurred in the absence of litigation misconduct. That standard generally demands that a district court assess and allocate specific litigation expenses—yet still allows it to exercise discretion and judgment. Fox v. Vice, 563 U. S. 826 . And in exceptional cases, that standard allows a court to avoid segregating individual expense items by shifting all of a party’s fees, from either the start or some midpoint of a suit. Pp. 5–9. (b) Here, the parties largely agree about the pertinent law but dispute what it means for this case. Goodyear contends that it requires throwing out the fee award and instructing the trial court to consider the matter anew. The Haegers maintain, to the contrary, that the award can stand because both courts below articulated and applied the appropriate but-for causation standard, or, even if they did not, the fee award in fact passes a but-for test. The Haegers’ defense of the lower courts’ reasoning is a non-starter: Neither court used the correct legal standard. The District Court specifically disclaimed the need for a causal link on the ground that this was a “truly egregious” case. 906 F. Supp. 2d, at 975. And the Ninth Circuit found that the trial court could grant all attorney’s fees incurred “during the time when [Goodyear was] acting in bad faith,” 813 F. 3d 1233, 1249—a temporal, not causal, limitation. A sanctioning court must determine which fees were incurred because of, and solely because of, the misconduct at issue, and no such finding lies behind the $2.7 million award made and affirmed below. Nor is this Court inclined to fill in the gap, as the Haegers urge. As an initial matter, the Haegers have not shown that this litigation would have settled as soon as Goodyear divulged the heat-test results (a showing that would justify an all-fees award from the moment Goodyear was supposed to disclose). Further, they cannot demonstrate that Goodyear’s non-disclosure so permeated the suit as to make that misconduct a but-for cause of every subsequent legal expense, totaling the full $2.7 million. Although the District Court considered causation in arriving at its back-up award of $2 million, it is unclear whether its understanding of that requirement corresponds to the appropriate standard—an uncertainty pointing toward throwing out the fee award and instructing the trial court to consider the matter anew. However, the Haegers contend that Goodyear has waived any ability to challenge the contingent award since the $2 million sum reflects Goodyear’s own submission that only about $700,000 of the fees sought would have been incurred regardless of the company’s behavior. The Court of Appeals did not address that issue, and this Court declines to decide it in the first instance. The possibility of waiver should therefore be the initial order of business on remand. Pp. 9–13. 813 F. 3d 1233, reversed and remanded. Kagan, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case.

In this case, we consider a federal court’s inherent authority to sanction a litigant for bad-faith conduct by ordering it to pay the other side’s legal fees. We hold that such an order is limited to the fees the innocent party incurred solely because of the misconduct—or put another way, to the fees that party would not have incurred but for the bad faith. A district court has broad discretion to calculate fee awards under that standard. But because the court here granted legal fees beyond those resulting from the litigation misconduct, its award cannot stand. I Respondents Leroy, Donna, Barry, and Suzanne Haeger sued the Goodyear Tire & Rubber Company (among other defendants) after the family’s motorhome swerved off the road and flipped over.[1] The Haegers alleged that the failure of a Goodyear G159 tire on the vehicle caused the accident: Their theory was that the tire was not designed to withstand the level of heat it generated when used on a motorhome at highway speeds. Discovery in the case lasted several years—and itself generated considerable heat. The Haegers repeatedly asked Goodyear to turn over internal test results for the G159, but the company’s responses were both slow in coming and unrevealing in content. After making the District Court referee some of their more contentious discovery battles, the parties finally settled the case (for a still-undisclosed sum) on the eve of trial. Some months later, the Haegers’ lawyer learned from a newspaper article that, in another lawsuit involving the G159, Goodyear had disclosed a set of test results he had never seen. That data indicated that the G159 got unusually hot at speeds of between 55 and 65 miles per hour. In ensuing correspondence, Goodyear conceded withholding the information from the Haegers even though they had requested (both early and often) “all testing data” related to the G159. Record in No. 2:05–cv–2046 (D Ariz.), Doc. 938, p. 8; see id., Doc. 938–1, at 24, 36; id., Doc. 1044–2, at 25 (filed under seal). The Haegers accordingly sought sanctions for discovery fraud, claiming that “Goodyear knowingly concealed crucial ‘internal heat test’ records related to the [G159’s] defective design.” Id., Doc. 938, at 1. That conduct, the Haegers urged, entitled them to attorney’s fees and costs expended in the litigation. See id., at 14. The District Court agreed to make such an award in the exercise of its inherent power to sanction litigation misconduct.[2] The court’s assessment of Goodyear’s actions was harsh (and is not contested here). Goodyear, the court found, had engaged in a “years-long course” of bad-faith behavior. 906 F. Supp. 2d 938, 972 (D Ariz. 2012). By withholding the G159’s test results at every turn, the company and its lawyers had made “repeated and deliberate attempts to frustrate the resolution of this case on the merits.” Id., at 971. But because the case had already settled, the court had limited options. It could not take the measure it most wished: an “entry of default judgment” against Goodyear. Id., at 972. All it could do for the Haegers was to order Goodyear to reimburse them for attorney’s fees and costs paid during the suit. But that award, in the District Court’s view, could be comprehensive, covering both expenses that could be causally tied to Goodyear’s misconduct and those that could not. The court calculated that the Haegers had spent $2.7 million in legal fees and costs since the moment, early in the litigation, when Goodyear made its first dishonest discovery response. And the court awarded the Haegers that entire sum. In the “usual[ ]” case, the court reasoned, “sanctions under a [c]ourt’s inherent power must be limited to the amount [of legal fees] caused by the misconduct.” Id., at 974–975 (emphasis deleted). But this case was not the usual one: Here, “the sanctionable conduct r[ose] to a truly egregious level.” Id., at 975. And when a litigant behaves that badly, the court opined, “all of the attorneys’ fees incurred in the case [can] be awarded,” without any need to find a “causal link between [those expenses and] the sanctionable conduct.” Ibid. As further support for its decision, the court considered the chances that full and timely disclosure of the test results would have affected Goodyear’s settlement calculus. “While there is some uncertainty,” the court stated, “the case more likely than not would have settled much earlier.” Id., at 972. Perhaps sensing thin ice, the District Court also made a “contingent award” in the event that the Court of Appeals reversed its preferred one. App. to Pet. for Cert. 180a. Here, the District Court recognized the possibility that a “linkage between [Goodyear’s] misconduct and [the Haegers’] harm is required.” Ibid. If so, the court stated, its fee award should be reduced to $2 million. The deduction of $700,000, which was based on estimates Goodyear offered, represented fees that the Haegers incurred in developing claims against other defendants and proving their own medical damages. See App. 69. A divided Ninth Circuit panel affirmed the full $2.7 million award. According to the majority, the District Court acted properly in “award[ing] the amount [it] reasonably believed” the Haegers expended in attorney’s fees and costs “during the time when [Goodyear was] acting in bad faith.” 813 F. 3d 1233, 1250 (2016). Or repeated in just slightly different words: The District Court “did not abuse its discretion” in “award[ing] the Haegers all their attorneys’ fees and costs in prosecuting the action once [Goodyear] began flouting [its] discovery obligations.” Id., at 1249. Judge Watford disagreed. He would have demanded a “causal link between Goodyear’s misconduct and the fees awarded.” Id., at 1255 (dissenting opinion). The only part of the District Court’s opinion that might support such a connection, Judge Watford noted, was its hypothesis that disclosure of the test results would have produced an earlier settlement, and thus obviated the need for further legal expenses. But Judge Watford thought that theory unpersuasive: Because Goodyear would still have had plausible defenses to the Haegers’ suit, “[i]t’s anyone’s guess how the litigation would have proceeded” had timely disclosure occurred. Ibid. Accordingly, Judge Watford would have reversed the District Court for awarding fees beyond those “sustained as a result of Goodyear’s misconduct.” Id., at 1256. The Court of Appeals’ decision created a split of authority: Other Circuits have insisted on limiting sanctions like this one to fees or costs that are causally related to a litigant’s misconduct.[3] We therefore granted certiorari. 579 U. S. ___ (2016). II Federal courts possess certain “inherent powers,” not conferred by rule or statute, “to manage their own affairs so as to achieve the orderly and expeditious disposition of cases.” Link v. Wabash R. Co., 370 U. S. 626 –631 (1962). That authority includes “the ability to fashion an appropriate sanction for conduct which abuses the judicial process.” Chambers v. NASCO, Inc., 501 U. S. 32 –45 (1991). And one permissible sanction is an “assessment of attorney’s fees”—an order, like the one issued here, instructing a party that has acted in bad faith to reimburse legal fees and costs incurred by the other side. Id., at 45. This Court has made clear that such a sanction, when imposed pursuant to civil procedures, must be compensatory rather than punitive in nature. See Mine Workers v. Bagwell, 512 U. S. 821 –830 (1994) (distinguishing compensatory from punitive sanctions and specifying the procedures needed to impose each kind).[4] In other words, the fee award may go no further than to redress the wronged party “for losses sustained”; it may not impose an additional amount as punishment for the sanctioned party’s misbehavior. Id., at 829 (quoting United States v. Mine Workers, 330 U. S. 258, 304 (1947) ). To level that kind of separate penalty, a court would need to provide procedural guarantees applicable in criminal cases, such as a “beyond a reasonable doubt” standard of proof. See id., at 826, 832–834, 838–839. When (as in this case) those criminal-type protections are missing, a court’s shifting of fees is limited to reimbursing the victim. That means, pretty much by definition, that the court can shift only those attorney’s fees incurred because of the misconduct at issue. Compensation for a wrong, after all, tracks the loss resulting from that wrong. So as we have previously noted, a sanction counts as compensatory only if it is “calibrate[d] to [the] damages caused by” the bad-faith acts on which it is based. Id., at 834. A fee award is so calibrated if it covers the legal bills that the litigation abuse occasioned. But if an award extends further than that—to fees that would have been incurred without the misconduct—then it crosses the boundary from compensation to punishment. Hence the need for a court, when using its inherent sanctioning authority (and civil procedures), to establish a causal link—between the litigant’s misbehavior and legal fees paid by the opposing party.[5] That kind of causal connection, as this Court explained in another attorney’s fees case, is appropriately framed as a but-for test: The complaining party (here, the Haegers) may recover “only the portion of his fees that he would not have paid but for” the misconduct. Fox v. Vice, 563 U. S. 826, 836 (2011) ; see Paroline v. United States, 572 U. S. ___, ___ (2014) (slip op., at 12) (“The traditional way to prove that one event was a factual cause of another is to show that the latter would not have occurred ‘but for’ the former”). In Fox, a prevailing defendant sought reimbursement under a fee-shifting statute for legal expenses incurred in defending against several frivolous claims. See 563 U. S., at 830; 42 U. S. C. §1988. The trial court granted fees for all legal work relating to those claims—regardless of whether the same work would have been done (for example, the same depositions taken) to contest the non-frivolous claims in the suit. We made clear that was wrong. When a “defendant would have incurred [an] expense in any event[,] he has suffered no incremental harm from the frivolous claim,” and so the court lacks a basis for shifting the expense. Fox, 563 U. S., at 836. Substitute “discovery abuse” for “frivolous claim” in that sentence, and the same thing goes in this case. Or otherwise said (and again borrowing from Fox), when “the cost[ ] would have been incurred in the absence of” the discovery violation, then the court (possessing only the power to compensate for harm the misconduct has caused) must leave it alone. Id., at 838. This but-for causation standard generally demands that a district court assess and allocate specific litigation expenses—yet still allows it to exercise discretion and judgment. The court’s fundamental job is to determine whether a given legal fee—say, for taking a deposition or drafting a motion—would or would not have been incurred in the absence of the sanctioned conduct. The award is then the sum total of the fees that, except for the misbehavior, would not have accrued. See id., at 837–838 (providing illustrative examples). But as we stressed in Fox, trial courts undertaking that task “need not, and indeed should not, become green-eyeshade accountants” (or whatever the contemporary equivalent is). Id., at 838. “The essential goal” in shifting fees is “to do rough justice, not to achieve auditing perfection.” Ibid. Accordingly, a district court “may take into account [its] overall sense of a suit, and may use estimates in calculating and allocating an attorney’s time.” Ibid. The court may decide, for example,that all (or a set percentage) of a particular category of expenses—say, for expert discovery—were incurred solely because of a litigant’s bad-faith conduct. And such judgments, in light of the trial court’s “superior understanding of the litigation,” are entitled to substantial deference on appeal. Hensley v. Eckerhart, 461 U. S. 424, 437 (1983) . In exceptional cases, the but-for standard even permits a trial court to shift all of a party’s fees, from either the start or some midpoint of a suit, in one fell swoop. Chambers v. NASCO offers one illustration. There, we approved such an award because literally everything the defendant did—“his entire course of conduct” throughout, and indeed preceding, the litigation—was “part of a sordid scheme” to defeat a valid claim. 501 U. S., at 51, 57 (brackets omitted). Thus, the district court could reasonably conclude that all legal expenses in the suit “were caused . . . solely by [his] fraudulent and brazenly unethical efforts.” Id., at 58. Or to flip the example: If a plaintiff initiates a case in complete bad faith, so that every cost of defense is attributable only to sanctioned behavior, the court may again make a blanket award. And similarly, if a court finds that a lawsuit, absent litigation misconduct, would have settled at a specific time—for example, when a party was legally required to disclose evidence fatal to its position—then the court may grant all fees incurred from that moment on. In each of those scenarios, a court escapes the grind of segregating individual expense items (a deposition here, a motion there)—or even categories of such items (again, like expert discovery)—but only because all fees in the litigation, or a phase of it, meet the applicable test: They would not have been incurred except for the misconduct. III It is an oddity of this case that both sides agree with just about everything said in the last six paragraphs about the pertinent law. Do legal fees awarded under a court’s inherent sanctioning authority have to be compensatory rather than punitive when civil litigation procedures are used? The Haegers and Goodyear alike say yes. Does that mean the fees awarded must be causally related to the sanctioned party’s misconduct? A joint yes on that too. More specifically, does the appropriate causal test limit the fees, a la Fox, to those that would not have been incurred but for the bad faith? No argument there either. And in an exceptional case, such as Chambers, could that test produce an award extending as far as all of the wronged party’s legal fees? Once again, agreement (if with differing degrees of enthusiasm). See Brief for Petitioner 17, 23–24, 31; Brief for Respondents 17–18, 22–23; Tr. of Oral Arg. 34–35, 46–47. All the parties really argue about here is what that law means for this case. Goodyear contends that it requires throwing out the trial court’s fee award and instructing the court to consider the matter anew. The Haegers maintain, to the contrary, that the award can stand. They initially contend—pointing to a couple of passages from the Ninth Circuit’s opinion—that both courts below articulated and applied the very but-for causation standard we have laid out. See Brief for Respondents 17–18 (highlighting the Ninth Circuit’s statements that Goodyear’s “bad faith conduct caused significant harm” and that the District Court “determine[d] the appropriate amount of fees to award as sanctions to compensate the [Haegers] for the damages they suffered as a result of [Goodyear’s] bad faith”). And even if we reject that view, the Haegers continue, we may uphold the fee award on the ground that it in fact passes a but-for test. That standard is satisfied (so they say) for either of two reasons. First, because the case would have settled as soon as Goodyear disclosed the requested heat-test results, thus putting an end to the Haegers’ legal bills. Or second, because (settlement prospects aside) the withholding of that data so infected the lawsuit as to account for each and every expense the Haegers subsequently incurred. See id., at 14–15, 22, 26. The Haegers’ defense of the lower courts’ reasoning is a non-starter: Neither of them used the correct legal standard. As earlier recounted, the District Court specifically disclaimed the “usual[ ]” need to find a “causal link” between misconduct and fees when the sanctioned party’s behavior was bad enough—in the court’s words, when it “r[ose] to a truly egregious level.” 906 F. Supp. 2d, at 975 (emphasis deleted); see supra, at 3. In such circumstances, the court thought, it could award “all” fees, including those that would have been incurred in the absence of the misconduct. 906 F. Supp. 2d, at 975. And the court confirmed that approach even while conceding that it might be wrong: By issuing a “contingent award” of $2 million, meant to go into effect if the Ninth Circuit demanded a causal “linkage between the misconduct and harm,” the District Court made clear that its primary, $2.7 million award was not so confined. App. to Pet. for Cert. 180a; see supra, at 4. Still, the Court of Appeals left the larger sanction in place, because it too mistook what findings were needed to support that award. In the Ninth Circuit’s view, the trial court could grant all attorney’s fees incurred “during the time when [Goodyear was] acting in bad faith.” 813 F. 3d, at 1250 (emphasis added); see id., at 1249 (permitting an award of fees incurred “once [Goodyear] began flouting [its] discovery obligations” (emphasis added)); supra, at 4. But that is a temporal limitation, not a causal one; and, like the District Court’s “egregiousness” requirement, it is wide of the mark. A sanctioning court must determine which fees were incurred because of, and solely because of, the misconduct at issue (however serious, or concurrent with a lawyer’s work, it might have been). No such finding lies behind the $2.7 million award made and affirmed below. Nor are we tempted to fill in that gap, as the Haegers have invited us to do. As an initial matter, the Haegers have not shown that this litigation would have settled as soon as Goodyear divulged the heat-test results (thus justifying an all-fees award from the moment it was supposed to disclose, see supra, at 8–9). Even the District Court did not go quite that far: In attempting to buttress its comprehensive award, it said only (and after expressing “some uncertainty”) that the suit probably would have settled “much earlier.” 906 F. Supp. 2d, at 972. And that more limited finding is itself subject to grave doubt, even taking into account the deference owed to the trial court. As Judge Watford reasoned, the test results, although favorable to the Haegers’ version of events, did not deprive Goodyear of colorable defenses. In particular, Goodyear still could have argued, as it had from the beginning, that “the Haegers’ own tire, which had endured more than 40,000 miles of wear and tear, failed because it struck road debris.” 813 F. 3d, at 1256 (dissenting opinion). And indeed, that is pretty much the course Goodyear took in another suit alleging that the G159 caused a motorhome accident. See Schalmo v. Goodyear, No. 51–2006–CA–2064–WS (Fla. Cir. Ct., 6th Cir., Pasco County). In that case (as Judge Watford again observed), Goodyear produced the very test results at issue here, yet still elected to go to trial. See 813 F. 3d, at 1256. So we do not think the record allows a finding, as would support the $2.7 million award, that disclosure of the heat-test results would have led straightaway to a settlement. Further, the Haegers cannot demonstrate that Goodyear’s non-disclosure so permeated the suit as to make that misconduct a but-for cause of every subsequent legal expense, totaling the full $2.7 million. If nothing else, the District Court’s back-up fee award belies that theory. After introducing a causal element into the equation, the court found that the $700,000 of fees that the Haegers incurred in litigating against other defendants and proving their own medical damages had nothing to do with Goodyear’s discovery decisions. See App. to Pet. for Cert. 180a; supra, at 4. The Haegers have failed to offerany concrete reason for questioning that judgment, and we do not see how they could. At a minimum, then, the sanction order could not force Goodyear to reimburse those expenses—because, again, the Haegers would have paid them even had the company behaved immaculately in every respect. That leaves the question whether the contingent $2 million award should now stand—or, alternatively, whether the District Court must reconsider from scratch which fees to shift. In the absence of any waiver issue, we would insist on the latter course. Although the District Court considered causation in arriving at its back-up award, we cannot tell from its sparse discussion whether its understanding of that requirement corresponds to the standard we have described. That uncertainty points toward demanding a do-over, under the unequivocally right legal rules. But the Haegers contend that Goodyear has waived any ability to challenge the $2 million award. In their view, that sum reflected Goodyear’s own submission—which it may not now amend—that only about $700,000 of the fees sought would have been incurred “regardless of Goodyear’s behavior.” App. 69; see Brief for Respondents 41; supra, at 4. The Court of Appeals did not previously address that issue, and we decline to decide it in the first instance. See Cutter v. Wilkinson, 544 U. S. 709, 718, n. 7 (2005) (“[W]e are a court of review, not of first view”). The possibility of waiver should therefore be the initial order of business below. If a waiver is found, that is the end of this case. If not, the District Court must reassess fees in line with a but-for causation requirement. For these reasons, we reverse the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 The additional defendants named in the Haegers’ complaint were Gulf Stream Coach, the manufacturer of the motorhome, and Spartan Motors, the manufacturer of the vehicle’s chassis. In the course of the litigation, the Haegers reached a settlement with Gulf Stream, and the District Court granted Spartan’s motion for summary judgment. 2 The court reasoned that no statute or rule enabled it to reach all the offending behavior. Sanctions under Federal Rule of Civil Procedure 11, the court thought, should not be imposed after final judgment in a case. See 906 F. Supp. 2d 938, 973, n. 24 (D Ariz. 2012). And sanctions under 28 U. S. C. §1927, it noted, could address the wrongdoing of only Goodyear’s attorneys, rather than of Goodyear itself. See 906 F. Supp. 2d, at 973. 3 See, e.g., Plaintiffs’ Baycol Steering Comm. v. Bayer Corp., 419 F. 3d 794, 808 (CA8 2005); Bradley v. American Household, Inc., 378 F. 3d 373, 378 (CA4 2004); United States v. Dowell, 257 F. 3d 694, 699 (CA7 2001). 4 Bagwell also addressed “coercive” sanctions, designed to make a party comply with a court order. 512 U. S., at 829. That kind of sanction is not at issue here. 5 Rule-based and statutory sanction regimes similarly require courts to find such a causal connection before shifting fees. For example, the Federal Rules of Civil Procedure provide that a district court may order a party to pay attorney’s fees “caused by” discovery misconduct, Rule 37(b)(2)(C), or “directly resulting from” misrepresentations in pleadings, motions, and other papers, Rule 11(c)(4). And under 28 U. S. C. §1927, a court may require an attorney who unreasonably multiplies proceedings to pay attorney’s fees incurred “because of” that misconduct. Those provisions confirm the need to establish a causal link between misconduct and fees when acting under inherent authority, given that such undelegated powers should be exercised with especial “restraint and discretion.” Roadway Express, Inc. v. Piper, 447 U. S. 752, 764 (1980) .

582.US.2016_16-349

The Fair Debt Collection Practices Act authorizes private lawsuits and weighty fines designed to deter the wayward practices of “debt collector[s],” a term embracing anyone who “regularly collects or attempts to collect . . . debts owed or due . . . another.” 15 U. S. C. §1692a(6). The complaint filed in this case alleges that CitiFinancial Auto loaned money to petitioners seeking to buy cars; that petitioners defaulted on those loans; and that respondent Santander then purchased the defaulted loans from CitiFinancial and sought to collect in ways petitioners believe violated the Act. The district court and Fourth Circuit held that Santander didn’t qualify as a debt collector because it did not regularly seek to collect debts “owed . . . another” but sought instead only to collect debts that it purchased and owned. Held: A company may collect debts that it purchased for its own account, like Santander did here, without triggering the statutory definition in dispute. By defining debt collectors to include those who regularly seek to collect debts “owed . . . another,” the statute’s plain language seems to focus on third party collection agents regularly collecting for a debt owner—not on a debt owner seeking to collect debts for itself. Petitioners’ arguments to the contrary do not dislodge the statute’s plain meaning. Petitioners point out that the word “owed” is the past participle of the verb “to owe,” and so suggest that the debt collector definition must exclude loan originators (who never seek to collect debts previously owed someone else) but embrace debt purchasers like Santander (who necessarily do). But past participles like “owed” are routinely used as adjectives to describe the present state of a thing. Congress also used the word “owed” to refer to present debt relationships in neighboring provisions of the Act, and petitioners have not rebutted the presumption that identical words in the same statute carry the same meaning. Neither would reading the word “owed” to refer to present debt relationships render any of the Act’s provisions surplusage, contrary to what petitioners suggest. Petitioners also contend that their interpretation best furthers the Act’s perceived purposes because, they primarily argue, if Congress had been aware of defaulted debt purchasers like Santander it would have treated them like traditional debt collectors because they pose similar risks of abusive collection practices. But it is not this Court’s job to rewrite a constitutionally valid text under the banner of speculation about what Congress might have done had it faced a question that, on everyone’s account, it never faced. And neither are petitioners’ policy arguments unassailable, as reasonable legislators might contend both ways on the question of how defaulted debt purchasers should be treated. This fact suggests for certain but one thing: that these are matters for Congress, not this Court, to resolve. Pp. 3–11. 817 F. 3d 131, affirmed. Gorsuch, J., delivered the opinion for a unanimous Court.

Disruptive dinnertime calls, downright deceit, and more besides drew Congress’s eye to the debt collection industry. From that scrutiny emerged the Fair Debt Collection Practices Act, a statute that authorizes private lawsuits and weighty fines designed to deter wayward collection practices. So perhaps it comes as little surprise that we now face a question about who exactly qualifies as a “debt collector” subject to the Act’s rigors. Everyone agrees that the term embraces the repo man—someone hired by a creditor to collect an outstanding debt. But what if you purchase a debt and then try to collect it for yourself—does that make you a “debt collector” too? That’s the nub of the dispute now before us. The parties approach the question from common ground. The complaint alleges that CitiFinancial Auto loaned money to petitioners seeking to buy cars; that petitioners defaulted on those loans; that respondent Santander then purchased the defaulted loans from CitiFinancial; and that Santander sought to collect in ways petitioners believe troublesome under the Act. The parties agree, too, that in deciding whether Santander’s conduct falls within the Act’s ambit we should look to statutory language defining the term “debt collector” to embrace anyone who “regularly collects or attempts to collect . . . debts owed or due . . . another.” 15 U. S. C. §1692a(6). Even when it comes to that question, the parties agree on at least part of an answer. Both sides accept that third party debt collection agents generally qualify as “debt collectors” under the relevant statutory language, while those who seek only to collect for themselves loans they originated generally do not. These results follow, the parties tell us, because debt collection agents seek to collect debts “owed . . . another,” while loan originators acting on their own account aim only to collect debts owed to themselves. All that remains in dispute is how to classify individuals and entities who regularly purchase debts originated by someone else and then seek to collect those debts for their own account. Does the Act treat the debt purchaser in that scenario more like the repo man or the loan originator? For their part, the district court and Fourth Circuit sided with Santander. They held that the company didn’t qualify as a debt collector because it didn’t regularly seek to collect debts “owed . . . another” but sought instead only to collect debts that it purchased and owned. At the same time, the Fourth Circuit acknowledged that some circuits faced with the same question have ruled otherwise—and it is to resolve this conflict that we took the case. Compare 817 F. 3d 131, 133–134, 137–138 (2016) (case below); Davidson v. Capital One Bank (USA), N. A., 797 F. 3d 1309, 1315–1316 (CA11 2015), with McKinney v. Caldeway Properties, Inc., 548 F. 3d 496, 501 (CA7 2008); FTC v. Check Investors, Inc., 502 F. 3d 159, 173–174 (CA3 2007). Before attending to that job, though, we pause to note two related questions we do not attempt to answer today. First, petitioners suggest that Santander can qualify as a debt collector not only because it regularly seeks to collect for its own account debts that it has purchased, but also because it regularly acts as a third party collection agent for debts owed to others. Petitioners did not, however, raise the latter theory in their petition for certiorari and neither did we agree to review it. Second, the parties briefly allude to another statutory definition of the term “debt collector”—one that encompasses those engaged “in any business the principal purpose of which is the collection of any debts.” §1692a(6). But the parties haven’t much litigated that alternative definition and in granting certiorari we didn’t agree to address it either. With these preliminaries by the board, we can turn to the much narrowed question properly before us. In doing so, we begin, as we must, with a careful examination of the statutory text. And there we find it hard to disagree with the Fourth Circuit’s interpretive handiwork. After all, the Act defines debt collectors to include those who regularly seek to collect debts “owed . . . another.” And by its plain terms this language seems to focus our attention on third party collection agents working for a debt owner—not on a debt owner seeking to collect debts for itself. Neither does this language appear to suggest that we should care how a debt owner came to be a debt owner—whether the owner originated the debt or came by it only through a later purchase. All that matters is whether the target of the lawsuit regularly seeks to collect debts for its own account or does so for “another.” And given that, it would seem a debt purchaser like Santander may indeed collect debts for its own account without triggering the statutory definition in dispute, just as the Fourth Circuit explained. Petitioners reply that this seemingly straightforward reading overlooks an important question of tense. They observe that the word “owed” is the past participle of the verb “to owe.” And this, they suggest, means the statute’s definition of debt collector captures anyone who regularly seeks to collect debts previously “owed . . . another.” So it is that, on petitioners’ account, the statute excludes from its compass loan originators (for they never seek to collect debts previously owed someone else) but embraces many debt purchasers like Santander (for in collecting purchased debts they necessarily seek to collect debts previously owed another). If Congress wanted to exempt all present debt owners from its debt collector definition, petitioners submit, it would have used the present participle “owing.” That would have better sufficed to do the job—to make clear that you must collect debts currently “owing . . . another” before implicating the Act. But this much doesn’t follow even as a matter of good grammar, let alone ordinary meaning. Past participles like “owed” are routinely used as adjectives to describe the present state of a thing—so, for example, burnt toast is inedible, a fallen branch blocks the path, and (equally) a debt owed to a current owner may be collected by him or her. See P. Peters, The Cambridge Guide to English Usage 409 (2004) (explaining that the term “past participle” is a “misnomer[ ], since” it “can occur in what is technically a present . . . tense”). Just imagine if you told a friend that you were seeking to “collect a debt owed to Steve.” Doesn’t it seem likely your friend would understand you as speaking about a debt currently owed to Steve, not a debt Steve used to own and that’s now actually yours? In the end, even petitioners find themselves forced to admit that past participles can and regularly do work just this way, as adjectives to describe the present state of the nouns they modify. See Brief for Petitioners 28; see also B. Garner, Modern English Usage 666 (4th ed. 2016) (while “owing . . . is an old and established usage . . . the more logical course is simply to write owed”). Widening our view to take in the statutory phrase in which the word “owed” appears—“owed or due . . . another”—serves to underscore the point. Petitioners acknowledge that the word “due” describes a debt currently due at the time of collection and not a debt that was due only in some previous period. Brief for Petitioners 26–28. So to rule for them we would have to suppose Congress set two words cheek by jowl in the same phrase but meant them to speak to entirely different periods of time. All without leaving any clue. We would have to read the phrase not as referring to “debts that are owed or due another” but as describing “debts that were owed or are due another.” And supposing such a surreptitious subphrasal shift in time seems to us a bit much. Neither are we alone in that assessment, for even petitioners acknowledge that theirs “may not be the most natural interpretation of the phrase standing in isolation.” Id., at 26–27. Given that, you might wonder whether extending our gaze from the narrow statutory provision at issue to take in the larger statutory landscape might offer petitioners a better perspective. But it does not. Looking to other neighboring provisions in the Act, it quickly comes clear that Congress routinely used the word “owed” to refer to present (not past) debt relationships. For example, in one nearby subsection, Congress defined a creditor as someone “to whom a debt is owed.” 15 U. S. C. §1692a(4). In another subsection, too, Congress required a debt collector to identify “the creditor to whom the debt is owed.” §1692g(a)(2). Yet petitioners offer us no persuasive reason why the word “owed” should bear a different meaning here, in the subsection before us, or why we should abandon our usual presumption that “identical words used in different parts of the same statute” carry “the same meaning.” IBP, Inc. v. Alvarez, 546 U. S. 21, 34 (2005) . Still other contextual clues add to petitioners’ problems. While they suggest that the statutory definition before us implicitly distinguishes between loan originators and debt purchasers, a pass through the statute shows that when Congress wished to distinguish between originators and purchasers it left little doubt in the matter. In the very definitional section where we now find ourselves working, Congress expressly differentiated between a person “who offers” credit (the originator) and a person “to whom a debt is owed” (the present debt owner). §1692a(4). Elsewhere, Congress recognized the distinction between a debt “originated by” the collector and a debt “owed or due” another. §1692a(6)(F)(ii). And elsewhere still, Congress drew a line between the “original” and “current” creditor. §1692g(a)(5). Yet no similar distinction can be found in the language now before us. To the contrary, the statutory text at issue speaks not at all about originators and current debt owners but only about whether the defendant seeks to collect on behalf of itself or “another.” And, usually at least, when we’re engaged in the business of interpreting statutes we presume differences in language like this convey differences in meaning. See, e.g., Loughrin v. United States, 573 U. S. ___, ___ (2014). Even what may be petitioners’ best piece of contextual evidence ultimately proves unhelpful to their cause. Petitioners point out that the Act exempts from the definition of “debt collector” certain individuals who have “obtained” particular kinds of debt—for example, debts not yet in default or debts connected to secured commercial credit transactions. §§1692a(6)(F)(iii) and (F)(iv). And because these exemptions contemplate the possibility that someone might “obtain” a debt “owed or due . . . another,” petitioners submit, the word “owed” must refer only to a previous owner. Ibid. This conclusion, they say, necessarily follows because, once you have “obtained” a debt, that same debt just cannot be currently “owed or due” another. This last and quite essential premise of the argument, however, misses its mark. As a matter of ordinary English, the word “obtained” can (and often does) refer to taking possession of a piece of property without also taking ownership—so, for example, you might obtain a rental car or a hotel room or an apartment. See, e.g., 10 Oxford English Dictionary 669 (2d ed. 1989) (defining “obtain” to mean, among other things, “[t]o come into the possession or enjoyment of (something) by one’s own effort or by request”); Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. 519 –533 (2013) (distinguishing between ownership and obtaining possession). And it’s easy enough to see how you might also come to possess (obtain) a debt without taking ownership of it. You might, for example, take possession of a debt for servicing and collection even while the debt formally remains owed another. Or as a secured party you might take possession of a debt as collateral, again without taking full ownership of it. See, e.g., U. C. C. §9–207, 3 U. L. A. 197 (2010). So it simply isn’t the case that the statute’s exclusions imply that the phrase “owed . . . another” must refer to debts previously owed to another. By this point petitioners find themselves in retreat. Unable to show that debt purchasers regularly collecting for their own account always qualify as debt collectors, they now suggest that purchasers sometimes qualify as debt collectors. On their view, debt purchasers surely qualify as collectors at least when they regularly purchase and seek to collect defaulted debts—just as Santander allegedly did here. In support of this narrower and more particular understanding of the Act, petitioners point again to the fact that the statute excludes from the definition of “debt collector” certain persons who obtain debts before default. 15 U. S. C. §1692a(6)(F)(iii). This exclusion, petitioners now suggest, implies that the term “debt collector” must embrace those who regularly seek to collect debts obtained after default. Others aligned with petitioners also suggest that the Act treats everyone who attempts to collect a debt as either a “debt collector” or a “creditor,” but not both. And because the statutory definition of the term “creditor” excludes those who seek to collect a debt obtained “in default,” §1692a(4), they contend it again follows as a matter of necessary inference that these persons must qualify as debt collectors. But these alternative lines of inferential argument bear their own problems. For while the statute surely excludes from the debt collector definition certain persons who acquire a debt before default, it doesn’t necessarily follow that the definition must include anyone who regularly collects debts acquired after default. After all and again, under the definition at issue before us you have to attempt to collect debts owed another before you can ever qualify as a debt collector. And petitioners’ argument simply does not fully confront this plain and implacable textual prerequisite. Likewise, even spotting (without granting) the premise that a person cannot be both a creditor and a debt collector with respect to a particular debt, we don’t see why a defaulted debt purchaser like Santander couldn’t qualify as a creditor. For while the creditor definition excludes persons who “receive an assignment or transfer of a debt in default,” it does so only (and yet again) when the debt is assigned or transferred “solely for the purpose of facilitating collection of such debt for another.” Ibid. (emphasis added). So a company collecting purchased defaulted debt for its own account—like Santander—would hardly seem to be barred from qualifying as a creditor under the statute’s plain terms. Faced with so many obstacles in the text and structure of the Act, petitioners ask us to move quickly on to policy. Indeed, from the beginning that is the field on which they seem most eager to pitch battle. Petitioners assert that Congress passed the Act in large measure to add new incentives for independent debt collectors to treat consumers well. In their view, Congress excluded loan originators from the Act’s demands because it thought they already faced sufficient economic and legal incentives to good behavior. But, on petitioners’ account, Congress never had the chance to consider what should be done about those in the business of purchasing defaulted debt. That’s because, petitioners tell us, the “advent” of the market for defaulted debt represents “ ‘one of the most significant changes’ ” to the debt market generally since the Act’s passage in 1977. Brief for Petitioners 8 (quoting Consumer Financial Protection Bureau, Fair Debt Collection Practices Act: CFPB Annual Report 2014, p. 7 (2014)). Had Congress known this new industry would blossom, they say, it surely would have judged defaulted debt purchasers more like (and in need of the same special rules as) independent debt collectors. Indeed, petitioners contend that no other result would be consistent with the overarching congressional goal of deterring untoward debt collection practices. All this seems to us quite a lot of speculation. And while it is of course our job to apply faithfully the law Congress has written, it is never our job to rewrite a constitutionally valid statutory text under the banner of speculation about what Congress might have done had it faced a question that, on everyone’s account, it never faced. See Magwood v. Patterson, 561 U. S. 320, 334 (2010) (“We cannot replace the actual text with speculation as to Congress’ intent”). Indeed, it is quite mistaken to assume, as petitioners would have us, that “whatever” might appear to “further[ ] the statute’s primary objective must be the law.” Rodriguez v. United States, 480 U. S. 522, 526 (1987) (per curiam) (emphasis deleted). Legislation is, after all, the art of compromise, the limitations expressed in statutory terms often the price of passage, and no statute yet known “pursues its [stated] purpose[ ] at all costs.” Id., at 525–526. For these reasons and more besides we will not presume with petitioners that any result consistent with their account of the statute’s overarching goal must be the law but will presume more modestly instead “that [the] legislature says . . . what it means and means . . . what it says.” Dodd v. United States, 545 U. S. 353, 357 (2005) (internal quotation marks omitted; brackets in original). Even taken on its own terms, too, the speculation petitioners urge upon us is far from unassailable. After all, is it really impossible to imagine that reasonable legislators might contend both ways on the question whether defaulted debt purchasers should be treated more like loan originators than independent debt collection agencies? About whether other existing incentives (in the form of common law duties, other statutory and regulatory obligations, economic incentives, or otherwise) suffice to deter debt purchasers from engaging in certain undesirable collection activities? Couldn’t a reasonable legislator endorsing the Act as written wonder whether a large financial institution like Santander is any more or less likely to engage in abusive conduct than another large financial institution like CitiFinancial Auto? Especially where (as here) the institution says that its primary business is loan origination and not the purchase of defaulted debt? We do not profess sure answers to any of these questions, but observe only that the parties and their amici manage to present many and colorable arguments both ways on them all, a fact that suggests to us for certain but one thing: that these are matters for Congress, not this Court, to resolve. In the end, reasonable people can disagree with how Congress balanced the various social costs and benefits in this area. We have no difficulty imagining, for example, a statute that applies the Act’s demands to anyone collecting any debts, anyone collecting debts originated by another, or to some other class of persons still. Neither do we doubt that the evolution of the debt collection business might invite reasonable disagreements on whether Congress should reenter the field and alter the judgments it made in the past. After all, it’s hardly unknown for new business models to emerge in response to regulation, and for regulation in turn to address new business models. Constant competition between constable and quarry, regulator and regulated, can come as no surprise in our changing world. But neither should the proper role of the judiciary in that process—to apply, not amend, the work of the People’s representatives. The judgment of the Court of Appeals is Affirmed.

581.US.2016_16-142

Terry Honeycutt managed sales and inventory for a Tennessee hardware store owned by his brother, Tony Honeycutt. After they were indicted for federal drug crimes including conspiracy to distribute a product used in methamphetamine production, the Government sought judgments against each brother in the amount of $269,751.98 pursuant to the Comprehensive Forfeiture Act of 1984, which mandates forfeiture of “any property constituting, or derived from, any proceeds the person obtained, directly or indirectly, as the result of” certain drug crimes, 21 U. S. C. §853(a)(1). Tony pleaded guilty and agreed to forfeit $200,000. Terry went to trial and was convicted. Despite conceding that Terry had no controlling interest in the store and did not stand to benefit personally from the sales of the product, the Government asked the District Court to hold him jointly and severally liable for the profits from the illegal sales and sought a judgment of $69,751.98, the outstanding conspiracy profits. The District Court declined to enter a forfeiture judgment against Terry, reasoning that he was a salaried employee who had not received any profits from the sales. The Sixth Circuit reversed, holding that the brothers, as co-conspirators, were jointly and severally liable for any conspiracy proceeds. Held: Because forfeiture pursuant to §853(a)(1) is limited to property the defendant himself actually acquired as the result of the crime, that provision does not permit forfeiture with regard to Terry Honeycutt, who had no ownership interest in his brother’s store and did not personally benefit from the illegal sales. Pp. 3–11. (a) Section 853(a) limits forfeiture to property flowing from, §853(a)(1), or used in, §853(a)(2), the crime itself—providing the first clue that the statute does not countenance joint and several liability, which would require forfeiture of untainted property. It also defines forfeitable property solely in terms of personal possession or use. Section 853(a)(1), the provision at issue, limits forfeiture to property the defendant “obtained, directly or indirectly, as the result of” the crime. Neither the dictionary definition nor the common usage of the word “obtain” supports the conclusion that an individual “obtains” property that was acquired by someone else. And the adverbs “directly” and “indirectly” refer to how a defendant obtains the property; they do not negate the requirement that he obtain it at all. Sections 853(a)(2) and 853(a)(3) are in accord with this reading. Pp. 3–7. (b) Joint and several liability is also contrary to several other provisions of §853. Section 853(c), which applies to property “described in subsection (a),” applies to tainted property only. See Luis v. United States, 578 U. S. ___, ___. Section §853(e)(1) permits pretrial asset freezes to preserve the availability of property forfeitable under subsection (a), provided there is probable cause to think that a defendant has committed an offense triggering forfeiture and “the property at issue has the requisite connection to that crime.” Kaley v. United States, 571 U. S. ___, ___. Section 853(d) establishes a “rebuttable presumption” that property is subject to forfeiture only if the Government proves that the defendant acquired the property “during the period of the violation” and “there was no likely source for” the property but the crime. These provisions reinforce the statute’s application to tainted property acquired by the defendant and are thus incompatible with joint and several liability. Joint and several liability would also render futile §853(p)—the sole provision of §853 that permits the Government to confiscate property untainted by the crime. Pp. 7–9. (c) The plain text and structure of §853 leave no doubt that Congress did not, as the Government claims, incorporate the principle that conspirators are legally responsible for each other’s foreseeable actions in furtherance of their common plan. See Pinkerton v. United States, 328 U. S. 640 . Congress provided just one way for the Government to recoup substitute property when the tainted property itself is unavailable—the procedures outlined in §853(p). And as is clear from its text and structure, §853 maintains traditional in rem forfeiture’s focus on tainted property unless one of §853(p)’s preconditions exists. Pp. 9–10. 816 F. 3d 362, reversed. Sotomayor, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case.

A federal statute— 21 U. S. C. §853—mandates forfeit-ure of “any property constituting, or derived from, any proceeds the person obtained, directly or indirectly, as the result of” certain drug crimes. This case concerns how §853 operates when two or more defendants act as part of a conspiracy. Specifically, the issue is whether, under §853, a defendant may be held jointly and severally liable for property that his co-conspirator derived from the crime but that the defendant himself did not acquire. The Court holds that such liability is inconsistent with the statute’s text and structure. I Terry Michael Honeycutt managed sales and inventory for a Tennessee hardware store owned by his brother, Tony Honeycutt. After observing several “ ‘edgy looking folks’ ” purchasing an iodine-based water-purification product known as Polar Pure, Terry Honeycutt contacted the Chattanooga Police Department to inquire whether the iodine crystals in the product could be used to manufacture methamphetamine. App. to Pet. for Cert. 2a. An officer confirmed that individuals were using Polar Pure for this purpose and advised Honeycutt to cease selling it if the sales made Honeycutt “ ‘uncomfortable.’ ” Ibid. Notwithstanding the officer’s advice, the store continued to sell large quantities of Polar Pure. Although each bottle of Polar Pure contained enough iodine to purify 500 gallons of water, and despite the fact that most people have no legitimate use for the product in large quantities, the brothers sold as many as 12 bottles in a single transaction to a single customer. Over a 3-year period, the store grossed roughly $400,000 from the sale of more than 20,000 bottles of Polar Pure. Unsurprisingly, these sales prompted an investigation by the federal Drug Enforcement Administration along with state and local law enforcement. Authorities exe-cuted a search warrant at the store in November 2010 and seized its entire inventory of Polar Pure—more than 300 bottles. A federal grand jury indicted the Honeycutt brothers for various federal crimes relating to their sale of iodine while knowing or having reason to believe it would be used to manufacture methamphetamine. Pursuant to the Comprehensive Forfeiture Act of 1984, §303, 98Stat. 2045, 21 U. S. C. §853(a)(1), which mandates forfeiture of “any proceeds the person obtained, directly or indirectly, as the result of” drug distribution, the Government sought forfeiture money judgments against each brother in the amount of $269,751.98, which represented the hardware store’s profits from the sale of Polar Pure. Tony Honeycutt pleaded guilty and agreed to forfeit $200,000. Terry went to trial. A jury acquitted Terry Honeycutt of 3 charges but found him guilty of the remaining 11, including conspiring to and knowingly distributing iodine in violation of §§841(c)(2), 843(a)(6), and 846. The District Court sentenced Terry Honeycutt to 60 months in prison. Despite conceding that Terry had no “controlling interest in the store” and “did not stand to benefit personally,” the Government insisted that the District Court “hold [him] jointly liable for the profit from the illegal sales.” App. to Pet. for Cert. 60a–61a. The Government thus sought a money judgment of $69,751.98, the amount of the conspiracy profits outstanding after Tony Honeycutt’s forfeiture payment. The District Court declined to enter a forfeiture judgment, reasoning that Honeycutt was a salaried employee who had not person-ally received any profits from the iodine sales. The Court of Appeals for the Sixth Circuit reversed. As co-conspirators, the court held, the brothers are “ ‘jointly and severally liable for any proceeds of the conspiracy.’ ” 816 F. 3d 362, 380 (2016). The court therefore concluded that each brother bore full responsibility for the entire forfeiture judgment. Ibid. The Court granted certiorari to resolve disagreement among the Courts of Appeals regarding whether joint and several liability applies under §853.[1] 580 U. S. ___ (2016). II Criminal forfeiture statutes empower the Government to confiscate property derived from or used to facilitate criminal activity. Such statutes serve important governmental interests such as “separating a criminal from his ill-gotten gains,” “returning property, in full, to those wrongfully deprived or defrauded of it,” and “lessen[ing] the economic power” of criminal enterprises. Caplin & Drysdale, Chartered v. United States, 491 U. S. 617 –630 (1989). The statute at issue here—§853—mandates forfeiture with respect to persons convicted of certain serious drug crimes. The question presented is whether §853 embraces joint and several liability for forfeiture judgments. A creature of tort law, joint and several liability “applies when there has been a judgment against multiple defendants.” McDermott, Inc. v. AmClyde, 511 U. S. 202 –221 (1994). If two or more defendants jointly cause harm, each defendant is held liable for the entire amount of the harm; provided, however, that the plaintiff recover only once for the full amount. See Restatement (Second) of Torts §875 (1977). Application of that principle in the forfeiture context when two or more defendants conspire to violate the law would require that each defendant be held liable for a forfeiture judgment based not only on property that he used in or acquired because of the crime, but also on property obtained by his co-conspirator. An example is instructive. Suppose a farmer masterminds a scheme to grow, harvest, and distribute mari-juana on local college campuses. The mastermind recruits a college student to deliver packages and pays the student $300 each month from the distribution proceeds for his services. In one year, the mastermind earns $3 million. The student, meanwhile, earns $3,600. If joint and sev-eral liability applied, the student would face a forfeiture judgment for the entire amount of the conspiracy’s proceeds: $3 million. The student would be bound by that judgment even though he never personally acquired any proceeds beyond the $3,600. This case requires determination whether this form of liability is permitted under §853(a)(1). The Court holds that it is not. A Forfeiture under §853 applies to “any person” convicted of certain serious drug crimes. Section 853(a) limits the statute’s reach by defining the property subject to forfeit-ure in three separate provisions. An understanding of how these three provisions work to limit the operation of the statute is helpful to resolving the question in this case. First, the provision at issue here, §853(a)(1), limits forfeit-ure to “property constituting, or derived from, any proceeds the person obtained, directly or indirectly, as the result of” the crime. Second, §853(a)(2) restricts forfeiture to “property used, or intended to be used, in any manner or part, to commit, or to facilitate the commission of,” the crime. Finally, §853(a)(3) applies to persons “convicted of engaging in a continuing criminal enterprise”—a form of conspiracy—and requires forfeiture of “property described in paragraph (1) or (2)” as well as “any of [the defendant’s] interest in, claims against, and property or contractual rights affording a source of control over, the continuing criminal enterprise.” These provisions, by their terms, limit forfeiture under §853 to tainted property; that is, property flowing from (§853(a)(1)), or used in (§853(a)(2)), the crime itself. The limitations of §853(a) thus provide the first clue that the statute does not countenance joint and several liability, which, by its nature, would require forfeiture of untainted property. Recall, for example, the college student from the earlier hypothetical. The $3,600 he received for his part in the marijuana distribution scheme clearly falls within §853(a)(1): It is property he “obtained . . . as the result of” the crime. But if he were held jointly and severally liable for the proceeds of the entire conspiracy, he would owe the Government $3 million. Of the $3 million, $2,996,400 would have no connection whatsoever to the student’s participation in the crime and would have to be paid from the student’s untainted assets. Joint and several liability would thus represent a departure from §853(a)’s restriction of forfeiture to tainted property. In addition to limiting forfeiture to tainted property, §853(a) defines forfeitable property solely in terms of personal possession or use. This is most clear in the specific text of §853(a)(1)—the provision under which the Government sought forfeiture in this case. Section 853(a)(1) limits forfeiture to property the defendant “obtained . . . as the result of” the crime. At the time Congress enacted §853(a)(1), the verb “obtain” was defined as “to come into possession of” or to “get or acquire.” Random House Dictionary of the English Language 995 (1966); see also 7 Oxford English Dictionary 37 (1933) (defining “obtain” as “[t]o come into the possession or enjoyment of (something) by one’s own effort, or by request; to procure or gain, as the result of purpose and effort”). That definition persists today. See Black’s Law Dictionary 1247 (10th ed. 2014) (defining “obtain” as “[t]o bring into one’s own possession; to procure, esp. through effort”); cf. Sekhar v. United States, 570 U. S. ___, ___–___ (2013) (slip op., at 4–5) (“Obtaining property requires ‘. . . the acquisition of property’ ”). Neither the dictionary definition nor the common usage of the word “obtain” supports the conclusion that an individual “obtains” property that was acquired by someone else. Yet joint and several liability would mean just that: The college student would be presumed to have “obtained” the $3 million that the mastermind acquired. Section 853(a)(1) further provides that the forfeitable property may be “obtained, directly or indirectly.” The adverbs “directly” and “indirectly” modify—but do not erase—the verb “obtain.” In other words, these adverbs refer to how a defendant obtains the property; they do not negate the requirement that he obtain it at all. For instance, the marijuana mastermind might receive payments directly from drug purchasers, or he might arrange to have drug purchasers pay an intermediary such as the college student. In all instances, he ultimately “obtains” the property—whether “directly or indirectly.” The other provisions of §853(a) are in accord with the limitation of forfeiture to property the defendant himself obtained. Section 853(a)(2) mandates forfeiture of prop-erty used to facilitate the crime but limits forfeiture to “the person’s property.” Similarly, §853(a)(3) requires forfeit-ure of property related to continuing criminal enterprises, but contrary to joint and several liability principles, requires the defendant to forfeit only “his interest in” the enterprise. Section 853(a)’s limitation of forfeiture to tainted property acquired or used by the defendant, together with the plain text of §853(a)(1), foreclose joint and several liability for co-conspirators. B Joint and several liability is not only contrary to §853(a), it is—for the same reasons—contrary to several other provisions of §853. Two provisions expressly incorporate the §853(a) limitations. First, §853(c) provides that “[a]ll right, title, and interest in property described in subsection (a)”—e.g., tainted property obtained as the result of or used to facilitate the crime—“vests in the United States upon the commission of the act giving rise to forfeiture.” Consistent with its text, the Court has previously acknowledged that §853(c) applies to tainted property only. See Luis v. United States, 578 U. S. ___, ___ (2016) (slip op., at 8). Second, §853(e)(1) authorizes pretrial freezes “to preserve the availability of property described in subsection (a) . . . for forfeiture.” Pretrial restraints on forfeitable property are permitted only when the Government proves, at a hearing, that (1) the defendant has committed an offense triggering forfeiture, and (2) “the property at issue has the requisite connection to that crime.” Kaley v. United States, 571 U. S. ___, ___ (2014) (slip op., at 3); see alsoid., at ___, n. 11 (slip op., at 15, n. 11) (“[F]orfeiture applies only to specific assets”). Another provision, §853(d), does not reference subsection (a) but incorporates its requirements on its own terms. Section 835(d) establishes a “rebuttable presumption” that property is subject to forfeiture only if the Government proves that “such property was acquired by [the defendant] during the period of the violation” and that “there was no likely source for such property other than” the crime. Contrary to all of these provisions, joint and several liability would mandate forfeiture of untainted property that the defendant did not acquire as a result of the crime. It would also render futile one other provision of the statute. Section 853(p)—the sole provision of §853 that permits the Government to confiscate property untainted by the crime—lays to rest any doubt that the statute permits joint and several liability. That provision governs forfeiture of “substitute property” and applies “if any property described in subsection (a), as a result of any act or omission of the defendant” either: “(A) cannot be located upon the exercise of duediligence; “(B) has been transferred or sold to, or deposited with, a third party; “(C) has been placed beyond the jurisdiction of the court; “(D) has been substantially diminished in value; or “(E) has been commingled with other property which cannot be divided without difficulty.” §853(p)(1). Only if the Government can prove that one of these five conditions was caused by the defendant may it seize “any other property of the defendant, up to the value of” the tainted property—rather than the tainted property itself. §853(p)(2). This provision begins from the premise that the defendant once possessed tainted property as “described in subsection (a),” and provides a means for the Government to recoup the value of the property if it has been dissipated or otherwise disposed of by “any act or omission of the defendant.” §853(p)(1). Section 853(p)(1) demonstrates that Congress contemplated situations where the tainted property itself would fall outside the Government’s reach. To remedy that situation, Congress did not authorize the Government to confiscate substitute property from other defendants or co-conspirators; it authorized the Government to confiscate assets only from the defendant who initially acquired the property and who bears responsibility for its dissipation. Permitting the Government to force other co-conspirators to turn over untainted substitute property would allow the Government to circumvent Congress’ carefully constructed statutory scheme, which permits forfeiture of substitute property only when the requirements of §§853(p) and (a) are satisfied. There is no basis to read such an end run into the statute. III Against all of this, the Government asserts the “bedrock principle of conspiracy liability” under which “conspirators are legally responsible for each other’s foreseeable actions in furtherance of their common plan.” Brief for United States 9; see also Pinkerton v. United States, 328 U. S. 640 (1946) . Congress, according to the Government, must be presumed to have legislated against the background principles of conspiracy liability, and thus, “when the traceable proceeds of a conspiracy are unavailable, [§]853 renders conspirators jointly and severally liable for the amount of the proceeds foreseeably obtained by the conspiracy.” Brief for United States 10. Not so. The plain text and structure of §853 leave no doubt that Congress did not incorporate those background principles. Congress provided just one way for the Government to recoup substitute property when the tainted property itself is unavailable—the procedures outlined in §853(p). And, for all the Government makes of the background principles of conspiracy liability, it fails to fully engage with the most important background principles underlying §853: those of forfeiture. Traditionally, forfeiture was an action against the tainted property itself and thus proceeded in rem; that is,proceedings in which “[t]he thing [was] primarily considered as the offender, or rather the offence [was] attached primarily to the thing.” The Palmyra, 12 Wheat. 1, 14 (1827). The forfeiture “proceeding in rem st[ood] independent of, and wholly unaffected by any criminal proceeding in personam” against the defendant. Id., at 15. Congress altered this distinction in enacting §853 by effectively merging the in rem forfeiture proceeding with the in personam criminal proceeding and by expanding forfeiture to include not just the “thing” but “property . . . derived from . . . any proceeds” of the crime. §853(a)(1). But as is clear from its text and structure, §853 maintains traditional in rem forfeiture’s focus on tainted property unless one of the preconditions of §853(p) exists. For those who find it relevant, the legislative history confirms as much: Congress altered the traditional system in order to “improv[e] the procedures applicable in forfeiture cases.” S. Rep. No. 98–225, p. 192 (1983). By adopting an in personam aspect to criminal forfeiture, and providing for substitute-asset forfeiture, Congress made it easier for the Government to hold the defendant who acquired the tainted property responsible. Congress did not, however, enact any “significant expansion of the scope of property subject to forfeiture.” Ibid.[2] IV Forfeiture pursuant to §853(a)(1) is limited to property the defendant himself actually acquired as the result of the crime. In this case, the Government has conceded that Terry Honeycutt had no ownership interest in his brother’s store and did not personally benefit from the Polar Pure sales. App. to Pet. for Cert. 60a. The District Court agreed. Id., at 40a. Because Honeycutt never obtained tainted property as a result of the crime, §853 does not require any forfeiture. The judgment of the Court of Appeals for the Sixth Circuit is reversed. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 Compare United States v. Van Nguyen, 602 F. 3d 886, 904 (CA8 2010) (applying joint and several liability to forfeiture under §853); United States v. Pitt, 193 F. 3d 751, 765 (CA3 1999) (same); United States v. McHan, 101 F. 3d 1027 (CA4 1996) (same); and United States v. Benevento, 836 F. 2d 129, 130 (CA2 1988) (per curiam) (same), with United States v. Cano-Flores, 796 F. 3d 83, 91 (CADC 2015) (declining to apply joint and several liability under §853). 2 Section 853(o) directs that “the provisions of [§853] shall be liberally construed to effectuate its remedial purposes.” The Government points to this as license to read joint and several liability into the statute. But the Court cannot construe a statute in a way that negates its plain text, and here, Congress expressly limited forfeiture to tainted property that the defendant obtained. As explained above, that limitation is incompatible with joint and several liability.

581.US.2016_15-1031

The Uniformed Services Former Spouses’ Protection Act authorizes States to treat veterans’ “disposable retired pay” as community property divisible upon divorce, 10 U. S. C. §1408, but expressly excludes from its definition of “disposable retired pay” amounts deducted from that pay “as a result of a waiver . . . required by law in order to receive” disability benefits, §1408(a)(4)(B). The divorce decree of petitioner John Howell and respondent Sandra Howell awarded Sandra 50% of John’s future Air Force retirement pay, which she began to receive when John retired the following year. About 13 years later, the Department of Veterans Affairs found that John was partially disabled due to an earlier service-related injury. To receive disability pay, federal law required John to give up an equivalent amount of retirement pay. 38 U. S. C. §5305. By his election, John waived about $250 of his retirement pay, which also reduced the value of Sandra’s 50% share. Sandra petitioned the Arizona family court to enforce the original divorce decree and restore the value of her share of John’s total retirement pay. The court held that the original divorce decree had given Sandra a vested interest in the prewaiver amount of John’s retirement pay and ordered John to ensure that she receive her full 50% without regard for the disability waiver. The Arizona Supreme Court affirmed, holding that federal law did not pre-empt the family court’s order. Held: A state court may not order a veteran to indemnify a divorced spouse for the loss in the divorced spouse’s portion of the veteran’s retirement pay caused by the veteran’s waiver of retirement pay to receive service-related disability benefits. This Court’s decision in Mansell v. Mansell, 490 U. S. 581 , determines the outcome here. There, the Court held that federal law completely pre-empts the States from treating waived military retirement pay as divisible community property. Id., at 594–595. The Arizona Supreme Court attempted to distinguish Mansell by emphasizing the fact that the veteran’s waiver in that case took place before the divorce proceeding while the waiver here took place several years after the divorce. This temporal difference highlights only that John’s military pay at the time it came to Sandra was subject to a future contingency, meaning that the value of Sandra’s share of military retirement pay was possibly worth less at the time of the divorce. Nothing in this circumstance makes the Arizona courts’ reimbursement award to Sandra any the less an award of the portion of military pay that John waived in order to obtain disability benefits. That the Arizona courts referred to her interest in the waivable portion as having “vested” does not help: State courts cannot “vest” that which they lack the authority to give. Neither can the State avoid Mansell by describing the family court order as an order requiring John to “reimburse” or to “indemnify” Sandra, rather than an order dividing property, a semantic difference and nothing more. Regardless of their form, such orders displace the federal rule and stand as an obstacle to the accomplishment and execution of the purposes and objectives of Congress. Family courts remain free to take account of the contingency that some military retirement pay might be waived or take account of reductions in value when calculating or recalculating the need for spousal support. Here, however, the state courts made clear that the original divorce decree divided the whole of John’s military pay, and their decisions rested entirely upon the need to restore Sandra’s lost portion. Pp. 6–8. 238 Ariz. 407, 361 P. 3d 936, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Alito, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed an opinion concurring in part and concurring in the judgment. Gorsuch, J., took no part in the consideration or decision of the case.

A federal statute provides that a State may treat as community property, and divide at divorce, a military veteran’s retirement pay. See 10 U. S. C. §1408(c)(1). The statute, however, exempts from this grant of permission any amount that the Government deducts “as a result of a waiver” that the veteran must make “in order to receive” disability benefits. §1408(a)(4)(B). We have held that a State cannot treat as community property, and divide at divorce, this portion (the waived portion) of the veteran’s retirement pay. See Mansell v. Mansell, 490 U. S. 581 –595 (1989). In this case a State treated as community property and awarded to a veteran’s spouse upon divorce a portion of the veteran’s total retirement pay. Long after the divorce, the veteran waived a share of the retirement pay in order to receive nontaxable disability benefits from the Federal Government instead. Can the State subsequently increase, pro rata, the amount the divorced spouse receives each month from the veteran’s retirement pay in order to indemnify the divorced spouse for the loss caused by the veteran’s waiver? The question is complicated, but the answer is not. Our cases and the statute make clear that the answer to the indemnification question is “no.” I A The Federal Government has long provided retirement pay to those veterans who have retired from the Armed Forces after serving, e.g., 20 years or more. It also provides disabled members of the Armed Forces with disability benefits. In order to prevent double counting, however, federal law typically insists that, to receive disability benefits, a retired veteran must give up an equivalent amount of retirement pay. And, since retirement pay is taxable while disability benefits are not, the veteran often elects to waive retirement pay in order to receive disability benefits. See 10 U. S. C. §3911 et seq. (Army retirement benefits); §6321 et seq. (Navy and Marines retirement benefits); §8911 et seq. (Air Force retirement benefits); 38 U. S. C. §5305 (requiring a waiver to receive disability benefits); §5301(a)(1) (exempting disability benefits from taxation). See generally McCarty v. McCarty, 453 U. S. 210 –215 (1981) (describing the military’s nondisability retirement system). In 1981 we considered federal military retirement pay alone, i.e., not in the context of pay waived to receive disability benefits. The question was whether a State could consider any of a veteran’s retirement pay to be a form of community property, divisible at divorce. The Court concluded that the States could not. See McCarty, supra. We noted that the relevant legislative history referred to military retirement pay as a “ ‘personal entitlement.’ ” Id., at 224. We added that other language in the statute as well as its history made “clear that Congress intended that military retired pay ‘actually reach the beneficiary.’ ” Id., at 228. We found a “conflict between the terms of the federal retirement statutes and the [state-conferred] community property right.” Id., at 232. And we concluded that the division of military retirement pay by the States threatened to harm clear and substantial federal interests. Hence federal law pre-empted the state law. Id., at 235. In 1982 Congress responded by passing the Uniformed Services Former Spouses’ Protection Act, 10 U. S. C. §1408. Congress wrote that a State may treat veterans’ “disposable retired pay” as divisible property, i.e., community property divisible upon divorce. §1408(c)(1). But the new Act expressly excluded from its definition of “dispos-able retired pay” amounts deducted from that pay “as a result of a waiver . . . required by law in order to receive” disability benefits. §1408(a)(4)(B). (A recent amendment to the statute renumbered the waiver provision. It now appears at §1408(a)(4)(A)(ii). See Pub. L. 114–328, §641(a), 130Stat. 2164.) In 1989 we interpreted the new federal language in Mansell, 490 U. S. 581 . Major Gerald E. Mansell and his wife had divorced in California. At the time of the divorce, they entered into a “property settlement which provided, in part, that Major Mansell would pay Mrs. Mansell 50 percent of his total military retirement pay, including that portion of retirement pay waived so that Major Mansell could receive disability benefits.” Id., at 586. The divorce decree incorporated this settlement and permitted the division. Major Mansell later moved to modify the decree so that it would omit the portion of the retirement pay that he had waived. The California courts refused to do so. But this Court reversed. It held that federal law forbade California from treating the waived portion as community property divisible at divorce. Justice Thurgood Marshall, writing for the Court, pointed out that federal law, as construed in McCarty, “completely pre-empted the application of state community property law to military retirement pay.” 490 U. S., at 588. He noted that Congress could “overcome” this pre-emption “by enacting an affirmative grant of authority giving the States the power to treat military retirement pay as community property.” Ibid. He recognized that Congress, with its new Act, had done that, but only to a limited extent. The Act provided a “precise and limited” grant of the power to divide federal military retirement pay. Ibid. It did not “gran[t]” the States “the authority to treattotal retired pay as community property.” Id., at 589. Rather, Congress excluded from its grant of authority the disability-related waived portion of military retirement pay. Hence, in respect to the waived portion of retirement pay, McCarty, with its rule of federal pre-emption, still applies. Ibid. B John Howell, the petitioner, and Sandra Howell, the respondent, were divorced in 1991, while John was serving in the Air Force. Anticipating John’s eventual retirement, the divorce decree treated John’s future retirement pay as community property. It awarded Sandra “as her sole and separate property FIFTY PERCENT (50%) of [John’s] military retirement when it begins.” App. to Pet. for Cert. 41a. It also ordered John to pay child support of $585 per month and spousal maintenance of $150 per month until the time of John’s retirement. In 1992 John retired from the Air Force and began to receive military retirement pay, half of which went to Sandra. About 13 years later the Department of Veterans Affairs found that John was 20% disabled due to a service-related shoulder injury. John elected to receive disability benefits and consequently had to waive about $250 per month of the roughly $1,500 of military retirement pay he shared with Sandra. Doing so reduced the amount of retirement pay that he and Sandra received by about $125 per month each. In re Marriage of Howell, 238 Ariz. 407, 408, 361 P. 3d 936, 937 (2015) Sandra then asked the Arizona family court to enforce the original decree, in effect restoring the value of her share of John’s total retirement pay. The court held that the original divorce decree had given Sandra a “vested” interest in the prewaiver amount of that pay, and ordered John to ensure that Sandra “receive her full 50% of the military retirement without regard for the disability.” App. to Pet. for Cert. 28a. The Arizona Supreme Court affirmed the family court’s decision. See 238 Ariz. 407, 361 P. 3d 936. It asked whether the family court could “order John to indemnify Sandra for the reduction” of her share of John’s military retirement pay. Id., at 409, 361 P. 3d, at 938. It wrote that the family court order did not “divide” John’s waived military retirement pay, the order did not require John “to rescind” his waiver, nor did the order “direct him to pay any amount to Sandra from his disability pay.” Id., at 410, 361 P. 3d, at 939. Rather the family court simply ordered John to “reimburse” Sandra for “reducing . . . her share” of military retirement pay. Ibid. The high court concluded that because John had made his waiver after, rather than before, the family court divided his military retirement pay, our decision in Mansell did not control the case, and thus federal law did not preempt the family court’s reimbursement order. 238 Ariz., at 410, 361 P. 3d, at 939. Because different state courts have come to different conclusions on the matter, we granted John Howell’s petition for certiorari. Compare Glover v. Ranney, 314 P. 3d 535, 539–540 (Alaska 2013); Krapf v. Krapf, 439 Mass. 97, 106–107, 786 N. E. 2d 318, 325–326 (2003); and Johnson v. Johnson, 37 S. W. 3d 892, 897–898 (Tenn. 2001), with Mallard v. Burkhart, 95 So. 3d 1264, 1269–1272 (Miss. 2012); and Youngbluth v. Youngbluth, 2010 VT 40, 188 Vt. 53, 62–65, 6 A. 3d 677, 682–685. II This Court’s decision in Mansell determines the outcome here. In Mansell, the Court held that federal law completely pre-empts the States from treating waived military retirement pay as divisible community property. 490 U. S., at 594–595. Yet that which federal law pre-empts is just what the Arizona family court did here. App. to Pet. for Cert. 28a, 35a (finding that the divorce decree gave Sandra a “vested” interest in John’s retirement pay and ordering that Sandra receive her share “without regard for the disability”). The Arizona Supreme Court, the respondent, and the Solicitor General try to distinguish Mansell. But we do not find their efforts convincing. The Arizona Supreme Court, like several other state courts, emphasized the fact that the veteran’s waiver in Mansell took place before the divorce proceeding; the waiver here took place several years after the divorce proceedings. See 238 Ariz., at 410, 361 P. 3d, at 939; see also Abernethy v. Fishkin, 699 So. 2d 235, 240 (Fla. 1997) (noting that a veteran had not yet waived retirement pay at the time of the divorce and permitting indemnification in light of the parties’ “intent to maintain level monthly payments pursuant to their property settlement agreement”). Hence here, as the Solicitor General emphasizes, the nonmilitary spouse and the family court were likely to have assumed that a full share of the veteran’s retirement pay would remain available after the assets were distributed. Nonetheless, the temporal difference highlights only that John’s military retirement pay at the time it came to Sandra was subject to later reduction (should John exercise a waiver to receive disability benefits to which he is entitled). The state court did not extinguish (and most likely would not have had the legal power to extinguish) that future contingency. The existence of that contingency meant that the value of Sandra’s share of military retirement pay was possibly worth less—perhaps less than Sandra and others thought—at the time of the divorce. So too is an ownership interest in property (say, A’s property interest in Blackacre) worth less if it is subject to defeasance or termination upon the occurrence of a later event (say, B’s death). See generally Restatement (Third) of Property §24.3 (2010) (describing property interests that are defeasible); id., §25.3, and Comment a (describing contingent future interests subject to divestment). We see nothing in this circumstance that makes the reimbursement award to Sandra any the less an award of the portion of military retirement pay that John waived in order to obtain disability benefits. And that is the portion that Congress omitted from the Act’s definition of “disposable retired pay,” namely, the portion that federal law prohibits state courts from awarding to a divorced veteran’s former spouse. Mansell, supra, at 589. That the Arizona courts referred to Sandra’s interest in the waiv-able portion as having “vested” does not help. Statecourts cannot “vest” that which (under governing federal law) they lack the authority to give. Cf. 38 U. S. C. §5301(a)(1) (providing that disability benefits are gener-ally nonassignable). Accordingly, while the divorce decree might be said to “vest” Sandra with an immediate right to half of John’s military retirement pay, that interest is, at most, contingent, depending for its amount on a subsequent condition: John’s possible waiver of that pay. Neither can the State avoid Mansell by describing the family court order as an order requiring John to “reimburse” or to “indemnify” Sandra, rather than an order that divides property. The difference is semantic and nothing more. The principal reason the state courts have given for ordering reimbursement or indemnification is that they wish to restore the amount previously awarded as community property, i.e., to restore that portion of retirement pay lost due to the postdivorce waiver. And we note that here, the amount of indemnification mirrors the waived retirement pay, dollar for dollar. Regardless of their form, such reimbursement and indemnification orders displace the federal rule and stand as an obstacle to the accomplishment and execution of the purposes and objectives of Congress. All such orders are thus pre-empted. The basic reasons McCarty gave for believing that Congress intended to exempt military retirement pay from state community property laws apply a fortiori to disability pay. See 453 U. S., at 232–235 (describing the federal interests in attracting and retaining military personnel). And those reasons apply with equal force to a veteran’s postdivorce waiver to receive disability benefits to which he or she has become entitled. We recognize, as we recognized in Mansell, the hardship that congressional pre-emption can sometimes work on divorcing spouses. See 490 U. S., at 594. But we note that a family court, when it first determines the value of a family’s assets, remains free to take account of the contingency that some military retirement pay might be waived, or, as the petitioner himself recognizes, take account of reductions in value when it calculates or recalculates the need for spousal support. See Rose v. Rose, 481 U. S. 619 –634, and n. 6 (1987); 10 U. S. C. §1408(e)(6). We need not and do not decide these matters, for here the state courts made clear that the original divorce decree divided the whole of John’s military retirement pay, and their decisions rested entirely upon the need to restore Sandra’s lost portion. Consequently, the determination of the Supreme Court of Arizona must be reversed. See Mansell, supra, at 594. III The judgment of the Supreme Court of Arizona is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.

581.US.2016_15-1189

A United States patent entitles the patent holder to “exclude others from making, using, offering for sale, or selling [its] invention throughout the United States or importing the invention into the United States.” 35 U. S. C. §154(a). Whoever engages in one of these acts “without authority” from the patentee may face liability for patent infringement. §271(a). When a patentee sells one of its products, however, the patentee can no longer control that item through the patent laws—its patent rights are said to “exhaust.” Respondent Lexmark International, Inc. designs, manufactures, and sells toner cartridges to consumers in the United States and abroad. It owns a number of patents that cover components of those cartridges and the manner in which they are used. When Lexmark sells toner cartridges, it gives consumers two options: One option is to buy a toner cartridge at full price, with no restrictions. The other option is to buy a cartridge at a discount through Lexmark’s “Return Program.” In exchange for the lower price, customers who buy through the Return Program must sign a contract agreeing to use the cartridge only once and to refrain from transferring the cartridge to anyone but Lexmark. Companies known as remanufacturers acquire empty Lexmark toner cartridges—including Return Program cartridges—from purchasers in the United States, refill them with toner, and then resell them. They do the same with Lexmark cartridges that they acquire from purchasers overseas and import into the United States. Lexmark sued a number of these remanufacturers, including petitioner Impression Products, Inc., for patent infringement with respect to two groups of cartridges. The first group consists of Return Program cartridges that Lexmark had sold within the United States. Lexmark argued that, because it expressly prohibited reuse and resale of these cartridges, Impression Products infringed the Lexmark patents when it refurbished and resold them. The second group consists of all toner cartridges that Lexmark had sold abroad and that Impression Products imported into the country. Lexmark claimed that it never gave anyone authority to import these cartridges, so Impression Products infringed its patent rights by doing just that. Impression Products moved to dismiss on the grounds that Lexmark’s sales, both in the United States and abroad, exhausted its patent rights in the cartridges, so Impression Products was free to refurbish and resell them, and to import them if acquired overseas. The District Court granted the motion to dismiss as to the domestic Return Program cartridges, but denied the motion as to the cartridges sold abroad. The Federal Circuit then ruled for Lexmark with respect to both groups of cartridges. Beginning with the Return Program cartridges that Lexmark sold domestically, the Federal Circuit held that a patentee may sell an item and retain the right to enforce, through patent infringement lawsuits, clearly communicated, lawful restrictions on post-sale use or resale. Because Impression Products knew about Lexmark’s restrictions and those restrictions did not violate any laws, Lexmark’s sales did not exhaust its patent rights, and it could sue Impression Products for infringement. As for the cartridges that Lexmark sold abroad, the Federal Circuit held that, when a patentee sells a product overseas, it does not exhaust its patent rights over that item. Lexmark was therefore free to sue for infringement when Impression Products imported cartridges that Lexmark had sold abroad. Judge Dyk, joined by Judge Hughes, dissented. Held: 1. Lexmark exhausted its patent rights in the Return Program cartridges that it sold in the United States. A patentee’s decision to sell a product exhausts all of its patent rights in that item, regardless of any restrictions the patentee purports to impose. As a result, even if the restrictions in Lexmark’s contracts with its customers were clear and enforceable under contract law, they do not entitle Lexmark to retain patent rights in an item that it has elected to sell. Pp. 5–13. (a) The Patent Act grants patentees the “right to exclude others from making, using, offering for sale, or selling [their] invention[s].” 35 U. S. C. §154(a). For over 160 years, the doctrine of patent exhaustion has imposed a limit on that right to exclude: When a patentee sells an item, that product “is no longer within the limits of the [patent] monopoly” and instead becomes the “private, individual property” of the purchaser. Bloomer v. McQuewan, 14 How. 539, 549–550. If the patentee negotiates a contract restricting the purchaser’s right to use or resell the item, it may be able to enforce that restriction as a matter of contract law, but may not do so through a patent infringement lawsuit. The exhaustion rule marks the point where patent rights yield to the common law principle against restraints on alienation. The Patent Act promotes innovation by allowing inventors to secure the financial rewards for their inventions. Once a patentee sells an item, it has secured that reward, and the patent laws provide no basis for restraining the use and enjoyment of the product. Allowing further restrictions would run afoul of the “common law’s refusal to permit restraints on the alienation of chattels.” Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. 519 . As Lord Coke put it in the 17th century, if an owner restricts the resale or use of an item after selling it, that restriction “is voide, because . . . it is against Trade and Traffique, and bargaining and contracting betweene man and man.” 1 E. Coke, Institutes of the Laws of England §360, p. 223 (1628). Congress enacted and has repeatedly revised the Patent Act against the backdrop of this hostility toward restraints on alienation, which is reflected in the exhaustion doctrine. This Court accordingly has long held that, even when a patentee sells an item under an express, otherwise lawful restriction, the patentee does not retain patent rights in that product. See, e.g., Quanta Computer, Inc. v. LG Electronics, Inc., 553 U. S. 617 . And that well-settled line of precedent allows for only one answer in this case: Lexmark cannot bring a patent infringement suit against Impression Products with respect to the Return Program cartridges sold in the United States because, once Lexmark sold those cartridges, it exhausted its right to control them through the patent laws. Pp. 5–9. (b) The Federal Circuit reached a different result because it started from the premise that the exhaustion doctrine is an interpretation of the patent infringement statute, which prohibits anyone from using or selling a patented article “without authority” from the patentee. According to the Federal Circuit, exhaustion reflects a default rule that selling an item “presumptively grant[s] ‘authority’ for the purchaser to use it and resell it.” 816 F. 3d 721, 742. But if a patentee withholds some authority by expressly limiting the purchaser’s rights, the patentee may enforce that restriction through patent infringement lawsuits. See id., at 741. The problem with the Federal Circuit’s logic is that the exhaustion doctrine is not a presumption about the authority that comes along with a sale; it is a limit on the scope of the patentee’s rights. The Patent Act gives patentees a limited exclusionary power, and exhaustion extinguishes that power. A purchaser has the right to use, sell, or import an item because those are the rights that come along with ownership, not because it purchased authority to engage in those practices from the patentee. Pp. 9–13. 2. Lexmark also sold toner cartridges abroad, which Impression Products acquired from purchasers and imported into the United States. Lexmark cannot sue Impression Products for infringement with respect to these cartridges. An authorized sale outside the United States, just as one within the United States, exhausts all rights under the Patent Act. The question about international exhaustion of intellectual property rights has arisen in the context of copyright law. Under the first sale doctrine, when a copyright owner sells a lawfully made copy of its work, it loses the power to restrict the purchaser’s right “to sell or otherwise dispose of . . . that copy.” 17 U. S. C. §109(a). In Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. 519 , this Court held that the first sale doctrine applies to copies of works made and sold abroad. Central to that decision was the fact that the first sale doctrine has its roots in the common law principle against restraints on alienation. Because that principle makes no geographical distinctions and the text of the Copyright Act did not provide such a distinction, a straightforward application of the first sale doctrine required concluding that it applies overseas. Applying patent exhaustion to foreign sales is just as straightforward. Patent exhaustion, too, has its roots in the antipathy toward restraints on alienation, and nothing in the Patent Act shows that Congress intended to confine that principle to domestic sales. Differentiating between the patent exhaustion and copyright first sale doctrines would also make little theoretical or practical sense: The two share a “strong similarity . . . and identity of purpose,” Bauer & Cie v. O’Donnell, 229 U. S. 1 , and many everyday products are subject to both patent and copyright protections. Lexmark contends that a foreign sale does not exhaust patent rights because the Patent Act limits a patentee’s power to exclude others from making, using, selling, or importing its products to acts that occur in the United States. Because those exclusionary powers do not apply abroad, the patentee may not be able to sell its products overseas for the same price as it could in the United States, and therefore is not sure to receive the reward guaranteed by American patent laws. Without that reward, says Lexmark, there should be no exhaustion. The territorial limit on patent rights is no basis for distinguishing copyright protections; those do not have extraterritorial effect either. Nor does the territorial limit support Lexmark’s argument. Exhaustion is a distinct limit on the patent grant, which is triggered by the patentee’s decision to give a patented item up for whatever fee it decides is appropriate. The patentee may not be able to command the same amount for its products abroad as it does in the United States. But the Patent Act does not guarantee a particular price. Instead, the Patent Act just ensures that the patentee receives one reward—of whatever it deems to be satisfactory compensation—for every item that passes outside the scope of its patent monopoly. This Court’s decision in Boesch v. Gräff, 133 U. S. 697 , is not to the contrary. That decision did not, as Lexmark contends, exempt all foreign sales from patent exhaustion. Instead, it held that a sale abroad does not exhaust a patentee’s rights when the patentee had nothing to do with the transaction. That just reaffirms the basic premise that only the patentee can decide whether to make a sale that exhausts its patent rights in an item. Finally, the United States advocates what it views as a middle-ground position: that a foreign sale exhausts patent rights unless the patentee expressly reserves those rights. This express-reservation rule is based on the idea that overseas buyers expect to be able to use and resell items freely, so exhaustion should be the presumption. But, at the same time, lower courts have long allowed patentees to expressly reserve their rights, so that option should remain open to patentees. The sparse and inconsistent decisions the Government cites, however, provide no basis for any expectation, let alone a settled one, that patentees can reserve rights when they sell abroad. The theory behind the express-reservation rule also wrongly focuses on the expectations of the patentee and purchaser during a sale. More is at stake when it comes to patent exhaustion than the dealings between the parties, which can be addressed through contracts. Instead, exhaustion occurs because allowing patent rights to stick to an already-sold item as it travels through the market would violate the principle against restraints on alienation. As a result, restrictions and location are irrelevant for patent exhaustion; what matters is the patentee’s decision to make a sale. Pp. 13–18. 816 F. 3d 721, reversed and remanded. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Thomas, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Ginsburg, J., filed an opinion concurring in part and dissenting in part. Gorsuch, J., took no part in the consideration or decision of the case.

A United States patent entitles the patent holder (the “patentee”), for a period of 20 years, to “exclude others from making, using, offering for sale, or selling [its] invention throughout the United States or importing the invention into the United States.” 35 U. S. C. §154(a). Whoever engages in one of these acts “without authority” from the patentee may face liability for patent infringement. §271(a). When a patentee sells one of its products, however, the patentee can no longer control that item through the patent laws—its patent rights are said to “exhaust.” The purchaser and all subsequent owners are free to use or resell the product just like any other item of personal property, without fear of an infringement lawsuit. This case presents two questions about the scope of the patent exhaustion doctrine: First, whether a patentee that sells an item under an express restriction on the purchaser’s right to reuse or resell the product may enforce that restriction through an infringement lawsuit. And second, whether a patentee exhausts its patent rights by selling its product outside the United States, where American patent laws do not apply. We conclude that a patentee’s decision to sell a product exhausts all of its patent rights in that item, regardless of any restrictions the patentee purports to impose or the location of the sale. I The underlying dispute in this case is about laser printers—or, more specifically, the cartridges that contain the powdery substance, known as toner, that laser printers use to make an image appear on paper. Respondent Lexmark International, Inc. designs, manufactures, and sells toner cartridges to consumers in the United States and around the globe. It owns a number of patents that cover components of those cartridges and the manner in which they are used. When toner cartridges run out of toner they can be refilled and used again. This creates an opportunity for other companies—known as remanufacturers—to acquire empty Lexmark cartridges from purchasers in the United States and abroad, refill them with toner, and then resell them at a lower price than the new ones Lexmark puts on the shelves. Not blind to this business problem, Lexmark structures its sales in a way that encourages customers to return spent cartridges. It gives purchasers two options: One is to buy a toner cartridge at full price, with no strings attached. The other is to buy a cartridge at roughly 20-percent off through Lexmark’s “Return Program.” A customer who buys through the Return Program still owns the cartridge but, in exchange for the lower price, signs a contract agreeing to use it only once and to refrain from transferring the empty cartridge to anyone but Lexmark. To enforce this single-use/no-resale restriction, Lexmark installs a microchip on each Return Program cartridge that prevents reuse once the toner in the cartridge runs out. Lexmark’s strategy just spurred remanufacturers toget more creative. Many kept acquiring empty Return Program cartridges and developed methods to counteract the effect of the microchips. With that technologicalobstacle out of the way, there was little to prevent the re-manufacturers from using the Return Program cartridges in their resale business. After all, Lexmark’s contractual single-use/no-resale agreements were with the initial customers, not with downstream purchasers like the remanufacturers. Lexmark, however, was not so ready to concede that its plan had been foiled. In 2010, it sued a number of remanufacturers, including petitioner Impression Products, Inc., for patent infringement with respect to two groups of cartridges. One group consists of Return Program cartridges that Lexmark sold within the United States. Lexmark argued that, because it expressly prohibited reuse and resale of these cartridges, the remanufacturers infringed the Lexmark patents when they refurbished and resold them. The other group consists of all toner cartridges that Lexmark sold abroad and that remanufacturers imported into the country. Lexmark claimed that it never gave anyone authority to import these cartridges, so the remanufacturers ran afoul of its patent rights by doing just that. Eventually, the lawsuit was whittled down to one defendant, Impression Products, and one defense: that Lexmark’s sales, both in the United States and abroad, exhausted its patent rights in the cartridges, so Impression Products was free to refurbish and resell them, and to import them if acquired abroad. Impression Products filed separate motions to dismiss with respect to both groups of cartridges. The District Court granted the motion as to the domestic Return Program cartridges, but denied the motion as to the cartridges Lexmark sold abroad. Both parties appealed. The Federal Circuit considered the appeals en banc and ruled for Lexmark with respect to both groups of cartridges. The court began with the Return Program cartridgesthat Lexmark sold in the United States. Relying on its decision in Mallinckrodt, Inc. v. Medipart, Inc., 976 F. 2d 700 (1992), the Federal Circuit held that a patentee may sell an item and retain the right to enforce, through patent infringement lawsuits, “clearly communicated, . . . lawful restriction[s] as to post-sale use or resale.” 816 F. 3d 721, 735 (2016). The exhaustion doctrine, the court reasoned, derives from the prohibition on making, using, selling, or importing items “without authority.” Id., at 734 (quoting 35 U. S. C. §271(a)). When you purchase an item you presumptively also acquire the authority to use or resell the item freely, but that is just a presumption; the same authority does not run with the item when the seller restricts post-sale use or resale. 816 F. 3d, at 742. Because the parties agreed that Impression Products knew about Lexmark’s restrictions and that those restrictions did not violate any laws, the Federal Circuit concluded that Lexmark’s sales had not exhausted all of its patent rights, and that the company could sue for infringement when Impression Products refurbished and resold Return Program cartridges. As for the cartridges that Lexmark sold abroad, the Federal Circuit once again looked to its precedent. In Jazz Photo Corp. v. International Trade Commission, 264 F. 3d 1094 (2001), the court had held that a patentee’s decision to sell a product abroad did not terminate its ability to bring an infringement suit against a buyer that “import[ed] the article and [sold] . . . it in the United States.” 816 F. 3d, at 726–727. That rule, the court concluded, makes good sense: Exhaustion is justified when a patentee receives “the reward available from [selling in] American markets,” which does not occur when the patentee sells overseas, where the American patent offers no protection and therefore cannot bolster the price of the patentee’s goods. Id., at 760–761. As a result, Lexmark was free to exercise its patent rights to sue Impression Products for bringing the foreign-sold cartridges to market in the United States. Judge Dyk, joined by Judge Hughes, dissented. In their view, selling the Return Program cartridges in the United States exhausted Lexmark’s patent rights in those items because any “authorized sale of a patented article . . . free[s] the article from any restrictions on use or sale based on the patent laws.” Id., at 775–776. As for the foreign cartridges, the dissenters would have held that a sale abroad also results in exhaustion, unless the seller “explicitly reserve[s] [its] United States patent rights” at the time of sale. Id., at 774, 788. Because Lexmark failed to make such an express reservation, its foreign sales exhausted its patent rights. We granted certiorari to consider the Federal Circuit’s decisions with respect to both domestic and international exhaustion, 580 U. S. ___ (2016), and now reverse. II A First up are the Return Program cartridges that Lexmark sold in the United States. We conclude that Lexmark exhausted its patent rights in these cartridges the moment it sold them. The single-use/no-resale restrictions in Lexmark’s contracts with customers may have been clear and enforceable under contract law, but they do not entitle Lexmark to retain patent rights in an item that it has elected to sell. The Patent Act grants patentees the “right to exclude others from making, using, offering for sale, or selling [their] invention[s].” 35 U. S. C. §154(a). For over 160 years, the doctrine of patent exhaustion has imposed a limit on that right to exclude. See Bloomer v. McQuewan, 14 How. 539 (1853). The limit functions automatically: When a patentee chooses to sell an item, that product “is no longer within the limits of the monopoly” and instead becomes the “private, individual property” of the purchaser, with the rights and benefits that come along with ownership. Id., at 549–550. A patentee is free to set the price and negotiate contracts with purchasers, but may not, “by virtue of his patent, control the use or disposition” of the product after ownership passes to the purchaser. United States v. Univis Lens Co., 316 U. S. 241, 250 (1942) (emphasis added). The sale “terminates all patent rights to that item.” Quanta Computer, Inc. v. LG Electronics, Inc., 553 U. S. 617, 625 (2008) . This well-established exhaustion rule marks the point where patent rights yield to the common law principle against restraints on alienation. The Patent Act “promote[s] the progress of science and the useful arts by granting to [inventors] a limited monopoly” that allows them to “secure the financial rewards” for their inventions. Univis, 316 U. S., at 250. But once a patentee sells an item, it has “enjoyed all the rights secured” by that limited monopoly. Keeler v. Standard Folding Bed Co., 157 U. S. 659, 661 (1895) . Because “the purpose of the patent law is fulfilled . . . when the patentee has received his reward for the use of his invention,” that law furnishes “no basis for restraining the use and enjoyment of the thing sold.” Univis, 316 U. S., at 251. We have explained in the context of copyright law that exhaustion has “an impeccable historic pedigree,” tracing its lineage back to the “common law’s refusal to permit restraints on the alienation of chattels.” Kirtsaeng v. John Wiley & Sons, Inc., 568 U. S. 519, 538 (2013) . As Lord Coke put it in the 17th century, if an owner restricts the resale or use of an item after selling it, that restriction “is voide, because . . . it is against Trade and Traffique, and bargaining and contracting betweene man and man.” 1 E. Coke, Institutes of the Laws of England §360, p. 223 (1628); see J. Gray, Restraints on the Alienation of Prop-erty §27, p. 18 (2d ed. 1895) (“A condition or conditional limitation on alienation attached to a transfer of the entire interest in personalty is as void as if attached to a fee simple in land”). This venerable principle is not, as the Federal Circuit dismissively viewed it, merely “one common-law jurisdiction’s general judicial policy at one time toward anti-alienation restrictions.” 816 F. 3d, at 750. Congress enacted and has repeatedly revised the Patent Act against the backdrop of the hostility toward restraints on alienation. That enmity is reflected in the exhaustion doctrine. The patent laws do not include the right to “restrain[ ] . . . further alienation” after an initial sale; such conditions have been “hateful to the law from Lord Coke’s day to ours” and are “obnoxious to the public interest.” Straus v. Victor Talking Machine Co., 243 U. S. 490, 501 (1917) . “The inconvenience and annoyance to the public that an opposite conclusion would occasion are too obvious to require illustration.” Keeler, 157 U. S., at 667. But an illustration never hurts. Take a shop that restores and sells used cars. The business works because the shop can rest assured that, so long as those bringing in the cars own them, the shop is free to repair and resell those vehicles. That smooth flow of commerce would sputter if companies that make the thousands of parts that go into a vehicle could keep their patent rights after the first sale. Those companies might, for instance, restrict resale rights and sue the shop owner for patent infringement. And even if they refrained from imposing such restrictions, the very threat of patent liability would force the shop to invest in efforts to protect itself from hidden lawsuits. Either way, extending the patent rights beyond the first sale would clog the channels of commerce, with little benefit from the extra control that the patentees retain. And advances in technology, along with increasingly complex supply chains, magnify the problem. See Brief for Costco Wholesale Corp. et al. as Amici Curiae 7–9; Brief for Intel Corp. et al. as Amici Curiae 17, n. 5 (“A generic smartphone assembled from various high-tech components could practice an estimated 250,000 patents”). This Court accordingly has long held that, even when a patentee sells an item under an express restriction, the patentee does not retain patent rights in that product. In Boston Store of Chicago v. American Graphophone Co., for example, a manufacturer sold graphophones—one of the earliest devices for recording and reproducing sounds—to retailers under contracts requiring those stores to resell at a specific price. 246 U. S. 8 –18 (1918). When the manufacturer brought a patent infringement suit against a retailer who sold for less, we concluded that there was “no room for controversy” about the result: By selling the item, the manufacturer placed it “beyond the confines of the patent law, [and] could not, by qualifying restrictions as to use, keep [it] under the patent monopoly.” Id., at 20, 25. Two decades later, we confronted a similar arrangement in United States v. Univis Lens Co. There, a company that made eyeglass lenses authorized an agent to sell its products to wholesalers and retailers only if they promised to market the lenses at fixed prices. The Government filed an antitrust lawsuit, and the company defended its arrangement on the ground that it was exercising authority under the Patent Act. We held that the initial sales “relinquish[ed] . . . the patent monopoly with respect to the article[s] sold,” so the “stipulation . . . fixing resale prices derive[d] no support from the patent and must stand on the same footing” as restrictions on unpatented goods. 316 U. S., at 249–251. It is true that Boston Store and Univis involved resale price restrictions that, at the time of those decisions, violated the antitrust laws. But in both cases it was the sale of the items, rather than the illegality of the restrictions, that prevented the patentees from enforcing those resale price agreements through patent infringement suits. And if there were any lingering doubt that patent exhaustion applies even when a sale is subject to an express, otherwise lawful restriction, our recent decision in Quanta Computer, Inc. v. LG Electronics, Inc. settled the matter. In that case, a technology company—with authorization from the patentee—sold microprocessors under contracts requiring purchasers to use those processors with other parts that the company manufactured. One buyer disregarded the restriction, and the patentee sued for infringement. Without so much as mentioning the lawfulness of the contract, we held that the patentee could not bring an infringement suit because the “authorized sale . . . took its products outside the scope of the patent monopoly.” 553 U. S., at 638. Turning to the case at hand, we conclude that this well-settled line of precedent allows for only one answer: Lexmark cannot bring a patent infringement suit against Impression Products to enforce the single-use/no-resale provision accompanying its Return Program cartridges. Once sold, the Return Program cartridges passed outside of the patent monopoly, and whatever rights Lexmark retained are a matter of the contracts with its purchasers, not the patent law. B The Federal Circuit reached a different result largely because it got off on the wrong foot. The “exhaustion doctrine,” the court believed, “must be understood as an interpretation of” the infringement statute, which prohibits anyone from using or selling a patented article “without authority” from the patentee. 816 F. 3d, at 734 (quoting 35 U. S. C. §271(a)). Exhaustion reflects a default rule that a patentee’s decision to sell an item “presumptively grant[s] ‘authority’ to the purchaser to use it and resell it.” 816 F. 3d, at 742. But, the Federal Circuit explained, the patentee does not have to hand over the full “bundle of rights” every time. Id., at 741 (internal quotation marks omitted). If the patentee expressly withholds a stick from the bundle—perhaps by restricting the purchaser’s resale rights—the buyer never acquires that withheld authority, and the patentee may continue to enforce its right to exclude that practice under the patent laws. The misstep in this logic is that the exhaustion doctrine is not a presumption about the authority that comes along with a sale; it is instead a limit on “the scope of the patentee’s rights.” United States v. General Elec. Co., 272 U. S. 476, 489 (1926) (emphasis added). The right to use, sell, or import an item exists independently of the Patent Act. What a patent adds—and grants exclusively to the pat-entee—is a limited right to prevent others from engaging in those practices. See Crown Die & Tool Co. v. Nye Tool & Machine Works, 261 U. S. 24, 35 (1923) . Exhaustion extinguishes that exclusionary power. See Bloomer, 14 How., at 549 (the purchaser “exercises no rights created by the act of Congress, nor does he derive title to [the item] by virtue of the . . . exclusive privilege granted to the patentee”). As a result, the sale transfers the right to use, sell, or import because those are the rights that come along with ownership, and the buyer is free and clear of an infringement lawsuit because there is no exclusionary right left to enforce. The Federal Circuit also expressed concern that preventing patentees from reserving patent rights when they sell goods would create an artificial distinction between such sales and sales by licensees. Patentees, the court explained, often license others to make and sell their products, and may place restrictions on those licenses. A computer developer could, for instance, license a manufacturer to make its patented devices and sell them only for non-commercial use by individuals. If a licensee breaches the license by selling a computer for commercial use, the patentee can sue the licensee for infringement. And, in the Federal Circuit’s view, our decision in General Talking Pictures Corp. v. Western Elec. Co., 304 U. S. 175 , aff’d on reh’g, 305 U. S. 124 (1938) , established that—when a patentee grants a license “under clearly stated restrictions on post-sale activities” of those who purchase products from the licensee—the patentee can also sue for infringement those purchasers who knowingly violate the restrictions. 816 F. 3d, at 743–744. If patentees can employ licenses to impose post-sale restrictions on purchasers that are enforceable through infringement suits, the court concluded, it would make little sense to prevent patentees from doing so when they sell directly to consumers. The Federal Circuit’s concern is misplaced. A patentee can impose restrictions on licensees because a license does not implicate the same concerns about restraints on alienation as a sale. Patent exhaustion reflects the principle that, when an item passes into commerce, it should not be shaded by a legal cloud on title as it moves through the marketplace. But a license is not about passing title to a product, it is about changing the contours of the patentee’s monopoly: The patentee agrees not to exclude a licensee from making or selling the patented invention, expanding the club of authorized producers and sellers. See General Elec. Co., 272 U. S., at 489–490. Because the patentee is exchanging rights, not goods, it is free to relinquish only a portion of its bundle of patent protections. A patentee’s authority to limit licensees does not, as the Federal Circuit thought, mean that patentees can use licenses to impose post-sale restrictions on purchasers that are enforceable through the patent laws. So long as a licensee complies with the license when selling an item, the patentee has, in effect, authorized the sale. That licensee’s sale is treated, for purposes of patent exhaustion, as if the patentee made the sale itself. The result: The sale exhausts the patentee’s rights in that item. See Hobbie v. Jennison, 149 U. S. 355 –363 (1893). A license may require the licensee to impose a restriction on purchasers, like the license limiting the computer manufacturer to selling for non-commercial use by individuals. But if the licensee does so—by, perhaps, having each customer sign a contract promising not to use the computers in business—the sale nonetheless exhausts all patent rights in the item sold. See Motion Picture Patents Co. v. Universal Film Mfg. Co., 243 U. S. 502 –507, 516 (1917). The purchasers might not comply with the restriction, but the only recourse for the licensee is through contract law, just as if the patentee itself sold the item with a restriction. General Talking Pictures involved a fundamentally different situation: There, a licensee “knowingly ma[de] . . . sales . . . outside the scope of its license.” 304 U. S., at 181–182 (emphasis added). We treated the sale “as if no license whatsoever had been granted” by the patentee, which meant that the patentee could sue both the licensee and the purchaser—who knew about the breach—for infringement. General Talking Pictures Corp. v. Western Elec. Co., 305 U. S. 124, 127 (1938) . This does not mean that patentees can use licenses to impose post-sale restraints on purchasers. Quite the contrary: The licensee infringed the patentee’s rights because it did not comply with the terms of its license, and the patentee could bring a patent suit against the purchaser only because the purchaser participated in the licensee’s infringement. General Talking Pictures, then, stands for the modest principle that, if a patentee has not given authority for a licensee to make a sale, that sale cannot exhaust the patentee’s rights. In sum, patent exhaustion is uniform and automatic. Once a patentee decides to sell—whether on its own or through a licensee—that sale exhausts its patent rights, regardless of any post-sale restrictions the patentee purports to impose, either directly or through a license. III Our conclusion that Lexmark exhausted its patent rights when it sold the domestic Return Program cartridges goes only halfway to resolving this case. Lexmark also sold toner cartridges abroad and sued Impression Products for patent infringement for “importing [Lexmark’s] invention into the United States.” 35 U. S. C. §154(a). Lexmark contends that it may sue for infringement with respect to all of the imported cartridges—not just those in the Return Program—because a foreign sale does not trigger patent exhaustion unless the patentee “expressly or implicitly transfer[s] or license[s]” its rights. Brief for Respondent 36–37. The Federal Circuit agreed, but we do not. An authorized sale outside the United States, just as one within the United States, exhausts all rights under the Patent Act. This question about international exhaustion of intellectual property rights has also arisen in the context of copyright law. Under the “first sale doctrine,” which is codified at 17 U. S. C. §109(a), when a copyright owner sells a lawfully made copy of its work, it loses the power to restrict the purchaser’s freedom “to sell or otherwise dispose of . . . that copy.” In Kirtsaeng v. John Wiley & Sons, Inc., we held that this “ ‘first sale’ [rule] applies to copies of a copyrighted work lawfully made [and sold] abroad.” 568 U. S., at 525. We began with the text of §109(a), but it was not decisive: The language neither “restrict[s] the scope of [the] ‘first sale’ doctrine geographically,” nor clearly embraces international exhaustion. Id., at 528–533. What helped tip the scales for global exhaustion was the fact that the first sale doctrine originated in “the common law’s refusal to permit restraints on the alienation of chattels.” Id., at 538. That “common-law doctrine makes no geographical distinctions.” Id., at 539. The lack of any textual basis for distinguishing between domestic and international sales meant that “a straightforward application” of the first sale doctrine required the conclusion that it applies overseas. Id., at 540 (internal quotation marks omitted). Applying patent exhaustion to foreign sales is just as straightforward. Patent exhaustion, too, has its roots in the antipathy toward restraints on alienation, see supra, at 6–8, and nothing in the text or history of the Patent Act shows that Congress intended to confine that borderless common law principle to domestic sales. In fact, Congress has not altered patent exhaustion at all; it remains an unwritten limit on the scope of the patentee’s monopoly. See Astoria Fed. Sav. & Loan Assn. v. Solimino, 501 U. S. 104, 108 (1991) (“[W]here a common-law principle is well established, . . . courts may take it as given that Congress has legislated with an expectation that the principle will apply except when a statutory purpose to the contrary is evident” (internal quotation marks omitted)). And differentiating the patent exhaustion and copyright first sale doctrines would make little theoretical or practical sense: The two share a “strong similarity . . . and identity of purpose,” Bauer & Cie v. O’Donnell, 229 U. S. 1, 13 (1913) , and many everyday products—“automobiles, microwaves, calculators, mobile phones, tablets, and personal computers”—are subject to both patent and copyright protections, see Kirtsaeng, 568 U. S., at 545; Brief for Costco Wholesale Corp. et al. as Amici Curiae 14–15. There is a “historic kinship between patent law and copyright law,” Sony Corp. of America v. Universal City Studios, Inc., 464 U. S. 417, 439 (1984) , and the bond between the two leaves no room for a rift on the question of international exhaustion. Lexmark sees the matter differently. The Patent Act, it points out, limits the patentee’s “right to exclude others” from making, using, selling, or importing its products to acts that occur in the United States. 35 U. S. C. §154(a). A domestic sale, it argues, triggers exhaustion because the sale compensates the patentee for “surrendering [those] U. S. rights.” Brief for Respondent 38. A foreign sale is different: The Patent Act does not give patentees exclusionary powers abroad. Without those powers, a patentee selling in a foreign market may not be able to sell its product for the same price that it could in the United States, and therefore is not sure to receive “the reward guaranteed by U. S. patent law.” Id., at 39 (internal quotation marks omitted). Absent that reward, says Lexmark, there should be no exhaustion. In short, there is no patent exhaustion from sales abroad because there are no patent rights abroad to exhaust. The territorial limit on patent rights is, however, no basis for distinguishing copyright protections; those protections “do not have any extraterritorial operation” either. 5 M. Nimmer & D. Nimmer, Copyright §17.02, p. 17–26 (2017). Nor does the territorial limit support the premise of Lexmark’s argument. Exhaustion is a separate limit on the patent grant, and does not depend on the patentee receiving some undefined premium for selling the right to access the American market. A purchaser buys an item, not patent rights. And exhaustion is triggered by the patentee’s decision to give that item up and receive whatever fee it decides is appropriate “for the article and the invention which it embodies.” Univis, 316 U. S., at 251. The patentee may not be able to command the same amount for its products abroad as it does in the United States. But the Patent Act does not guarantee a particular price, much less the price from selling to American consumers. Instead, the right to exclude just ensures that the patentee receives one reward—of whatever amount the patentee deems to be “satisfactory compensation,” Keeler, 157 U. S., at 661—for every item that passes outside the scope of the patent monopoly. This Court has addressed international patent exhaustion in only one case, Boesch v. Gräff, decided over 125 years ago. All that case illustrates is that a sale abroad does not exhaust a patentee’s rights when the patentee had nothing to do with the transaction. Boesch—from the days before the widespread adoption of electrical lighting—involved a retailer who purchased lamp burners from a manufacturer in Germany, with plans to sell them in the United States. The manufacturer had authority to make the burners under German law, but there was a hitch: Two individuals with no ties to the German manufacturer held the American patent to that invention. These patentees sued the retailer for infringement when the retailer imported the lamp burners into the United States, and we rejected the argument that the German manufacturer’s sale had exhausted the American patentees’ rights. The German manufacturer had no permission to sell in the United States from the American patentees, and the American patentees had not exhausted their patent rights in the products because they had not sold them to anyone, so “purchasers from [the German manufacturer] could not be thereby authorized to sell the articles in the United States.” 133 U. S. 697, 703 (1890) . Our decision did not, as Lexmark contends, exempt all foreign sales from patent exhaustion. See Brief for Respondent 44–45. Rather, it reaffirmed the basic premise that only the patentee can decide whether to make a sale that exhausts its patent rights in an item. The American patentees did not do so with respect to the German products, so the German sales did not exhaust their rights. Finally, the United States, as an amicus, advocates what it views as a middle-ground position: that “a foreign sale authorized by the U. S. patentee exhausts U. S. pat-ent rights unless those rights are expressly reserved.” Brief for United States 7–8. Its position is largely based on policy rather than principle. The Government thinks that an overseas “buyer’s legitimate expectation” is that a “sale conveys all of the seller’s interest in the patented article,” so the presumption should be that a foreign sale triggers exhaustion. Id., at 32–33. But, at the same time, “lower courts long ago coalesced around” the rule that “a patentee’s express reservation of U. S. patent rights at the time of a foreign sale will be given effect,” so that option should remain open to the patentee. Id., at 22 (emphasis deleted). The Government has little more than “long ago” on its side. In the 1890s, two circuit courts—in cases involving the same company—did hold that patentees may use express restrictions to reserve their patent rights in connection with foreign sales. See Dickerson v. Tinling, 84 F. 192, 194–195 (CA8 1897); Dickerson v. Matheson, 57 F. 524, 527 (CA2 1893). But no “coalesc[ing]” ever took place: Over the following hundred-plus years, only a smattering of lower court decisions mentioned this express-reservation rule for foreign sales. See, e.g., Sanofi, S. A. v. Med-Tech Veterinarian Prods., Inc., 565 F. Supp. 931, 938 (NJ 1983). And in 2001, the Federal Circuit adopted its blanket rule that foreign sales do not trigger exhaustion, even if the patentee fails to expressly reserve its rights. Jazz Photo, 264 F. 3d, at 1105. These sparse and inconsistent decisions provide no basis for any expectation, let alone a settled one, that patentees can reserve patent rights when they sell abroad. The theory behind the Government’s express-reservation rule also wrongly focuses on the likely expectations of the patentee and purchaser during a sale. Exhaustion does not arise because of the parties’ expectations about how sales transfer patent rights. More is at stake when it comes to patents than simply the dealings between the parties, which can be addressed through contract law. Instead, exhaustion occurs because, in a sale, the patentee elects to give up title to an item in exchange for payment. Allowing patent rights to stick remora-like to that item as it flows through the market would violate the principle against restraints on alienation. Exhaustion does not depend on whether the patentee receives a premium for selling in the United States, orthe type of rights that buyers expect to receive. As a result, restrictions and location are irrelevant; what matters is the patentee’s decision to make a sale. * * * The judgment of the United States Court of Appeals for the Federal Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.

581.US.2016_16-32

Respondents Beverly Wellner and Janis Clark—the wife and daughter, respectively, of Joe Wellner and Olive Clark—each held a power of attorney affording her broad authority to manage her family member’s affairs. When Joe and Olive moved into a nursing home operated by petitioner Kindred Nursing Centers L. P., Beverly and Janis used their powers of attorney to complete all necessary paperwork. As part of that process, each signed an arbitration agreement on her relative’s behalf providing that any claims arising from the relative’s stay at the facility would be resolved through binding arbitration. After Joe and Olive died, their estates (represented by Beverly and Janis) filed suits alleging that Kindred’s substandard care had caused their deaths. Kindred moved to dismiss the cases, arguing that the arbitration agreements prohibited bringing the disputes to court. The trial court denied Kindred’s motions, and the Kentucky Court of Appeals agreed that the suits could go forward. The Kentucky Supreme Court consolidated the cases and affirmed. The court initially found that the language of the Wellner power of attorney did not permit Beverly to enter into an arbitration agreement on Joe’s behalf, but that the Clark document gave Janis the capacity to do so on behalf of Olive. Nonetheless, the court held, both arbitration agreements were invalid because neither power of attorney specifically entitled the representative to enter into an arbitration agreement. Because the Kentucky Constitution declares the rights of access to the courts and trial by jury to be “sacred” and “inviolate,” the court determined, an agent could deprive her principal of such rights only if expressly provided in the power of attorney. Held: The Kentucky Supreme Court’s clear-statement rule violates the Federal Arbitration Act by singling out arbitration agreements for disfavored treatment. Pp. 4–10. (a) The FAA, which makes arbitration agreements “valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract,” 9 U. S. C. §2, establishes an equal-treatment principle: A court may invalidate an arbitration agreement based on “generally applicable contract defenses,” but not on legal rules that “apply only to arbitration or that derive their meaning from the fact that an agreement to arbitrate is at issue,” AT&T Mobility LLC v. Concepcion, 563 U. S. 333 . The Act thus preempts any state rule that discriminates on its face against arbitration or that covertly accomplishes the same objective by disfavoring contracts that have the defining features of arbitration agreements. The Kentucky Supreme Court’s clear-statement rule fails to put arbitration agreements on an equal plane with other contracts. By requiring an explicit statement before an agent can relinquish her principal’s right to go to court and receive a jury trial, the court did exactly what this Court has barred: adopt a legal rule hinging on the primary characteristic of an arbitration agreement. Pp. 4–7. (b) In support of the decision below, respondents argue that the clear-statement rule affects only contract formation, and that the FAA does not apply to contract formation questions. But the Act’s text says otherwise. The FAA cares not only about the “enforce[ment]” of arbitration agreements, but also about their initial “valid[ity]”—that is, about what it takes to enter into them. 9 U. S. C. §2. Precedent confirms the point. In Concepcion, the Court noted the impermissibility of applying a contract defense like duress “in a fashion that disfavors arbitration.” 563 U. S., at 341. That discussion would have made no sense if the FAA had nothing to say about contract formation, because duress involves “unfair dealing at the contract formation stage.” Morgan Stanley Capital Group Inc. v. Public Util. Dist. No. 1 of Snohomish Cty., 554 U. S. 527 . Finally, respondents’ view would make it trivially easy for States to undermine the Act. Pp. 7–9. (c) Because the Kentucky Supreme Court invalidated the Clark-Kindred arbitration agreement based exclusively on the clear-statement rule, the court must now enforce that agreement. But because it is unclear whether the court’s interpretation of the Wellner document was wholly independent of its rule, the court should determine on remand whether it adheres, in the absence of the rule, to its prior reading of that power of attorney. Pp. 9–10. 478 S. W. 3d 306, reversed in part, vacated in part, and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, Alito, and Sotomayor, JJ., joined. Thomas, J., filed a dissenting opinion. Gorsuch, J., took no part in the consideration or decision of the case.

The Federal Arbitration Act (FAA or Act) requires courts to place arbitration agreements “on equal footing with all other contracts.” DIRECTV, Inc. v. Imburgia, 577 U. S. ___, ___ (2015) (slip op., at 6) (quoting Buckeye Check Cashing, Inc. v. Cardegna, 546 U. S. 440, 443 (2006) ); see 9 U. S. C. §2. In the decision below, the Kentucky Supreme Court declined to give effect to two arbitration agreements executed by individuals holding “powers of attorney”—that is, authorizations to act on behalf of others. According to the court, a general grant of power (even if seemingly comprehensive) does not permit a legal representative to enter into an arbitration agreement for someone else; to form such a contract, the representative must possess specific authority to “waive his principal’s fundamental constitutional rights to access the courts [and] to trial by jury.” Extendicare Homes, Inc. v. Whisman, 478 S. W. 3d 306, 327 (2015). Because that rule singles out arbitration agreements for disfavored treatment, we hold that it violates the FAA. I Petitioner Kindred Nursing Centers L. P. operates nursing homes and rehabilitation centers. Respondents Beverly Wellner and Janis Clark are the wife and daughter, respectively, of Joe Wellner and Olive Clark, two now-deceased residents of a Kindred nursing home called the Winchester Centre. At all times relevant to this case, Beverly and Janis each held a power of attorney, designating her as an “attorney-in-fact” (the one for Joe, the other for Olive) and affording her broad authority to manage her family member’s affairs. In the Wellner power of attorney, Joe gave Beverly the authority, “in my name, place and stead,” to (among other things) “institute legal proceedings” and make “contracts of every nature in relation to both real and personal property.” App. 10–11. In the Clark power of attorney, Olive provided Janis with “full power . . . to transact, handle, and dispose of all matters affecting me and/or my estate in any possible way,” including the power to “draw, make, and sign in my name any and all . . . contracts, deeds, or agreements.” Id., at 7. Joe and Olive moved into the Winchester Centre in 2008, with Beverly and Janis using their powers of attorney to complete all necessary paperwork. As part of that process, Beverly and Janis each signed an arbitration agreement with Kindred on behalf of her relative. The two contracts, worded identically, provided that “[a]ny and all claims or controversies arising out of or in any way relating to . . . the Resident’s stay at the Facility” would be resolved through “binding arbitration” rather than a lawsuit. Id., at 14, 21. When Joe and Olive died the next year, their estates (represented again by Beverly and Janis) brought separate suits against Kindred in Kentucky state court. The complaints alleged that Kindred had delivered substandard care to Joe and Olive, causing their deaths. Kindred moved to dismiss the cases, arguing that the arbitration agreements Beverly and Janis had signed prohibited bringing their disputes to court. But the trial court denied Kindred’s motions, and the Kentucky Court of Appeals agreed that the estates’ suits could go forward. See App. to Pet. for Cert. 125a–126a, 137a–138a. The Kentucky Supreme Court, after consolidating the cases, affirmed those decisions by a divided vote. See 478 S. W. 3d, at 313. The court began with the language of the two powers of attorney. The Wellner document, the court stated, did not permit Beverly to enter into an arbitration agreement on Joe’s behalf. In the court’s view, neither the provision authorizing her to bring legal proceedings nor the one enabling her to make property-related contracts reached quite that distance. See id., at 325–326; supra, at 2. By contrast, the court thought, the Clark power of attorney extended that far and beyond. Under that document, after all, Janis had the capacity to “dispose of all matters” affecting Olive. See supra, at 2. “Given this extremely broad, universal delegation of authority,” the court acknowledged, “it would be impossible to say that entering into [an] arbitration agreement was not covered.” 478 S. W. 3d, at 327. And yet, the court went on, both arbitration agreements—Janis’s no less than Beverly’s—were invalid. That was because a power of attorney could not entitle a representative to enter into an arbitration agreement without specifically saying so. The Kentucky Constitution, the court explained, protects the rights of access to the courts and trial by jury; indeed, the jury guarantee is the sole right the Constitution declares “sacred” and “inviolate.” Id., at 328–329. Accordingly, the court held, an agent could deprive her principal of an “adjudication by judge or jury” only if the power of attorney “expressly so provide[d].” Id., at 329. And that clear-statement rule—so said the court—complied with the FAA’s demands. True enough that the Act precludes “singl[ing] out arbitration agreements.” Ibid. (internal quotation marks omitted). But that was no problem, the court asserted, because its rule would apply not just to those agreements, but also to some other contracts implicating “fundamental constitutional rights.” Id., at 328. In the future, for example, the court would bar the holder of a “non-specific” power of attorney from entering into a contract “bind[ing] the principal to personal servitude.” Ibid. Justice Abramson dissented, in an opinion joined by two of her colleagues. In their view, the Kentucky Supreme Court’s new clear-statement rule was “clearly not . . . applicable to ‘any contract’ but [instead] single[d] out arbitration agreements for disfavored treatment.” Id., at 344–345. Accordingly, the dissent concluded, the rule “r[a]n afoul of the FAA.” Id., at 353. We granted certiorari. 580 U. S. ___ (2016). II A The FAA makes arbitration agreements “valid, irrevocable, and enforceable, save upon such grounds as exist at law or in equity for the revocation of any contract.” 9 U. S. C. §2. That statutory provision establishes an equal-treatment principle: A court may invalidate an arbitration agreement based on “generally applicable contract defenses” like fraud or unconscionability, but not on legal rules that “apply only to arbitration or that derive their meaning from the fact that an agreement to arbitrate is at issue.” AT&T Mobility LLC v. Concepcion, 563 U. S. 333, 339 (2011) . The FAA thus preempts any state rule discriminating on its face against arbitration—for example, a “law prohibit[ing] outright the arbitration of a particular type of claim.” Id., at 341. And not only that: The Act also displaces any rule that covertly accomplishes the same objective by disfavoring contracts that (oh so coinciden-tally) have the defining features of arbitration agreements. In Concepcion, for example, we described a hypothetical state law declaring unenforceable any contract that “disallow[ed] an ultimate disposition [of a dispute] by a jury.” Id., at 342. Such a law might avoid referring to arbitration by name; but still, we explained, it would “rely on the uniqueness of an agreement to arbitrate as [its] basis”—and thereby violate the FAA. Id., at 341 (quoting Perry v. Thomas, 482 U. S. 483 , n. 9 (1987)). The Kentucky Supreme Court’s clear-statement rule, in just that way, fails to put arbitration agreements on an equal plane with other contracts. By the court’s own account, that rule (like the one Concepcion posited) serves to safeguard a person’s “right to access the courts and to trial by jury.” 478 S. W. 3d, at 327; see supra, at 3–4. In ringing terms, the court affirmed the jury right’s unsurpassed standing in the State Constitution: The framers, the court explained, recognized “that right and that right alone as a divine God-given right” when they made it “the only thing” that must be “ ‘held sacred’ ” and “ ‘inviolate.’ ” 478 S. W. 3d, at 328–329 (quoting Ky. Const. §7). So it was that the court required an explicit statement before an attorney-in-fact, even if possessing broad delegated powers, could relinquish that right on another’s behalf. See 478 S. W. 3d, at 331 (“We say only that an agent’s authority to waive his principal’s constitutional right to access the courts and to trial by jury must be clearly expressed by the principal”). And so it was that the court did exactly what Concepcion barred: adopt a legal rule hinging on the primary characteristic of an arbitration agreement—namely, a waiver of the right to go to court and receive a jury trial. See 563 U. S., at 341–342; see also 478 S. W. 3d, at 353 (Abramson, J., dissenting) (noting that the jury-trial right at the core of “the majority’s new rule” is “the one right that just happens to be correlative to the right to arbitrate” (emphasis deleted)). Such a rule is too tailor-made to arbitration agreements—subjecting them, by virtue of their defining trait, to uncommon barriers—to survive the FAA’s edict against singling out those contracts for disfavored treatment.[1] And the state court’s sometime-attempt to cast the rule in broader terms cannot salvage its decision. The clear-statement requirement, the court suggested, could also apply when an agent endeavored to waive other “fundamental constitutional rights” held by a principal. 478 S. W. 3d, at 331; see supra, at 4. But what other rights, really? No Kentucky court, so far as we know, has ever before demanded that a power of attorney explicitly confer authority to enter into contracts implicating constitutional guarantees. Nor did the opinion below indicate that such a grant would be needed for the many routine contracts—executed day in and day out by legal representatives—meeting that description. For example, the Kentucky Constitution protects the “inherent and inalienable” rights to “acquir[e] and protect[ ] property” and to “freely communicat[e] thoughts and opinions.” Ky. Const. §1. But the state court nowhere cautioned that an attorney-in-fact would now need a specific authorization to, say, sell her principal’s furniture or commit her principal to a non-disclosure agreement. (And were we in the business of giving legal advice, we would tell the agent not to worry.) Rather, the court hypothesized a slim set of both patently objectionable and utterly fanciful contracts that would be subject to its rule: No longer could a representative lacking explicit authorization waive her “principal’s right to worship freely” or “consent to an arranged marriage” or “bind [her] principal to personal servitude.” 478 S. W. 3d, at 328; see supra, at 4. Placing arbitration agreements within that class reveals the kind of “hostility to arbitration” that led Congress to enact the FAA. Concepcion, 563 U. S., at 339. And doing so only makes clear the arbitration-specific character of the rule, much as if it were made applicable to arbitration agreements and black swans.[2] B The respondents, Janis and Beverly, primarily advance a different argument—based on the distinction between contract formation and contract enforcement—to support the decision below. Kentucky’s clear-statement rule, they begin, affects only contract formation, because it bars agents without explicit authority from entering into arbitration agreements. And in their view, the FAA has “no application” to “contract formation issues.” Supp. Brief for Respondents 1. The Act, to be sure, requires a State to enforce all arbitration agreements (save on generally applicable grounds) once they have come into being. But, the respondents claim, States have free rein to decide—irrespective of the FAA’s equal-footing principle—whether such contracts are validly created in the first instance. See id., at 3 (“The FAA’s statutory framework applies only after a court has determined that a valid arbitration agreement was formed”). Both the FAA’s text and our case law interpreting it say otherwise. The Act’s key provision, once again, states that an arbitration agreement must ordinarily be treated as “valid, irrevocable, and enforceable.” 9 U. S. C. §2; see supra, at 4. By its terms, then, the Act cares not only about the “enforce[ment]” of arbitration agreements, but also about their initial “valid[ity]”—that is, about what it takes to enter into them. Or said otherwise: A rule selectively finding arbitration contracts invalid because improperly formed fares no better under the Act than a rule selectively refusing to enforce those agreements once properly made. Precedent confirms that point. In Concepcion, we noted the impermissibility of applying a contract defense like duress “in a fashion that disfavors arbitration.” 563 U. S., at 341. But the doctrine of duress, as we have elsewhere explained, involves “unfair dealing at the contract formation stage.” Morgan Stanley Capital Group Inc. v. Public Util. Dist. No. 1 of Snohomish Cty., 554 U. S. 527, 547 (2008) . Our discussion of duress would have made no sense if the FAA, as the respondents contend, had nothing to say about contract formation. And still more: Adopting the respondents’ view would make it trivially easy for States to undermine the Act—indeed, to wholly defeat it. As the respondents have acknowledged, their reasoning would allow States to pronounce any attorney-in-fact incapable of signing an arbitration agreement—even if a power of attorney specifically authorized her to do so. See Tr. of Oral Arg. 27. (After all, such a rule would speak to only the contract’s formation.) And why stop there? If the respondents were right, States could just as easily declare everyone incompetent to sign arbitration agreements. (That rule too would address only formation.) The FAA would then mean nothing at all—its provisions rendered helpless to prevent even the most blatant discrimination against arbitration. III As we did just last Term, we once again “reach a conclusion that . . . falls well within the confines of (and goes no further than) present well-established law.” DIRECTV, 577 U. S., at ___ (slip op., at 10). The Kentucky Supreme Court specially impeded the ability of attorneys-in-fact to enter into arbitration agreements. The court thus flouted the FAA’s command to place those agreements on an equal footing with all other contracts. Our decision requires reversing the Kentucky Supreme Court’s judgment in favor of the Clark estate. As noted earlier, the state court held that the Clark power of attorney was sufficiently broad to cover executing an arbitration agreement. See supra, at 3. The court invalidated the agreement with Kindred only because the power of attorney did not specifically authorize Janis to enter into it on Olive’s behalf. In other words, the decision below was based exclusively on the clear-statement rule that we have held violates the FAA. So the court must now enforce the Clark-Kindred arbitration agreement. By contrast, our decision might not require such a result in the Wellner case. The Kentucky Supreme Court began its opinion by stating that the Wellner power of attorney was insufficiently broad to give Beverly the authority to execute an arbitration agreement for Joe. See supra, at 3. If that interpretation of the document is wholly independent of the court’s clear-statement rule, then nothing we have said disturbs it. But if that rule at all influenced the construction of the Wellner power of attorney, then the court must evaluate the document’s meaning anew. The court’s opinion leaves us uncertain as to whether such an impermissible taint occurred. We therefore vacate the judgment below and return the case to the state court for further consideration. See Marmet Health Care Center, Inc. v. Brown, 565 U. S. 530, 534 (2012) (per curiam) (vacating and remanding another arbitration decision because we could not tell “to what degree [an] alternative holding was influenced by” the state court’s erroneous, arbitration-specific rule). On remand, the court should determine whether it adheres, in the absence of its clear-statement rule, to its prior reading of the Wellner power of attorney. For these reasons, we reverse in part and vacate in part the judgment of the Kentucky Supreme Court, and we remand the case for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 Making matters worse, the Kentucky Supreme Court’s clear-statement rule appears not to apply to other kinds of agreements relinquishing the right to go to court or obtain a jury trial. Nothing in the decision below (or elsewhere in Kentucky law) suggests that explicit authorization is needed before an attorney-in-fact can sign a settlement agreement or consent to a bench trial on her principal’s behalf. See 478 S. W. 3d, at 325 (discussing the Wellner power of attorney’s provision for “managing a claim in litigation” without insisting that such commitments would require a clearer grant). Mark that as yet another indication that the court’s demand for specificity in powers of attorney arises from the suspect status of arbitration rather than the sacred status of jury trials. 2 We do not suggest that a state court is precluded from announcing a new, generally applicable rule of law in an arbitration case. We simply reiterate here what we have said many times before—that the rule must in fact apply generally, rather than single out arbitration.

581.US.2016_16-529

The Securities and Exchange Commission (SEC or Commission) possesses authority to investigate violations of federal securities laws and to commence enforcement actions in federal district court if its investigations uncover evidence of wrongdoing. Initially, the Commission’s statutory authority in enforcement actions was limited to seeking an injunction barring future violations. Beginning in the 1970’s, federal district courts, at the request of the Commission, began ordering disgorgement in SEC enforcement proceedings. Although Congress has since authorized the Commission to seek monetary civil penalties, the Commission has continued to seek disgorgement. This Court has held that 28 U. S. C. §2462, which establishes a 5-year limitations period for “an action, suit or proceeding for the enforcement of any civil fine, penalty, or forfeiture,” applies when the Commission seeks monetary civil penalties. See Gabelli v. SEC, 568 U. S. 442 . In 2009, the Commission brought an enforcement action, alleging that petitioner Charles Kokesh violated various securities laws by concealing the misappropriation of $34.9 million from four business-development companies from 1995 to 2009. The Commission sought monetary civil penalties, disgorgement, and an injunction barring Kokesh from future violations. After a jury found that Kokesh’s actions violated several securities laws, the District Court determined that §2462’s 5-year limitations period applied to the monetary civil penalties. With respect to the $34.9 million disgorgement judgment, however, the court concluded that §2462 did not apply because disgorgement is not a “penalty” within the meaning of the statute. The Tenth Circuit affirmed, holding that disgorgement was neither a penalty nor a forfeiture. Held: Because SEC disgorgement operates as a penalty under §2462, any claim for disgorgement in an SEC enforcement action must be commenced within five years of the date the claim accrued. Pp. 5–11. (a) The definition of “penalty” as a “punishment, whether corporal or pecuniary, imposed and enforced by the State, for a crime or offen[s]e against its laws,” Huntington v. Attrill, 146 U. S. 657 , gives rise to two principles. First, whether a sanction represents a penalty turns in part on “whether the wrong sought to be redressed is a wrong to the public, or a wrong to the individual.” Id., at 668. Second, a pecuniary sanction operates as a penalty if it is sought “for the purpose of punishment, and to deter others from offending in like manner” rather than to compensate victims. Ibid. This Court has applied these principles in construing the term “penalty,” holding, e.g., that a statute providing a compensatory remedy for a private wrong did not impose a “penalty,” Brady v. Daly, 175 U. S. 148 . Pp. 5–7. (b) The application of these principles here readily demonstrates that SEC disgorgement constitutes a penalty within the meaning of §2462. First, SEC disgorgement is imposed by the courts as a consequence for violating public laws, i.e., a violation committed against the United States rather than an aggrieved individual. Second, SEC disgorgement is imposed for punitive purposes. Sanctions imposed for the purpose of deterring infractions of public laws are inherently punitive because “deterrence [is] not [a] legitimate nonpunitive governmental objectiv[e].” Bell v. Wolfish, 441 U. S. 520 , n. 20. Finally, SEC disgorgement is often not compensatory. Disgorged profits are paid to the district courts, which have discretion to determine how the money will be distributed. They may distribute the funds to victims, but no statute commands them to do so. When an individual is made to pay a noncompensatory sanction to the government as a consequence of a legal violation, the payment operates as a penalty. See Porter v. Warner Holding Co., 328 U. S. 395 . Pp. 7–9. (c) The Government responds that SEC disgorgement is not punitive but a remedial sanction that operates to restore the status quo. It is not clear, however, that disgorgement simply returns the defendant to the place he would have occupied had he not broken the law. It sometimes exceeds the profits gained as a result of the violation. And, as demonstrated here, SEC disgorgement may be ordered without consideration of a defendant’s expenses that reduced the amount of illegal profit. In such cases, disgorgement does not simply restore the status quo; it leaves the defendant worse off and is therefore punitive. Although disgorgement may serve compensatory goals in some cases, “sanctions frequently serve more than one purpose.” Austin v. United States, 509 U. S. 602 . Because they “go beyond compensation, are intended to punish, and label defendants wrongdoers” as a consequence of violating public laws, Gabelli, 568 U. S., at 451–452, disgorgement orders represent a penalty and fall within §2462’s 5-year limitations period. Pp. 9–11. 834 F. 3d 1158, reversed. Sotomayor, J., delivered the opinion for a unanimous Court.

A 5-year statute of limitations applies to any “action, suit or proceeding for the enforcement of any civil fine, penalty, or forfeiture, pecuniary or otherwise.” 28 U. S. C. §2462. This case presents the question whether §2462 applies to claims for disgorgement imposed as a sanction for violating a federal securities law. The Court holds that it does. Disgorgement in the securities-enforcement context is a “penalty” within the meaning of §2462, and so disgorgement actions must be commenced within five years of the date the claim accrues. I A After rampant abuses in the securities industry led to the 1929 stock market crash and the Great Depression, Congress enacted a series of laws to ensure that “the highest ethical standards prevail in every facet of the securities industry.”[1] SEC v. Capital Gains Research Bureau, Inc., 375 U. S. 180 –187 (1963) (internal quotation marks omitted). The second in the series—the Securities Exchange Act of 1934—established the Securities and Exchange Commission (SEC or Commission) to enforce federal securities laws. Congress granted the Com-mission power to prescribe “ ‘rules and regulations . . . as necessary or appropriate in the public interest or for the protection of investors.’ ” Blue Chip Stamps v. Manor Drug Stores, 421 U. S. 723, 728 (1975) . In addition to rulemaking, Congress vested the Commission with “broad authority to conduct investigations into possible violations of the federal securities laws.” SEC v. Jerry T. O’Brien, Inc., 467 U. S. 735, 741 (1984) . If an investigation uncovers evidence of wrongdoing, the Commission may initiate enforcement actions in federal district court. Initially, the only statutory remedy available to the SEC in an enforcement action was an injunction barring future violations of securities laws. See 1 T. Hazen, Law of Securities Regulation §1:37 (7th ed., rev. 2016). In the absence of statutory authorization for monetary remedies, the Commission urged courts to order disgorgement as an exercise of their “inherent equity power to grant relief ancillary to an injunction.” SEC v. Texas Gulf Sulphur Co., 312 F. Supp. 77, 91 (SDNY 1970), aff’d in part and rev’d in part, 446 F. 2d 1301 (CA2 1971). Generally, disgorgement is a form of “[r]estitution measured by the defendant’s wrongful gain.” Restatement (Third) of Restitution and Unjust Enrichment §51, Comment a, p. 204 (2010) (Restatement (Third)). Disgorgement requires that the defendant give up “those gains . . . properly attribut-able to the defendant’s interference with the claimant’s legally protected rights.” Ibid. Beginning in the 1970’s, courts ordered disgorgement in SEC enforcement proceedings in order to “deprive . . . defendants of their profits in order to remove any monetary reward for violating” securities laws and to “protect the investing public by providing an effective deterrent to future violations.” Texas Gulf, 312 F. Supp., at 92. In 1990, as part of the Securities Enforcement Remedies and Penny Stock Reform Act, Congress authorized the Commission to seek monetary civil penalties. 104Stat. 932, codified at 15 U. S. C. §77t(d). The Act left the Commission with a full panoply of enforcement tools: It may promulgate rules, investigate violations of those rules and the securities laws generally, and seek monetary penal-ties and injunctive relief for those violations. In the years since the Act, however, the Commission has con-tinued its practice of seeking disgorgement in enforcement proceedings. This Court has already held that the 5-year statute of limitations set forth in 28 U. S. C. §2462 applies when the Commission seeks statutory monetary penalties. See Gabelli v. SEC, 568 U. S. 442, 454 (2013) . The question here is whether §2462, which applies to any “action, suit or proceeding for the enforcement of any civil fine, penalty, or forfeiture, pecuniary or otherwise,” also applies when the SEC seeks disgorgement. B Charles Kokesh owned two investment-adviser firms that provided investment advice to business-development companies. In late 2009, the Commission commenced an enforcement action in Federal District Court alleging that between 1995 and 2009, Kokesh, through his firms, misappropriated $34.9 million from four of those development companies. The Commission further alleged that, in order to conceal the misappropriation, Kokesh caused the filing of false and misleading SEC reports and proxy statements. The Commission sought civil monetary penalties, disgorgement, and an injunction barring Kokesh from violating securities laws in the future. After a 5-day trial, a jury found that Kokesh’s actions violated the Investment Company Act of 1940, 15 U. S. C. §80a–36; the Investment Advisers Act of 1940, 15 U. S. C. §§80b–5, 80b–6; and the Securities Exchange Act of 1934, 15 U. S. C. §§78m, 78n. The District Court then turned to the task of imposing penalties sought by the Commission. As to the civil monetary penalties, the District Court determined that §2462’s 5-year limitations period pre-cluded any penalties for misappropriation occurring prior to October 27, 2004—that is, five years prior to the date the Commission filed the complaint. App. to Pet. for Cert. 26a. The court ordered Kokesh to pay a civil penalty of $2,354,593, which represented “the amount of funds that [Kokesh] himself received during the limitations period.” Id., at 31a–32a. Regarding the Commission’s request for a $34.9 million disgorgement judgment—$29.9 million of which resulted from violations outside the limitations period—the court agreed with the Commission that because disgorgement is not a “penalty” within the meaning of §2462, no limitations period applied. The court therefore entered a disgorgement judgment in the amount of $34.9 million and ordered Kokesh to pay an additional $18.1 million in prejudgment interest. The Court of Appeals for the Tenth Circuit affirmed. 834 F. 3d 1158 (2016). It agreed with the District Court that disgorgement is not a penalty, and further found that disgorgement is not a forfeiture. Id., at 1164–1167. The court thus concluded that the statute of limitations in §2462 does not apply to SEC disgorgement claims. This Court granted certiorari, 580 U. S. ___ (2017), to resolve disagreement among the Circuits over whether disgorgement claims in SEC proceedings are subject to the 5-year limitations period of §2462.[2] II Statutes of limitations “se[t] a fixed date when exposure to the specified Government enforcement efforts en[d].” Gabelli, 568 U. S., at 448. Such limits are “ ‘vital to the welfare of society’ ” and rest on the principle that “ ‘even wrongdoers are entitled to assume that their sins may be forgotten.’ ” Id., at 449. The statute of limitations at issue here— 28 U. S. C. §2462—finds its roots in a law enacted nearly two centuries ago. 568 U. S., at 445. In its current form, §2462 establishes a 5-year limitations period for “an action, suit or proceeding for the enforcement of any civil fine, penalty, or forfeiture.” This limitations period applies here if SEC disgorgement qualifies as either a fine, penalty, or forfeiture. We hold that SEC disgorgement constitutes a penalty.[3] A A “penalty” is a “punishment, whether corporal or pecuniary, imposed and enforced by the State, for a crime or offen[s]e against its laws.” Huntington v. Attrill, 146 U. S. 657, 667 (1892) . This definition gives rise to two principles. First, whether a sanction represents a penalty turns in part on “whether the wrong sought to be redressed is a wrong to the public, or a wrong to the individual.” Id., at 668. Although statutes creating private causes of action against wrongdoers may appear—or even be labeled—penal, in many cases “neither the liability imposed nor the remedy given is strictly penal.” Id., at 667. This is because “[p]enal laws, strictly and properly, are those imposing punishment for an offense committed against the State.” Ibid. Second, a pecuniary sanction operates as a penalty only if it is sought “for the purpose of punishment, and to deter others from offending in like manner”—as opposed to compensating a victim for his loss. Id., at 668. The Court has applied these principles in construing the term “penalty.” In Brady v. Daly, 175 U. S. 148 (1899) , for example, a playwright sued a defendant in Federal Circuit Court under a statute providing that copyright infringers “ ‘shall be liable for damages . . . not less than one hundred dollars for the first [act of infringement], and fifty dollars for every subsequent performance, as to the court shall appear to be just.’ ” Id., at 153. The defendant argued that the Circuit Court lacked jurisdiction on the ground that a separate statute vested district courts with exclusive jurisdiction over actions “to recover a penalty.” Id., at 152. To determine whether the statutory damages represented a penalty, this Court noted first that the statute provided “for a recovery of damages for an act which violates the rights of the plaintiff, and gives the right of action solely to him” rather than the public generally, and second, that “the whole recovery is given to the proprietor, and the statute does not provide for a recovery by any other person.” Id., at 154, 156. By providing a compensatory remedy for a private wrong, the Court held, the statute did not impose a “penalty.” Id., at 154. Similarly, in construing the statutory ancestor of §2462, the Court utilized the same principles. In Meeker v. Lehigh Valley R. Co., 236 U. S. 412 –422 (1915), the Interstate Commerce Commission, a now-defunct federal agency charged with regulating railroads, ordered a railroad company to refund and pay damages to a shipping company for excessive shipping rates. The railroad company argued that the action was barred by Rev. Stat. §1047, Comp. Stat. 1913, §1712 (now 28 U. S. C. §2462), which imposed a 5-year limitations period upon any “ ‘suit or prosecution for a penalty or forfeiture, pecuniary or otherwise, accruing under the laws of the United States.’ ” 236 U. S., at 423. The Court rejected that argument, reasoning that “the words ‘penalty or forfeiture’ in [the statute] refer to something imposed in a punitive way for an infraction of a public law.” Ibid. A penalty, the Court held, does “not include a liability imposed [solely] for the purpose of redressing a private injury.” Ibid. Because the liability imposed was compensatory and paid entirely to a private plaintiff, it was not a “penalty” within the meaning of the statute of limitations. Ibid.; see also Gabelli, 568 U. S., at 451–452 (“[P]enalties” in the context of §2462 “go beyond compensation, are intended to punish, and label defendants wrongdoers”). B Application of the foregoing principles readily demonstrates that SEC disgorgement constitutes a penalty within the meaning of §2462. First, SEC disgorgement is imposed by the courts as a consequence for violating what we described in Meeker as public laws. The violation for which the remedy is sought is committed against the United States rather than an aggrieved individual—this is why, for example, a securities-enforcement action may proceed even if victims do not support or are not parties to the prosecution. As the Government concedes, “[w]hen the SEC seeks disgorgement, it acts in the public interest, to remedy harm to the public at large, rather than standing in the shoes of particular injured parties.” Brief for United States 22. Courts agree. See, e.g., SEC v. Rind, 991 F. 2d 1486, 1491 (CA9 1993) (“[D]isgorgement actions further the Commission’s public policy mission of protecting investors and safeguarding the integrity of the markets”); SEC v. Teo, 746 F. 3d 90, 102 (CA3 2014) (“[T]he SEC pursues [disgorgement] ‘independent of the claims of individual investors’ ” in order to “ ‘promot[e] economic and social policies’ ”). Second, SEC disgorgement is imposed for punitive purposes. In Texas Gulf—one of the first cases requiring disgorgement in SEC proceedings—the court emphasized the need “to deprive the defendants of their profits in order to . . . protect the investing public by providing an effective deterrent to future violations.” 312 F. Supp., at 92. In the years since, it has become clear that deterrence is not simply an incidental effect of disgorgement. Rather, courts have consistently held that “[t]he primary purpose of disgorgement orders is to deter violations of the securities laws by depriving violators of their ill-gotten gains.” SEC v. Fischbach Corp., 133 F. 3d 170, 175 (CA2 1997); see also SEC v. First Jersey Securities, Inc., 101 F. 3d 1450, 1474 (CA2 1996) (“The primary purpose of disgorgement as a remedy for violation of the securities laws is to deprive violators of their ill-gotten gains, thereby effectuating the deterrence objectives of those laws”); Rind, 991 F. 2d, at 1491 (“ ‘The deterrent effect of [an SEC] enforcement action would be greatly undermined if securities law violators were not required to disgorge illicit profits’ ”). Sanctions imposed for the purpose of deterring infractions of public laws are inherently punitive because “deterrence [is] not [a] legitimate nonpunitive governmental objectiv[e].” Bell v. Wolfish, 441 U. S. 520, 539, n. 20 (1979) ; see also United States v. Bajakajian, 524 U. S. 321, 329 (1998) (“Deterrence . . . has traditionally been viewed as a goal of punishment”). Finally, in many cases, SEC disgorgement is not compensatory. As courts and the Government have employed the remedy, disgorged profits are paid to the district court, and it is “within the court’s discretion to determine how and to whom the money will be distributed.” Fischbach Corp., 133 F. 3d, at 175. Courts have required disgorgement “regardless of whether the disgorged funds will be paid to such investors as restitution.” Id., at 176; see id., at 175 (“Although disgorged funds may often go to compensate securities fraud victims for their losses, such compensation is a distinctly secondary goal”). Some disgorged funds are paid to victims; other funds are dispersed to the United States Treasury. See, e.g., id., at 171 (affirming distribution of disgorged funds to Treasury where “no party before the court was entitled to the funds and . . . the persons who might have equitable claims were too dispersed for feasible identification and payment”); SEC v. Lund, 570 F. Supp. 1397, 1404–1405 (CD Cal. 1983) (ordering disgorgement and directing trustee to disperse funds to victims if “feasible” and to disperse any remaining money to the Treasury). Even though district courts may distribute the funds to the victims, they have not identified any statutory command that they do so. When an individual is made to pay a noncompensatory sanction to the Government as a consequence of a legal violation, the payment operates as a penalty. See Porter v. Warner Holding Co., 328 U. S. 395, 402 (1946) (distinguishing between restitution paid to an aggrieved party and penalties paid to the Government). SEC disgorgement thus bears all the hallmarks of a penalty: It is imposed as a consequence of violating a public law and it is intended to deter, not to compensate. The 5-year statute of limitations in §2462 therefore applies when the SEC seeks disgorgement. C The Government’s primary response to all of this is that SEC disgorgement is not punitive but “remedial” in that it “lessen[s] the effects of a violation” by “ ‘restor[ing] the status quo.’ ” Brief for Respondent 17. As an initial matter, it is not clear that disgorgement, as courts have applied it in the SEC enforcement context, simply returns the defendant to the place he would have occupied had he not broken the law. SEC disgorgement sometimes exceeds the profits gained as a result of the violation. Thus, for example, “an insider trader may be ordered to disgorge not only the unlawful gains that accrue to the wrongdoer directly, but also the benefit that accrues to third parties whose gains can be attributed to the wrongdoer’s conduct.” SEC v. Contorinis, 743 F. 3d 296, 302 (CA2 2014). Individuals who illegally provide confidential trading information have been forced to disgorge profits gained by individuals who received and traded based on that information—even though they never received any profits. Ibid; see also SEC v. Warde, 151 F. 3d 42, 49 (CA2 1998) (“A tippee’s gains are attributable to the tipper, regardless whether benefit accrues to the tipper”); SEC v. Clark, 915 F. 2d 439, 454 (CA9 1990) (“[I]t is well settled that a tipper can be required to disgorge his tippees’ profits”). And, as demonstrated by this case, SEC disgorgement sometimes is ordered without consideration of a defendant’s expenses that reduced the amount of illegal profit. App. to Pet. for Cert. 43a; see Restatement (Third) §51, Comment h, at 216 (“As a general rule, the defendant is entitled to a deduction for all marginal costs incurred in producing the revenues that are subject to disgorgement. Denial of an otherwise appropriate deduction, by making the defendant liable in excess of net gains, results in a punitive sanction that the law of restitution normally attempts to avoid”). In such cases, disgorgement does not simply restore the status quo; it leaves the defendant worse off. The justification for this practice given by the court below demonstrates that disgorgement in this context is a punitive, rather than a remedial, sanction: Disgorgement, that court explained, is intended not only to “prevent the wrongdoer’s unjust enrichment” but also “to deter others’ violations of the securities laws.” App. to Pet. for Cert. 43a. True, disgorgement serves compensatory goals in some cases; however, we have emphasized “the fact that sanctions frequently serve more than one purpose.” Austin v. United States, 509 U. S. 602, 610 (1993) . “ ‘A civil sanction that cannot fairly be said solely to serve a remedial purpose, but rather can only be explained as also serving either retributive or deterrent purposes, is punishment, as we have come to understand the term.’ ” Id., at 621; cf. Bajakajian, 524 U. S., at 331, n. 6 (“[A] modern statutory forfeiture is a ‘fine’ for Eighth Amendment purposes if it constitutes punishment even in part”). Because disgorgement orders “go beyond compensation, are intended to punish, and label defendants wrongdoers” as a consequence of violating public laws, Gabelli, 568 U. S., at 451–452, they represent a penalty and thus fall within the 5-year statute of limitations of §2462. III Disgorgement, as it is applied in SEC enforcement proceedings, operates as a penalty under §2462. Accordingly, any claim for disgorgement in an SEC enforcement action must be commenced within five years of the date the claim accrued. The judgment of the Court of Appeals for the Tenth Circuit is reversed. It is so ordered.Notes 1 Each of these statutes—the Securities Act of 1933, 15 U. S. C. §77a et seq.; the Securities Exchange Act of 1934, 15 U. S. C. §78a et seq.; the Public Utility Holding Company Act of 1935, 15 U. S. C. §79 et seq.; the Trust Indenture Act of 1939, 15 U. S. C. §77aaa et seq.; the Investment Company Act of 1940, 15 U. S. C. §80a–1 et seq.; and the Investment Advisers Act of 1940, 15 U. S. C. §80b–1 et seq.—serves the “fundamental purpose” of “substitut[ing] a philosophy of full disclosure for the philosophy of caveat emptor and thus . . . achiev[ing] a high standard of business ethics in the securities industry.” SEC v. Capital Gains Research Bureau, Inc., 375 U. S. 180, 186 (1963) . 2 Compare SEC v. Graham, 823 F. 3d 1357, 1363 (CA11 2016) (holding that §2462 applies to SEC disgorgement claims), with Riordan v. SEC, 627 F. 3d 1230, 1234 (CADC 2010) (holding that §2462 does not apply to SEC disgorgement claims). 3 Nothing in this opinion should be interpreted as an opinion on whether courts possess authority to order disgorgement in SEC enforcement proceedings or on whether courts have properly applied disgorgement principles in this context The sole question presented in this case is whether disgorgement, as applied in SEC enforcement actions, is subject to §2462’s limitations period.

582.US.2016_16-327

Petitioner Jae Lee moved to the United States from South Korea with his parents when he was 13. In the 35 years he has spent in this country, he has never returned to South Korea, nor has he become a U. S. citizen, living instead as a lawful permanent resident. In 2008, federal officials received a tip from a confidential informant that Lee had sold the informant ecstasy and marijuana. After obtaining a warrant, the officials searched Lee’s house, where they found drugs, cash, and a loaded rifle. Lee admitted that the drugs were his, and a grand jury indicted him on one count of possessing ecstasy with intent to distribute. Lee retained counsel and entered into plea discussions with the Government. During the plea process, Lee repeatedly asked his attorney whether he would face deportation; his attorney assured him that he would not be deported as a result of pleading guilty. Based on that assurance, Lee accepted a plea and was sentenced to a year and a day in prison. Lee had in fact pleaded guilty to an “aggravated felony” under the Immigration and Nationality Act, 8 U. S. C. §1101(a)(43)(B), so he was, contrary to his attorney’s advice, subject to mandatory deportation as a result of that plea. See §1227(a)(2)(A)(iii). When Lee learned of this consequence, he filed a motion to vacate his conviction and sentence, arguing that his attorney had provided constitutionally ineffective assistance. At an evidentiary hearing, both Lee and his plea-stage counsel testified that “deportation was the determinative issue” to Lee in deciding whether to accept a plea, and Lee’s counsel acknowledged that although Lee’s defense to the charge was weak, if he had known Lee would be deported upon pleading guilty, he would have advised him to go to trial. A Magistrate Judge recommended that Lee’s plea be set aside and his conviction vacated. The District Court, however, denied relief, and the Sixth Circuit affirmed. Applying the two-part test for ineffective assistance claims from Strickland v. Washington, 466 U. S. 668 , the Sixth Circuit concluded that, while the Government conceded that Lee’s counsel had performed deficiently, Lee could not show that he was prejudiced by his attorney’s erroneous advice. Held: Lee has demonstrated that he was prejudiced by his counsel’s erroneous advice. Pp. 5–13. (a) When a defendant claims that his counsel’s deficient performance deprived him of a trial by causing him to accept a plea, the defendant can show prejudice by demonstrating a “reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” Hill v. Lockhart, 474 U. S. 52 . Lee contends that he can make this showing because he never would have accepted a guilty plea had he known the result would be deportation. The Government contends that Lee cannot show prejudice from accepting a plea where his only hope at trial was that something unexpected and unpredictable might occur that would lead to acquittal. Pp. 5–8. (b) The Government makes two errors in urging the adoption of a per se rule that a defendant with no viable defense cannot show prejudice from the denial of his right to trial. First, it forgets that categorical rules are ill suited to an inquiry that demands a “case-by-case examination” of the “totality of the evidence.” Williams v. Taylor, 529 U. S. 362 (internal quotation marks omitted); Strickland, 466 U. S., at 695. More fundamentally, it overlooks that the Hill v. Lockhart inquiry focuses on a defendant’s decisionmaking, which may not turn solely on the likelihood of conviction after trial. The decision whether to plead guilty also involves assessing the respective consequences of a conviction after trial and by plea. See INS v. St. Cyr, 533 U. S. 289 –323. When those consequences are, from the defendant’s perspective, similarly dire, even the smallest chance of success at trial may look attractive. For Lee, deportation after some time in prison was not meaningfully different from deportation after somewhat less time; he says he accordingly would have rejected any plea leading to deportation in favor of throwing a “Hail Mary” at trial. Pointing to Strickland, the Government urges that “[a] defendant has no entitlement to the luck of a lawless decisionmaker.” 466 U. S., at 695. That statement, however, was made in the context of discussing the presumption of reliability applied to judicial proceedings, which has no place where, as here, a defendant was deprived of a proceeding altogether. When the inquiry is focused on what an individual defendant would have done, the possibility of even a highly improbable result may be pertinent to the extent it would have affected the defendant’s decisionmaking. Pp. 8–10. (c) Courts should not upset a plea solely because of post hoc assertions from a defendant about how he would have pleaded but for his attorney’s deficiencies. Rather, they should look to contemporaneous evidence to substantiate a defendant’s expressed preferences. In the unusual circumstances of this case, Lee has adequately demonstrated a reasonable probability that he would have rejected the plea had he known that it would lead to mandatory deportation: Both Lee and his attorney testified that “deportation was the determinative issue” to Lee; his responses during his plea colloquy confirmed the importance he placed on deportation; and he had strong connections to the United States, while he had no ties to South Korea. The Government argues that Lee cannot “convince the court that a decision to reject the plea bargain would have been rational under the circumstances,” Padilla v. Kentucky, 559 U. S. 356 , since deportation would almost certainly result from a trial. Unlike the Government, this Court cannot say that it would be irrational for someone in Lee’s position to risk additional prison time in exchange for holding on to some chance of avoiding deportation. Pp. 10–13. 825 F. 3d 311, reversed and remanded. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion, in which Alito, J., joined except as to Part I. Gorsuch, J., took no part in the consideration or decision of the case.

Petitioner Jae Lee was indicted on one count of possessing ecstasy with intent to distribute. Although he has lived in this country for most of his life, Lee is not a United States citizen, and he feared that a criminal conviction might affect his status as a lawful permanent resident. His attorney assured him there was nothing to worry about—the Government would not deport him if he pleaded guilty. So Lee, who had no real defense to the charge, opted to accept a plea that carried a lesser prison sentence than he would have faced at trial. Lee’s attorney was wrong: The conviction meant that Lee was subject to mandatory deportation from this country. Lee seeks to vacate his conviction on the ground that, in accepting the plea, he received ineffective assistance of counsel in violation of the Sixth Amendment. Everyone agrees that Lee received objectively unreasonable representation. The question presented is whether he can show he was prejudiced as a result. I Jae Lee moved to the United States from South Korea in 1982. He was 13 at the time. His parents settled the family in New York City, where they opened a small coffee shop. After graduating from a business high school in Manhattan, Lee set out on his own to Memphis, Tennessee, where he started working at a restaurant. After three years, Lee decided to try his hand at running a business. With some assistance from his family, Lee opened the Mandarin Palace Chinese Restaurant in a Memphis suburb. The Mandarin was a success, and Lee eventually opened a second restaurant nearby. In the 35 years he has spent in the country, Lee has never returned to South Korea. He did not become a United States citizen, living instead as a lawful permanent resident. At the same time he was running his lawful businesses, Lee also engaged in some illegitimate activity. In 2008, a confidential informant told federal officials that Lee had sold the informant approximately 200 ecstasy pills and two ounces of hydroponic marijuana over the course of eight years. The officials obtained a search warrant for Lee’s house, where they found 88 ecstasy pills, three Valium tablets, $32,432 in cash, and a loaded rifle. Lee admitted that the drugs were his and that he had given ecstasy to his friends. A grand jury indicted Lee on one count of possessing ecstasy with intent to distribute in violation of 21 U. S. C. §841(a)(1). Lee retained an attorney and entered into plea discussions with the Government. The attorney advised Lee that going to trial was “very risky” and that, if he pleaded guilty, he would receive a lighter sentence than he would if convicted at trial. App. 167. Lee informed his attorney of his noncitizen status and repeatedly asked him whether he would face deportation as a result of the criminal proceedings. The attorney told Lee that he would not be deported as a result of pleading guilty. Lee v. United States, 825 F. 3d 311, 313 (CA6 2016). Based on that assurance, Lee accepted the plea and the District Court sentenced him to a year and a day in prison, though it deferred commencement of Lee’s sentence for two months so that Lee could manage his restaurants over the holiday season. Lee quickly learned, however, that a prison term was not the only consequence of his plea. Lee had pleaded guilty to what qualifies as an “aggravated felony” under the Immigration and Nationality Act, and a noncitizen convicted of such an offense is subject to mandatory deportation. See 8 U. S. C. §§1101(a)(43)(B), 1227(a)(2)(A)(iii); Calcano-Martinez v. INS, 533 U. S. 348 , n. 1 (2001). Upon learning that he would be deported after serving his sentence, Lee filed a motion under 28 U. S. C. §2255 to vacate his conviction and sentence, arguing that his attorney had provided constitutionally ineffective assistance. At an evidentiary hearing on Lee’s motion, both Lee and his plea-stage counsel testified that “deportation was the determinative issue in Lee’s decision whether to accept the plea.” Report and Recommendation in No. 2:10–cv–02698 (WD Tenn.), pp. 6–7 (Report and Recommendation). In fact, Lee explained, his attorney became “pretty upset because every time something comes up I always ask about immigration status,” and the lawyer “always said why [are you] worrying about something that you don’t need to worry about.” App. 170. According to Lee, the lawyer assured him that if deportation was not in the plea agreement, “the government cannot deport you.” Ibid. Lee’s attorney testified that he thought Lee’s case was a “bad case to try” because Lee’s defense to the charge was weak. Id., at 218–219. The attorney nonetheless acknowledged that if he had known Lee would be deported upon pleading guilty, he would have advised him to go to trial. Id., at 236, 244. Based on the hearing testimony, a Magistrate Judge recommended that Lee’s plea be set aside and his conviction vacated because he had received ineffective assistance of counsel. The District Court, however, denied relief. Applying our two-part test for ineffective assistance claims from Strickland v. Washington, 466 U. S. 668 (1984) , the District Court concluded that Lee’s counsel had performed deficiently by giving improper advice about the deportation consequences of the plea. But, “[i]n light of the overwhelming evidence of Lee’s guilt,” Lee “would have almost certainly” been found guilty and received “a significantly longer prison sentence, and subsequent deportation,” had he gone to trial. Order in No. 2:10–cv–02698 (WD Tenn.), p. 24 (Order). Lee therefore could not show he was prejudiced by his attorney’s erroneous advice. Viewing its resolution of the issue as debatable among jurists of reason, the District Court granted a certificate of appealability. The Court of Appeals for the Sixth Circuit affirmed the denial of relief. On appeal, the Government conceded that the performance of Lee’s attorney had been deficient. To establish that he was prejudiced by that deficient performance, the court explained, Lee was required to show “a reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” 825 F. 3d, at 313 (quoting Hill v. Lockhart, 474 U. S. 52, 59 (1985) ; internal quotation marks omitted). Lee had “no bona fide defense, not even a weak one,” so he “stood to gain nothing from going to trial but more prison time.” 825 F. 3d, at 313, 316. Relying on Circuit precedent holding that “no rational defendant charged with a deportable offense and facing overwhelming evidence of guilt would proceed to trial rather than take a plea deal with a shorter prison sentence,” the Court of Appeals concluded that Lee could not show prejudice. Id., at 314 (internal quotation marks omitted). We granted certiorari. 580 U. S. ___ (2016). II The Sixth Amendment guarantees a defendant the effective assistance of counsel at “critical stages of a criminal proceeding,” including when he enters a guilty plea. Lafler v. Cooper, 566 U. S. 156, 165 (2012) ; Hill, 474 U. S., at 58. To demonstrate that counsel was constitutionally ineffective, a defendant must show that counsel’s representation “fell below an objective standard of reasonableness” and that he was prejudiced as a result. Strickland, 466 U. S., at 688, 692. The first requirement is not at issue in today’s case: The Government concedes that Lee’s plea-stage counsel provided inadequate representation when he assured Lee that he would not be deported if he pleaded guilty. Brief for United States 15. The question is whether Lee can show he was prejudiced by that erroneous advice. A A claim of ineffective assistance of counsel will often involve a claim of attorney error “during the course of a legal proceeding”—for example, that counsel failed to raise an objection at trial or to present an argument on appeal. Roe v. Flores-Ortega, 528 U. S. 470, 481 (2000) . A defendant raising such a claim can demonstrate prejudice by showing “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id., at 482 (quoting Strickland, 466 U. S., at 694; internal quotation marks omitted). But in this case counsel’s “deficient performance arguably led not to a judicial proceeding of disputed reliability, but rather to the forfeiture of a proceeding itself.” Flores-Ortega, 528 U. S., at 483. When a defendant alleges his counsel’s deficient performance led him to accept a guilty plea rather than go to trial, we do not ask whether, had he gone to trial, the result of that trial “would have been different” than the result of the plea bargain. That is because, while we ordinarily “apply a strong presumption of reliability to judicial proceedings,” “we cannot accord” any such presumption “to judicial proceedings that never took place.” Id., at 482–483 (internal quotation marks omitted). We instead consider whether the defendant was prejudiced by the “denial of the entire judicial proceeding . . . to which he had a right.” Id., at 483. As we held in Hill v. Lockhart, when a defendant claims that his counsel’s deficient performance deprived him of a trial by causing him to accept a plea, the defendant can show prejudice by demonstrating a “reasonable probability that, but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” 474 U. S., at 59. The dissent contends that a defendant must also show that he would have been better off going to trial. That is true when the defendant’s decision about going to trial turns on his prospects of success and those are affected by the attorney’s error—for instance, where a defendant alleges that his lawyer should have but did not seek to suppress an improperly obtained confession. Premo v. Moore, 562 U. S. 115, 118 (2011) ; cf., e.g., Hill, 474 U. S., at 59 (discussing failure to investigate potentially exculpatory evidence). Not all errors, however, are of that sort. Here Lee knew, correctly, that his prospects of acquittal at trial were grim, and his attorney’s error had nothing to do with that. The error was instead one that affected Lee’s understanding of the consequences of pleading guilty. The Court confronted precisely this kind of error in Hill. See id., at 60 (“the claimed error of counsel is erroneous advice as to eligibility for parole”). Rather than asking how a hypothetical trial would have played out absent the error, the Court considered whether there was an adequate showing that the defendant, properly advised, would have opted to go to trial. The Court rejected the defendant’s claim because he had “alleged no special circumstances that might support the conclusion that he placed particular emphasis on his parole eligibility in deciding whether or not to plead guilty.” Ibid.[1] Lee, on the other hand, argues he can establish prejudice under Hill because he never would have accepted a guilty plea had he known that he would be deported as a result. Lee insists he would have gambled on trial, risking more jail time for whatever small chance there might be of an acquittal that would let him remain in the United States.[2] The Government responds that, since Lee had no viable defense at trial, he would almost certainly have lost and found himself still subject to deportation, with a lengthier prison sentence to boot. Lee, the Government contends, cannot show prejudice from accepting a plea where his only hope at trial was that something unexpected and unpredictable might occur that would lead to an acquittal. B The Government asks that we, like the Court of Appeals below, adopt a per se rule that a defendant with no viable defense cannot show prejudice from the denial of his right to trial. Brief for United States 26. As a general matter, it makes sense that a defendant who has no realistic defense to a charge supported by sufficient evidence will be unable to carry his burden of showing prejudice from accepting a guilty plea. But in elevating this general proposition to a per se rule, the Government makes two errors. First, it forgets that categorical rules are ill suited to an inquiry that we have emphasized demands a “case-by-case examination” of the “totality of the evidence.” Williams v. Taylor, 529 U. S. 362, 391 (2000) (internal quotation marks omitted); Strickland, 466 U. S., at 695. And, more fundamentally, the Government overlooks that the inquiry we prescribed in Hill v. Lockhart focuses on a defendant’s decisionmaking, which may not turn solely on the likelihood of conviction after trial. A defendant without any viable defense will be highly likely to lose at trial. And a defendant facing such long odds will rarely be able to show prejudice from accepting a guilty plea that offers him a better resolution than would be likely after trial. But that is not because the prejudice inquiry in this context looks to the probability of a conviction for its own sake. It is instead because defendants obviously weigh their prospects at trial in deciding whether to accept a plea. See Hill, 474 U. S., at 59. Where a defendant has no plausible chance of an acquittal at trial, it is highly likely that he will accept a plea if the Government offers one. But common sense (not to mention our precedent) recognizes that there is more to consider than simply the likelihood of success at trial. The decision whether to plead guilty also involves assessing the respective consequences of a conviction after trial and by plea. See INS v. St. Cyr, 533 U. S. 289 –323 (2001). When those consequences are, from the defendant’s perspective, similarly dire, even the smallest chance of success at trial may look attractive. For example, a defendant with no realistic defense to a charge carrying a 20-year sentence may nevertheless choose trial, if the prosecution’s plea offer is 18 years. Here Lee alleges that avoiding deportation was the determinative factor for him; deportation after some time in prison was not meaningfully different from deportation after somewhat less time. He says he accordingly would have rejected any plea leading to deportation—even if it shaved off prison time—in favor of throwing a “Hail Mary” at trial. The Government urges that, in such circumstances, the possibility of an acquittal after trial is “irrelevant to the prejudice inquiry,” pointing to our statement in Strickland that “[a] defendant has no entitlement to the luck of a lawless decisionmaker.” 466 U. S., at 695. That statement, however, was made in the context of discussing the presumption of reliability we apply to judicial proceedings. As we have explained, that presumption has no place where, as here, a defendant was deprived of a proceeding altogether. Flores-Ortega, 528 U. S., at 483. In a presumptively reliable proceeding, “the possibility of arbitrariness, whimsy, caprice, ‘nullification,’ and the like” must by definition be ignored. Strickland, 466 U. S., at 695. But where we are instead asking what an individual defendant would have done, the possibility of even a highly improbable result may be pertinent to the extent it would have affected his decisionmaking.[3] C “Surmounting Strickland’s high bar is never an easy task,” Padilla v. Kentucky, 559 U. S. 356, 371 (2010) , and the strong societal interest in finality has “special force with respect to convictions based on guilty pleas.” United States v. Timmreck, 441 U. S. 780, 784 (1979) . Courts should not upset a plea solely because of post hoc assertions from a defendant about how he would have pleaded but for his attorney’s deficiencies. Judges should instead look to contemporaneous evidence to substantiate a defendant’s expressed preferences. In the unusual circumstances of this case, we conclude that Lee has adequately demonstrated a reasonable probability that he would have rejected the plea had he known that it would lead to mandatory deportation. There is no question that “deportation was the determinative issue in Lee’s decision whether to accept the plea deal.” Report and Recommendation, at 6–7; see also Order, at 14 (noting Government did not dispute testimony to this effect). Lee asked his attorney repeatedly whether there was any risk of deportation from the proceedings, and both Lee and his attorney testified at the evidentiary hearing below that Lee would have gone to trial if he had known about the deportation consequences. See Report and Recommendation, at 12 (noting “the undisputed fact that had Lee at all been aware that deportation was possible as a result of his guilty plea, he would . . . not have pled guilty”), adopted in relevant part in Order, at 15. Lee demonstrated as much at his plea colloquy: When the judge warned him that a conviction “could result in your being deported,” and asked “[d]oes that at all affect your decision about whether you want to plead guilty or not,” Lee answered “Yes, Your Honor.” App. 103. When the judge inquired “[h]ow does it affect your decision,” Lee responded “I don’t understand,” and turned to his attorney for advice. Ibid. Only when Lee’s counsel assured him that the judge’s statement was a “standard warning” was Lee willing to proceed to plead guilty. Id., at 210.[4] There is no reason to doubt the paramount importance Lee placed on avoiding deportation. Deportation is always “a particularly severe penalty,” Padilla, 559 U. S., at 365 (internal quotation marks omitted), and we have “recognized that ‘preserving the client’s right to remain in the United States may be more important to the client than any potential jail sentence,’ ” id., at 368 (quoting St. Cyr, 533 U. S., at 322; alteration and some internal quotation marks omitted); see also Padilla, 559 U. S., at 364 (“[D]eportation is an integral part—indeed, sometimes the most important part—of the penalty that may be imposed on noncitizen defendants who plead guilty to specified crimes.” (footnote omitted)). At the time of his plea, Lee had lived in the United States for nearly three decades, had established two businesses in Tennessee, and was the only family member in the United States who could care for his elderly parents—both naturalized American citizens. In contrast to these strong connections to the United States, there is no indication that he had any ties to South Korea; he had never returned there since leaving as a child. The Government argues, however, that under Padilla v. Kentucky, a defendant “must convince the court that a decision to reject the plea bargain would have been rational under the circumstances.” Id., at 372. The Government contends that Lee cannot make that showing because he was going to be deported either way; going to trial would only result in a longer sentence before that inevitable consequence. See Brief for United States 13, 21–23. We cannot agree that it would be irrational for a defendant in Lee’s position to reject the plea offer in favor of trial. But for his attorney’s incompetence, Lee would have known that accepting the plea agreement would certainly lead to deportation. Going to trial? Almost certainly. If deportation were the “determinative issue” for an individual in plea discussions, as it was for Lee; if that individual had strong connections to this country and no other, as did Lee; and if the consequences of taking a chance at trial were not markedly harsher than pleading, as in this case, that “almost” could make all the difference. Balanced against holding on to some chance of avoiding deportation was a year or two more of prison time. See id., at 6. Not everyone in Lee’s position would make the choice to reject the plea. But we cannot say it would be irrational to do so. Lee’s claim that he would not have accepted a plea had he known it would lead to deportation is backed by substantial and uncontroverted evidence. Accordingly we conclude Lee has demonstrated a “reasonable probability that, but for [his] counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.” Hill, 474 U. S., at 59. * * * The judgment of the United States Court of Appeals for the Sixth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 The dissent also relies heavily on Missouri v. Frye, 566 U. S. 134 (2012) , and Lafler v. Cooper, 566 U. S. 156 (2012) . Those cases involved defendants who alleged that, but for their attorney’s incompetence, they would have accepted a plea deal—not, as here and as in Hill, that they would have rejected a plea. In both Frye and Lafler, the Court highlighted this difference: Immediately following the sentence that the dissent plucks from Frye, post, at 5 (opinion of Thomas, J.), the Court explained that its “application of Strickland to the instances of an uncommunicated, lapsed plea does nothing to alter the standard laid out in Hill.” 566 U. S., at 148 (“Hill was correctly decided and applies in the context in which it arose”). Lafler, decided the same day as Frye, reiterated that “[i]n contrast to Hill, here the ineffective advice led not to an offer’s acceptance but to its rejection.” 566 U. S., at 163. Frye and Lafler articulated a different way to show prejudice, suited to the context of pleas not accepted, not an additional element to the Hill inquiry. See Frye, 566 U. S., at 148 (“Hill does not . . . provide the sole means for demonstrating prejudice arising from the deficient performance of counsel during plea negotiations”). Contrary to the dissent’s assertion, post, at 8–9, we do not depart from Strickland’s requirement of prejudice. The issue is how the required prejudice may be shown. 2 Lee also argues that he can show prejudice because, had his attorney advised him that he would be deported if he accepted the Government’s plea offer, he would have bargained for a plea deal that did not result in certain deportation. Given our conclusion that Lee can show prejudice based on the reasonable probability that he would have gone to trial, we need not reach this argument. 3 The dissent makes much of the fact that Hill v. Lockhart, 474 U. S. 52 (1985) , also noted that courts should ignore the “idiosyncrasies of the particular decisionmaker.” Post, at 7 (quoting Hill, 474 U. S., at 60; internal quotation marks omitted). But Hill made this statement in discussing how courts should analyze “predictions of the outcome at a possible trial.” Id., at 59–60. As we have explained, assessing the effect of some types of attorney errors on defendants’ decisionmaking involves such predictions: Where an attorney error allegedly affects how a trial would have played out, we analyze that error’s effects on a defendant’s decisionmaking by making a prediction of the likely trial outcome. But, as Hill recognized, such predictions will not always be “necessary.” Id., at 60. Such a prediction is neither necessary nor appropriate where, as here, the error is one that is not alleged to be pertinent to a trial outcome, but is instead alleged to have affected a defendant’s understanding of the consequences of his guilty plea. 4 Several courts have noted that a judge’s warnings at a plea colloquy may undermine a claim that the defendant was prejudiced by his attorney’s misadvice. See, e.g., United States v. Newman, 805 F. 3d 1143, 1147 (CADC 2015); United States v. Kayode, 777 F. 3d 719, 728–729 (CA5 2014); United States v. Akinsade, 686 F. 3d 248, 253 (CA4 2012); Boyd v. Yukins, 99 Fed. Appx. 699, 705 (CA6 2004). The present case involves a claim of ineffectiveness of counsel extending to advice specifically undermining the judge’s warnings themselves, which the defendant contemporaneously stated on the record he did not understand. There has been no suggestion here that the sentencing judge’s statements at the plea colloquy cured any prejudice from the erroneous advice of Lee’s counsel.

581.US.2016_15-1500

Petitioners Brian and Michelle Lewis were driving on a Connecticut interstate when they were struck from behind by a vehicle driven by respondent William Clarke, a Mohegan Tribal Gaming Authority employee, who was transporting Mohegan Sun Casino patrons. The Lewises sued Clarke in his individual capacity in state court. Clarke moved to dismiss for lack of subject-matter jurisdiction, arguing that because he was an employee of the Gaming Authority—an arm of the Mohegan Tribe entitled to sovereign immunity—and was acting within the scope of his employment at the time of the accident, he was similarly entitled to sovereign immunity against suit. He also argued, in the alternative, that he should prevail because the Gaming Authority was bound by tribal law to indemnify him. The trial court denied Clarke’s motion, but the Supreme Court of Connecticut reversed, holding that tribal sovereign immunity barred the suit because Clarke was acting within the scope of his employment when the accident occurred. It did not consider whether Clarke should be entitled to sovereign immunity based on the indemnification statute. Held: 1. In a suit brought against a tribal employee in his individual capacity, the employee, not the tribe, is the real party in interest and the tribe’s sovereign immunity is not implicated. Pp. 5–8. (a) In the context of lawsuits against state and federal employees or entities, courts look to whether the sovereign is the real party in interest to determine whether sovereign immunity bars the suit, see Hafer v. Melo, 502 U. S. 21 . A defendant in an official-capacity action—where the relief sought is only nominally against the official and in fact is against the official’s office and thus the sovereign itself—may assert sovereign immunity. Kentucky v. Graham, 473 U. S. 159 . But an officer in an individual-capacity action—which seeks “to impose individual liability upon a government officer for actions taken under color of state law,” Hafer, 502 U. S., at 25—may be able to assert personal immunity defenses but not sovereign immunity, id., at 30–31. The Court does not reach Clarke’s argument that he is entitled to the personal immunity defense of official immunity, which Clarke raised for the first time on appeal. Pp. 5–7. (b) Applying these general rules in the context of tribal sovereign immunity, it is apparent that they foreclose Clarke’s sovereign immunity defense. This action arises from a tort committed by Clarke on a Connecticut interstate and is simply a suit against Clarke to recover for his personal actions. Clarke, not the Gaming Authority, is the real party in interest. The State Supreme Court extended sovereign immunity for tribal employees beyond what common-law sovereign immunity principles would recognize for either state or federal employees. Pp. 7–8. 2. An indemnification provision cannot, as a matter of law, extend sovereign immunity to individual employees who would otherwise not fall under its protective cloak. Pp. 8–12. (a) This conclusion follows naturally from the principles discussed above and previously applied to the different question whether a state instrumentality may invoke the State’s immunity from suit even when the Federal Government has agreed to indemnify that instrumentality against adverse judgments, Regents of Univ. of Cal. v. Doe, 519 U. S. 425 . There, this Court held that the indemnification provision did not divest the state instrumentality of Eleventh Amendment immunity, and its analysis turned on where the potential legal liability lay, not from whence the money to pay the damages award ultimately came. Here, the Connecticut courts exercise no jurisdiction over the Tribe or Gaming Authority, and their judgments will not bind the Tribe or its instrumentalities in any way. Moreover, indemnification is not a certainty, because Clarke will not be indemnified should the Gaming Authority determine that he engaged in “wanton, reckless, or malicious” activity. Mohegan Tribe Code §4–52. Pp. 8–10. (b) Courts have extended sovereign immunity to private healthcare insurance companies under certain circumstances, but those cases rest on the proposition that the fiscal intermediaries are essentially state instrumentalities, and Clarke offers no persuasive reason to depart from precedent and treat a lawsuit against an individual employee as one against a state instrumentality. Similarly, this Court has never held that a civil rights suit under 42 U. S. C. §1983 against a state officer in his individual capacity implicates the Eleventh Amendment and a State’s sovereign immunity from suit. Finally, this Court’s conclusion that indemnification provisions do not alter the real-party-in-interest analysis for sovereign immunity purposes is consistent with the practice that applies in the contexts of diversity of citizenship and joinder. Pp. 10–12. 320 Conn.706, 135 A. 3d 677, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Alito, and Kagan, JJ., joined. Thomas, J., and Ginsburg, J., filed opinions concurring in the judgment. Gorsuch, J., took no part in the consideration or decision of the case.

Indian tribes are generally entitled to immunity from suit. This Court has considered the scope of that immu-nity in a number of circumstances. This case presents an ordinary negligence action brought against a tribal employee in state court under state law. We granted certiorari to resolve whether an Indian tribe’s sovereign immu-nity bars individual-capacity damages actions against tribal employees for torts committed within the scope of their employment and for which the employees are indemnified by the tribe. We hold that, in a suit brought against a tribal employee in his individual capacity, the employee, not the tribe, is the real party in interest and the tribe’s sovereign immu-nity is not implicated. That an employee was acting within the scope of his employment at the time the tort was committed is not, on its own, sufficient to bar a suit against that employee on the basis of tribal sovereign immunity. We hold further that an indemnification provision does not extend a tribe’s sovereign immunity where it otherwise would not reach. Accordingly, we reverse and remand. I A The Mohegan Tribe of Indians of Connecticut traces its lineage back centuries. Originally part of the Lenni Lenape, the Tribe formed the independent Mohegan Tribe under the leadership of Sachem Uncas in the early 1600’s. M. Fawcett, The Lasting of the Mohegans 7, 11–13 (1995). In 1994, in accordance with the petition procedures established by the Bureau of Indian Affairs, the Tribe attained federal recognition.[1] See 59 Fed. Reg. 12140 (1994); Mohegan Const., Preamble and Art. II. As one means of maintaining its economic self-sufficiency, the Tribe entered into a Gaming Compact with the State of Connecticut pursuant to the Indian Gaming Regulatory Act, 102Stat. 2467, 25 U. S. C. §2701 et seq. The compact authorizes the Tribe to conduct gaming on its land, subject to certain conditions including establishment of the Gaming Disputes Court. See 59 Fed. Reg. 65130 (approving the Tribal-State Compact Between the Mohegan Indian Tribe and the State of Connecticut (May 17, 1994)); Mohegan Const., Art. XIII, §2; Mohegan Tribe Code 3–248(a) (Supp. 2016). The Mohegan Tribal Gaming Authority, an arm of the Tribe, exercises the powers of the Mohegan Tribe over tribal gaming activities. Mohegan Const., Art. XIII, §1; Mohegan Tribe Code §2–21. Of particular relevance here, Mohegan law sets out sovereign immunity and indemnification policies applicable to disputes arising from gaming activities. The Gaming Authority has waived its sovereign immunity and consented to be sued in the Mohegan Gaming Disputes Court. Mohegan Const., Art. XIII, §1; Mohegan Tribe Code §3–250(b). Neither the Tribe nor the Gaming Authority has consented to suit for claims arising under Connecticut state law. See Mohegan Const., Art. IX, §2(t); Mohegan Tribe Code §3–250(g); see also Blatchford v. Native Village of Noatak, 501 U. S. 775, 782 (1991) (observing that Indian tribes have not surrendered their immunity against suits by States). Further, Mohegan Tribe Code §4–52 provides that the Gaming Authority “shall save harmless and indemnify its Officer or Em-ployee from financial loss and expense arising out of any claim, demand, or suit by reason of his or her alleged negligence . . . if the Officer or Employee is found to have been acting in the discharge of his or her duties or within the scope of his or her employment.” The Gaming Authority does not indemnify employees who engage in “wanton, reckless or malicious” activity. Mohegan Tribe Code§4–52. B Petitioners Brian and Michelle Lewis were driving down Interstate 95 in Norwalk, Connecticut, when a limousine driven by respondent William Clarke hit their vehicle from behind. Clarke, a Gaming Authority employee, was transporting patrons of the Mohegan Sun Casino to their homes. For purposes of this appeal, it is undisputed that Clarke caused the accident. The Lewises filed suit against Clarke in his individual capacity in Connecticut state court, and Clarke moved to dismiss for lack of subject-matter jurisdiction on the basis of tribal sovereign immunity. See 2014 WL 5354956, *2 (Super. Ct. Conn., Sept. 10, 2014) (Cole-Chu, J.). Clarke argued that because the Gaming Authority, an arm of the Tribe, was entitled to sovereign immunity, he, an employee of the Gaming Authority acting within the scope of his employment at the time of the accident, was similarly entitled to sovereign immunity against suit. According to Clarke, denying the motion would abrogate the Tribe’s sovereign immunity. The trial court denied Clarke’s motion to dismiss. Id., at *8. The court agreed with the Lewises that the sovereign immunity analysis should focus on the remedy sought in their complaint. To that end, the court identified Clarke, not the Gaming Authority or the Tribe, as the real party in interest because the damages remedy sought was solely against Clarke and would in no way affect the Tribe’s ability to govern itself independently. The court therefore concluded that tribal sovereign immunity was not implicated. Id., at *2–*8. It also rejected Clarke’s alternative argument that because the Gaming Authority was obligated to indemnify him pursuant to Mohegan Tribe Code §4–52 and would end up paying the damages, he should prevail under the remedy analysis. Id., at *7. The trial court reasoned that a “voluntary undertaking cannot be used to extend sovereign immunity where it did not otherwise exist.” Ibid. The Supreme Court of Connecticut reversed, holding that tribal sovereign immunity did bar the suit. 320 Conn. 706, 135 A. 3d 677 (2016). The court agreed with Clarke that “because he was acting within the scope of his employment for the Mohegan Tribal Gaming Authority and the Mohegan Tribal Gaming Authority is an arm of the Mohegan Tribe, tribal sovereign immunity bars the plaintiffs’ claims against him.” Id., at 709, 135 A. 3d, at 680. Of particular significance to the court was ensuring that “plaintiffs cannot circumvent tribal immunity by merely naming the defendant, an employee of the tribe, when the complaint concerns actions taken within the scope of his duties and the complaint does not allege, nor have the plaintiffs offered any other evidence, that he acted outside the scope of his authority.” Id., at 720, 135 A. 3d, at 685. To do otherwise, the court reasoned, would “ ‘eviscerate’ ” the protections of tribal immunity. Id., at 717, 135 A. 3d, at 684 (alterations and internal quotation marks omitted). Because the court determined that Clarke was entitled to sovereign immunity on the sole basis that he was acting within the scope of his employment when the accident occurred, id., at 720, 135 A. 3d, at 685–686, it did not consider whether Clarke should be entitled to sovereign immunity on the basis of the indemnification statute. We granted certiorari to consider whether tribal sovereign immunity bars the Lewises’ suit against Clarke, 579 U. S. ___ (2016), and we now reverse the judgment of the Supreme Court of Connecticut. II Two issues require our resolution: (1) whether the sovereign immunity of an Indian tribe bars individual-capacity damages against tribal employees for torts committed within the scope of their employment; and (2) what role, if any, a tribe’s decision to indemnify its employees plays in this analysis. We decide this case under the framework of our precedents regarding tribal immunity. A Our cases establish that, in the context of lawsuits against state and federal employees or entities, courts should look to whether the sovereign is the real party in interest to determine whether sovereign immunity bars the suit. See Hafer v. Melo, 502 U. S. 21, 25 (1991) . In making this assessment, courts may not simply rely on the characterization of the parties in the complaint, but rather must determine in the first instance whether the remedy sought is truly against the sovereign. See, e.g., Ex parte New York, 256 U. S. 490 –502 (1921). If, for example, an action is in essence against a State even if the State is not a named party, then the State is the real party in interest and is entitled to invoke the Eleventh Amendment’s protection. For this reason, an arm or instrumentality of the State generally enjoys the same immunity as the sovereign itself. E.g., Regents of Univ. of Cal. v. Doe, 519 U. S. 425 –430 (1997). Similarly, lawsuits brought against employees in their official capacity “represent only another way of pleading an action against an entity of which an officer is an agent,” and they may also be barred by sovereign immunity. Kentucky v. Graham, 473 U. S. 159 –166 (1985) (internal quotation marks omitted). The distinction between individual- and official-capacity suits is paramount here. In an official-capacity claim, the relief sought is only nominally against the official and in fact is against the official’s office and thus the sovereign itself. Will v. Michigan Dept. of State Police, 491 U. S. 58, 71 (1989) ; Dugan v. Rank, 372 U. S. 609 –622 (1963). This is why, when officials sued in their official capacities leave office, their successors automatically assume their role in the litigation. Hafer, 502 U. S., at 25. The real party in interest is the government entity, not the named official. See Edelman v. Jordan, 415 U. S. 651 –665 (1974). “Personal-capacity suits, on the other hand, seek to impose individual liability upon a government officer for actions taken under color of state law.” Hafer, 502 U. S., at 25 (emphasis added); see also id., at 27–31 (discharged employees entitled to bring personal damages action against state auditor general); cf. Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388 (1971) . “[O]fficers sued in their personal capacity come to court as individuals,” Hafer, 502 U. S., at 27, and the real party in interest is the individual, not the sovereign. The identity of the real party in interest dictates what immunities may be available. Defendants in an official-capacity action may assert sovereign immunity. Graham, 473 U. S., at 167. An officer in an individual-capacity action, on the other hand, may be able to assert personal immunity defenses, such as, for example, absolute prosecutorial immunity in certain circumstances. Van de Kamp v. Goldstein, 555 U. S. 335 –344 (2009). But sovereign immunity “does not erect a barrier against suits to impose individual and personal liability.” Hafer, 502 U. S., at 30–31 (internal quotation marks omitted); see Alden v. Maine, 527 U. S. 706, 757 (1996) . B There is no reason to depart from these general rules in the context of tribal sovereign immunity. It is apparent that these general principles foreclose Clarke’s sovereign immunity defense in this case. This is a negligence action arising from a tort committed by Clarke on an interstate highway within the State of Connecticut. The suit is brought against a tribal employee operating a vehicle within the scope of his employment but on state lands, and the judgment will not operate against the Tribe. This is not a suit against Clarke in his official capacity. It is simply a suit against Clarke to recover for his personal actions, which “will not require action by the sovereign or disturb the sovereign’s property.” Larson v. Domestic and Foreign Commerce Corp., 337 U. S. 682, 687 (1949) . We are cognizant of the Supreme Court of Connecticut’s concern that plaintiffs not circumvent tribal sovereign immunity. But here, that immunity is simply not in play. Clarke, not the Gaming Authority, is the real party in interest. In ruling that Clarke was immune from this suit solely because he was acting within the scope of his employment, the court extended sovereign immunity for tribal employees beyond what common-law sovereign immunity principles would recognize for either state or federal employees. See, e.g., Graham, 473 U. S., at 167–168. The protection offered by tribal sovereign immunity here is no broader than the protection offered by state or federal sovereign immunity. Accordingly, under established sovereign immunity principles, the Gaming Authority’s immunity does not, in these circumstances, bar suit against Clarke.[2] III The conclusion above notwithstanding, Clarke argues that the Gaming Authority is the real party in interest here because it is required by Mohegan Tribe Code §4–52 to indemnify Clarke for any adverse judgment.[3] A We have never before had occasion to decide whether an indemnification clause is sufficient to extend a sovereign immunity defense to a suit against an employee in his individual capacity. We hold that an indemnification provision cannot, as a matter of law, extend sovereign immunity to individual employees who would otherwise not fall under its protective cloak. Our holding follows naturally from the principles discussed above. Indeed, we have applied these same principles to a different question before—whether a state instrumentality may invoke the State’s immunity from suit even when the Federal Government has agreed to indemnify that instrumentality against adverse judgments. In Regents of Univ. of Cal., an individual brought suit against the University of California, a public university of the State of California, for breach of contract related to his employment at a laboratory operated by the university pursuant to a contract with the Federal Government. We held that the indemnification provision did not divest the state instrumentality of Eleventh Amendment immunity. 519 U. S., at 426. Our analysis turned on where the potential legal liability lay, not from whence the money to pay the damages award ultimately came. Because the lawsuit bound the university, we held, the Eleventh Amendment applied to the litigation even though the damages award would ultimately be paid by the federal Department of Energy. Id., at 429–431. Our reasoning remains the same. The critical inquiry is who may be legally bound by the court’s adverse judgment, not who will ultimately pick up the tab.[4] Here, the Connecticut courts exercise no jurisdiction over the Tribe or the Gaming Authority, and their judgments will not bind the Tribe or its instrumentalities in any way. The Tribe’s indemnification provision does not somehow convert the suit against Clarke into a suit against the sovereign; when Clarke is sued in his individual capacity, he is held responsible only for his individual wrongdoing. Moreover, indemnification is not a certainty here. Clarke will not be indemnified by the Gaming Authority should it determine that he engaged in “wanton, reckless, or malicious” activity. Mohegan Tribe Code §4–52. That determination is not necessary to the disposition of the Lewises’ suit against Clarke in the Connecticut state courts, which is a separate legal matter. B Clarke notes that courts have extended sovereign immunity to private healthcare insurance companies under certain circumstances. See, e.g., Pani v. Empire Blue Cross Blue Shield, 152 F. 3d 67, 71–72 (CA2 1998); Pine View Gardens, Inc. v. Mutual of Omaha Ins. Co., 485 F. 2d 1073, 1074–1075 (CADC 1973); Brief for Respondent 19, n. 4. But, these cases rest on the proposition that the fiscal intermediaries are essentially state instrumentalities, as the governing regulations make clear. See 42 CFR §421.5(b) (2016) (providing that the Medicare Administrator “is the real party of interest in any litigation involving the administration of the program”). It is well established in our precedent that a suit against an arm or instrumentality of the State is treated as one against the State itself. See Regents of Univ. of Cal., 519 U. S., at 429. We have not before treated a lawsuit against an individual em-ployee as one against a state instrumentality, and Clarke offers no persuasive reason to do so now. Nor have we ever held that a civil rights suit under 42 U. S. C. §1983 against a state officer in his individual capacity implicates the Eleventh Amendment and a State’s sovereign immunity from suit.[5] Federal appellate courts that have considered the indemnity question have rejected the argument that an indemnity statute brings the Eleventh Amendment into play in §1983 actions. See, e.g., Stoner v. Wisconsin Dept. of Agriculture, Trade and Consumer Protection, 50 F. 3d 481, 482–483 (CA7 1995); Blalock v. Schwinden, 862 F. 2d 1352, 1354 (CA9 1988); Duckworth v. Franzen, 780 F. 2d 645, 650 (CA7 1985). These cases rely on the concern that originally drove the adoption of the Eleventh Amendment—the protection of the States against involuntary liability. See Hess v. Port Authority Trans-Hudson Corporation, 513 U. S. 30, 39, 48 (1994) . But States institute indemnification policies voluntarily. And so, indemnification provisions do not implicate one of the underlying rationales for state sovereign immunity—a government’s ability to make its own decisions about “the allocation of scarce resources.” Alden, 527 U. S., at 751. Finally, our conclusion that indemnification provisions do not alter the real-party-in-interest analysis for purposes of sovereign immunity is consistent with the practicethat applies in the contexts of diversity of citizenship and joinder. In assessing diversity jurisdiction, courts look to the real parties to the controversy. Navarro Savings Assn. v. Lee, 446 U. S. 458, 460 (1980) . Applying this principle, courts below have agreed that the fact that a third party indemnifies one of the named parties to the case does not, as a general rule, influence the diversity analysis. See, e.g., Corfield v. Dallas Glen Hills LP, 355 F. 3d 853, 865 (CA5 2003); E. R. Squibb & Sons, Inc. v. Accident & Cas. Ins. Co., 160 F. 3d 925, 936–937 (CA2 1998). They have similarly held that a party does not become a required party for joinder purposes under Federal Rule of Civil Procedure 19 simply by virtue of indemnifying one of the named parties. See, e.g., Gardiner v. Virgin Islands Water & Power Auth., 145 F. 3d 635, 641 (CA3 1998); Rochester Methodist Hospital v. Travelers Ins. Co., 728 F. 2d 1006, 1016–1017 (CA8 1984). In sum, although tribal sovereign immunity is implicated when the suit is brought against individual officers in their official capacities, it is simply not present when the claim is made against those employees in their individual capacities. An indemnification statute such as the one at issue here does not alter the analysis. Clarke may not avail himself of a sovereign immunity defense. IV The judgment of the Supreme Court of Connecticut is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 There are currently 567 federally recognized Indian and Alaska Native entities. 81 Fed. Reg. 26826–26832 (2016); see also Native Hawaiian Law: A Treatise 303–324 (M. MacKenzie ed. 2015) (discussing the existing relationships between the U. S. Government and federally recognized tribes and other indigenous groups in the United States); F. Cohen, Handbook of Federal Indian Law §§1.01–1.07 (2012 and Supp. 2015); V. Deloria & R. DeMallie, Documents of American Indian Diplomacy: Treaties, Agreements, and Conventions, 1775–1979 (1999). 2 There are, of course, personal immunity defenses distinct from sovereign immunity. E.g., Harlow v. Fitzgerald, 457 U. S. 800 –815 (1982). Clarke argues for the first time before this Court that one particular form of personal immunity is available to him here—official immunity. See Westfall v. Erwin, 484 U. S. 292 –297 (1988). That defense is not properly before us now, however, given that Clarke’s motion to dismiss was based solely on tribal sovereign immunity. See Travelers Casualty & Surety Co. of America v. Pacific Gas & Elec. Co., 549 U. S. 443, 455 (2007) . 3 As noted above, the Supreme Court of Connecticut did not reach whether Clarke should be entitled to sovereign immunity on the basis of the indemnification statute. We nevertheless consider the issue fairly included within the question presented, as it is a purely legal question that is an integral part of Clarke’s sovereign immunity argument and that was both raised to and passed on by the trial court. See Mitchell v. Forsyth, 472 U. S. 511, 530 (1985) (“[T]he purely legal question on which [petitioner’s] claim of immunity turns is appropriate for our immediate resolution notwithstanding that it was not addressed by the Court of Appeals” (internal quotation marks omitted)). 4 Our holding in Hess v. Port Authority Trans-Hudson Corporation, 513 U. S. 30 (1994) , is not to the contrary. There the immunity question turned on whether the Port Authority Trans-Hudson Corporation was a state agency cloaked with Eleventh Amendment immunity such that any judgment “must be paid out of a State’s treasury.” Id., at 48, 51–52 (emphasis added). Here, unlike in Hess, the damages judgment would not come from the sovereign. 5 A suit against a state officer in his official, rather than individual, capacity might implicate the Eleventh Amendment. See Kentucky v. Graham, 473 U. S. 159 –166 (1985).

580.US.2016_14-1538

Respondent Promega Corporation sublicensed the Tautz patent, which claims a toolkit for genetic testing, to petitioner Life Technologies Corporation and its subsidiaries (collectively Life Technologies) for the manufacture and sale of the kits for use in certain licensed law enforcement fields worldwide. One of the kit’s five components, an enzyme known as the Taq polymerase, was manufactured by Life Technologies in the United States and then shipped to the United Kingdom, where the four other components were made, for combination there. When Life Technologies began selling the kits outside the licensed fields of use, Promega sued, claiming that patent infringement liability was triggered under §271(f)(1) of the Patent Act, which prohibits the supply from the United States of “all or a substantial portion of the components of a patented invention” for combination abroad. The jury returned a verdict for Promega, but the District Court granted Life Technologies’ motion for judgment as a matter of law, holding that §271(f)(1)’s phrase “all or a substantial portion” did not encompass the supply of a single component of a multicomponent invention. The Federal Circuit reversed. It determined that a single important component could constitute a “substantial portion” of the components of an invention under §271(f)(1) and found the Taq polymerase to be such a component. Held: The supply of a single component of a multicomponent invention for manufacture abroad does not give rise to §271(f)(1) liability. Pp. 4–11. (a) Section 271(f)(1)’s phrase “substantial portion” refers to a quantitative measurement. Although the Patent Act itself does not define the term “substantial,” and the term’s ordinary meaning may refer either to qualitative importance or to quantitatively large size, the statutory context points to a quantitative meaning. Neighboring words “all” and “portion” convey a quantitative meaning, and nothing in the neighboring text points to a qualitative interpretation. More-over, a qualitative reading would render the modifying phrase “of the components” unnecessary the first time it is used in §271(f)(1). Only the quantitative approach thus gives meaning to each statutory provision. Promega’s proffered “case-specific approach,” which would require a factfinder to decipher whether the components at issue are a “substantial portion” under either a qualitative or a quantitative test, is rejected. Tasking juries with interpreting the statute’s meaning on an ad hoc basis would only compound, not resolve, the statute’s ambiguity. And Promega’s proposal to adopt an analytical framework that accounts for both the components’ quantitative and qualitative aspects is likely to complicate rather than aid the factfinder’s review. Pp. 4–8. (b) Under a quantitative approach, a single component cannot constitute a “substantial portion” triggering §271(f)(1) liability. This conclusion is reinforced by §271(f)’s text, context, and structure. Section 271(f)(1) consistently refers to the plural “components,” indicating that multiple components make up the substantial portion. Reading §271(f)(1) to cover any single component would also leave little room for §271(f)(2), which refers to “any component,” and would undermine §271(f)(2)’s express reference to a single component “especially made or especially adapted for use in the invention.” The better reading allows the two provisions to work in tandem and gives each provision its unique application. Pp. 8–10. (c) The history of §271(f) further bolsters this conclusion. Congress enacted §271(f) in response to Deepsouth Packing Co. v. Laitram Corp., 406 U. S. 518 , to fill a gap in the enforceability of patent rights by reaching components that are manufactured in the United States but assembled overseas. Consistent with Congress’s intent, a supplier may be liable under §271(f)(1) for supplying from the United States all or a substantial portion of the components of the invention or under §271(f)(2) for supplying a single component if it is especially made or especially adapted for use in the invention and not a staple article or commodity. But, as here, when a product is made abroad and all components but a single commodity article are supplied from abroad, the activity is outside the statute’s scope. Pp. 10–11. 773 F. 3d. 1338, reversed and remanded. Sotomayor, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Breyer, and Kagan, JJ., joined, and in which Thomas and Alito, JJ., joined as to all but Part II–C. Alito, J., filed an opinion concurring in part and concurring in the judgment, in which Thomas, J., joined. Roberts, C. J., took no part in the decision of the case.

This case concerns the intersection of international supply chains and federal patent law. Section 271(f)(1) of the Patent Act of 1952 prohibits the supply from theUnited States of “all or a substantial portion” of the components of a patented invention for combination abroad. 35 U. S. C. §271(f)(1). We granted certiorari to determine whether a party that supplies a single component of a multicomponent invention for manufacture abroad can be held liable for infringement under §271(f)(1). 579 U. S. ___ (2016). We hold that a single component does not constitute a substantial portion of the components that can give rise to liability under §271(f)(1). Because only a single component of the patented invention at issue here was supplied from the United States, we reverse and remand. I A We begin with an overview of the patent in dispute. Although the science behind the patent is complex, a basic understanding suffices to resolve the question presented by this case. The Tautz patent, U. S. Reissue Patent No. RE 37,984, claims a toolkit for genetic testing.[1] The kit is used to take small samples of genetic material—in the form of nucleotide sequences that make up the molecule deoxyribonu-cleic acid (commonly referred to as “DNA”)—and then synthesize multiple copies of a particular nucleotide sequence. This process of copying, known as amplification, generates DNA profiles that can be used by law enforcement agencies for forensic identification and by clinical and research institutions around the world. For purposes of this litigation, the parties agree that the kit covered by the Tautz patent contains five components: (1) a mixture of primers that mark the part of the DNA strand to be copied; (2) nucleotides for forming replicated strands of DNA; (3) an enzyme known as Taq polymerase; (4) a buffer solution for the amplification; and (5) control DNA.[2] Respondent Promega Corporation was the exclusive licensee of the Tautz patent. Petitioner Life Technologies Corporation manufactured genetic testing kits.[3] During the timeframe relevant here, Promega sublicensed the Tautz patent to Life Technologies for the manufacture and sale of the kits for use in certain licensed law enforcement fields worldwide. Life Technologies manufactured all but one component of the kits in the United Kingdom. It manufactured that component—the Taq polymerase—in the United States. Life Technologies shipped the Taq polymerase to its United Kingdom facility, where it was combined with the other four components of the kit. Four years into the agreement, Promega sued Life Technologies on the grounds that Life Technologies had infringed the patent by selling the kits outside the licensed fields of use to clinical and research markets. As relevant here, Promega alleged that Life Technologies’ supply of the Taq polymerase from the United States to its United Kingdom manufacturing facilities triggered liability under §271(f)(1). B At trial, the parties disputed the scope of §271(f)(1)’s prohibition against supplying all or a substantial portion of the components of a patented invention from the United States for combination abroad. Section 271(f)(1)’s full text reads: “Whoever without authority supplies or causes to be supplied in or from the United States all or a substantial portion of the components of a patented invention, where such components are uncombined in whole or in part, in such manner as to actively induce the combination of such components outside of the United States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.” The jury returned a verdict for Promega, finding that Life Technologies had willfully infringed the patent. Life Technologies then moved for judgment as a matter of law, contending that §271(f)(1) did not apply to its conduct because the phrase “all or a substantial portion” does not encompass the supply of a single component of a multicomponent invention. The District Court granted Life Technologies’ motion. The court agreed that there could be no infringement under §271(f)(1) because Promega’s evidence at trial “showed at most that one component of all of the accused products, [the Taq] polymerase, was supplied from the United States.” 2012 WL 12862829, *3 (WD Wis., Sept. 13, 2012) (Crabb, J.). Section 271(f)(1)’s reference to “a substantial portion of the components,” the District Court ruled, does not embrace the supply of a single component. Id., at *5. The Court of Appeals for the Federal Circuit reversed and reinstated the jury’s verdict finding Life Technologies liable for infringement.[4] 773 F. 3d 1338, 1353 (2014). As relevant here, the court held that “there are circumstances in which a party may be liable under §271(f)(1) for supplying or causing to be supplied a single component for combination outside the United States.” Ibid. The Federal Circuit concluded that the dictionary definition of “substantial” is “important” or “essential,” which it read to suggest that a single important component can be a “ ‘substantial portion of the components’ ” of a patented invention. Ibid. Relying in part on expert trial testimony that the Taq polymerase is a “ ‘main’ ” and “ ‘major’ ” component of the kits, the court ruled that the single Taq polymerase component was a substantial component as the term is used in §271(f)(1). Id., at 1356. II The question before us is whether the supply of a single component of a multicomponent invention is an infringing act under 35 U. S. C. §271(f)(1). We hold that it is not. A The threshold determination to be made is whether §271(f)(2)’s requirement of “a substantial portion” of the components of a patented invention refers to a quantitative or qualitative measurement. Life Technologies and the United States argue that the text of §271(f)(1) establishes a quantitative threshold, and that the threshold must be greater than one. Promega defends the Federal Circuit’s reading of the statute, arguing that a “substantial portion” of the components includes a single component if that component is sufficiently important to the invention. We look first to the text of the statute. Sebelius v. Cloer, 569 U. S. ___, ___ (2013) (slip op., at 6). The Patent Act itself does not define the term “substantial,” and so we turn to its ordinary meaning. Ibid. Here we find little help. All agree the term is ambiguous and, taken in isolation, might refer to an important portion or to a large portion. Brief for Petitioners 16; Brief for Respondent 18; Brief for United States as Amicus Curiae 12. “Substantial,” as it is commonly understood, may refer either to qualitative importance or to quantitatively large size. See, e.g., Webster’s Third New International Dictionary 2280 (defs. 1c, 2c) (1981) (Webster’s Third) (“important, essential,” or “considerable in amount, value, or worth”); 17 Oxford English Dictionary 67 (defs. 5a, 9) (2d ed. 1989) (OED) (“That is, constitutes, or involves an essential part, point, or feature; essential, material,” or “Of ample or considerable amount, quantity, or dimensions”). The context in which “substantial” appears in the statute, however, points to a quantitative meaning here. Its neighboring terms are the first clue. “[A] word is given more precise content by the neighboring words with which it is associated.” United States v. Williams, 553 U. S. 285, 294 (2008) . Both “all” and “portion” convey a quantitative meaning. “All” means the entire quantity, without reference to relative importance. See, e.g., Webster’s Third 54 (defs. 1a, 2a, 3) (“that is the whole amount or quantity of,” or “every member or individual component of,” or “the whole number or sum of”); 1 OED 324 (def. 2) (“The entire number of; the individual components of, without exception”). “Portion” likewise refers to some quantity less than all. Webster’s Third 1768 (defs. 1, 3a) (“an individual’s part or share of something,” or “a part of a whole”); 12 OED 154, 155 (def. 1a, 5a) (“The part (of anything) allotted or belonging to one person,” or “A part of any whole”). Conversely, there is nothing in the neighboring text to ground a qualitative interpretation. Moreover, the phrase “substantial portion” is modified by “of the components of a patented invention.” It is the supply of all or a substantial portion “of the components” of a patented invention that triggers liability for infringement. But if “substantial” has a qualitative meaning, then the more natural way to write the opening clause of the provision would be to not reference “the components” at all. Instead, the opening clause of §271(f)(1) could have triggered liability for the supply of “all or a substantial portion of . . . a patented invention, where [its] components are uncombined in whole or in part.” A qualitative reading would render the phrase “of the components” unnecessary the first time it is used in §271(f)(1). Whenever possible, however, we should favor an interpretation that gives meaning to each statutory provision. See Hibbs v. Winn, 542 U. S. 88, 101 (2004) . Only the quantitative approach does so here. Thus, “substantial,” in the context of §271(f)(1), is most reasonably read to connote a quantitative measure. Promega argues that a quantitative approach is too narrow, and invites the Court to instead adopt a “case-specific approach” that would require a factfinder to decipher whether the components at issue are a “substantial portion” under either a qualitative or quantitative test. Brief for Respondent 17, 42. We decline to do so. Having determined the phrase “substantial portion” is ambiguous, our task is to resolve that ambiguity, not to compound it by tasking juries across the Nation with interpreting the meaning of the statute on an ad hoc basis. See, e.g., Robinson v. Shell Oil Co., 519 U. S. 337 –346 (1997). As a more general matter, moreover, we cannot accept Promega’s suggestion that the Court adopt a different analytical framework entirely—one that accounts for both the quantitative and qualitative aspects of the components. Promega reads §271(f)(1) to mean that the answer to whether a given portion of the components is “substantial” depends not only on the number of components involved but also on their qualitative importance to the invention overall. At first blush, there is some appeal to the idea that, in close cases, a subjective analysis of the qualitative importance of a component may help determine whether it is a “substantial portion” of the components of a patent. But, for the reasons discussed above, the statute’s structure provides little support for a qualitative interpretation of the term.[5] Nor would considering the qualitative importance of a component necessarily help resolve close cases. To the contrary, it might just as easily complicate the factfinder’s review. Surely a great many components of an invention (if not every component) are important. Few inventions, including the one at issue here, would function at all without any one of their components. Indeed, Promega has not identified any component covered by the Tautz patent that would not satisfy Promega’s “importance” litmus test.[6] How are courts—or, for that matter, market participants attempting to avoid liability—to determine the relative importance of the components of an invention? Neither Promega nor the Federal Circuit offers an easy way to make this decision. Accordingly, we conclude that a quantitative interpretation hews most closely to the text of the statute and provides an administrable construction. B Having determined that the term “substantial portion” refers to a quantitative measurement, we must next decide whether, as a matter of law, a single component can ever constitute a “substantial portion” so as to trigger liability under §271(f)(1). The answer is no. As before, we begin with the text of the statute. Section 271(f)(1) consistently refers to “components” in the plural. The section is targeted toward the supply of all or a substantial portion “of the components,” where “such components” are uncombined, in a manner that actively induces the combination of “such components” outside the United States. Text specifying a substantial portion of “components,” plural, indicates that multiple components constitute the substantial portion. The structure of §271(f) reinforces this reading. Section 271(f)(2), which is §271(f)(1)’s companion provision, reads as follows: “Whoever without authority supplies or causes to be supplied in or from the United States any component of a patented invention that is especially made or especially adapted for use in the invention and not a staple article or commodity of commerce suitable for substantial noninfringing use, where such component is uncombined in whole or in part, knowing that such component is so made or adapted and intending that such component will be combined outside of theUnited States in a manner that would infringe the patent if such combination occurred within the United States, shall be liable as an infringer.” Reading §271(f)(1) to refer to more than one component allows the two provisions to work in tandem. Whereas §271(f)(1) refers to “components,” plural, §271(f)(2) refers to “any component,” singular. And, whereas §271(f)(1) speaks to whether the components supplied by a party constitute a substantial portion of the components,§271(f)(2) speaks to whether a party has supplied “any” noncommodity component “especially made or especially adapted for use in the invention.” We do not disagree with the Federal Circuit’s observation that the two provisions concern different scenarios. See 773 F. 3d, at 1354. As this Court has previously observed, §§271(f)(1) and 271(f)(2) “differ, among other things, on the quantity of components that must be ‘supplie[d] . . . from the United States’ for liability to attach.” Microsoft Corp. v. AT&T Corp., 550 U. S. 437, 454, n. 16 (2007) . But we do not draw the Federal Circuit’s conclusion from these different but related provisions. Reading §271(f)(1) to cover any single component would not only leave little room for §271(f)(2), but would also undermine §271(f)(2)’s express reference to a single component “especially made or especially adapted for use in the invention.”[7] Our conclusion that §271(f)(1) prohibits the supply of components, plural, gives each subsection its unique application.[8] See, e.g., Cloer, 569 U. S., at ___ (slip op.,at 6). Taken alone, §271(f)(1)’s reference to “components” might plausibly be read to encompass “component” in the singular. See 1 U. S. C. §1 (instructing that “words importing the plural include the singular,” “unless the context indicates otherwise”). But §271(f)’s text, context, and structure leave us to conclude that when Congress said “components,” plural, it meant plural, and when it said “component,” singular, it meant singular. We do not today define how close to “all” of the components “a substantial portion” must be. We hold only that one component does not constitute “all or a substantial portion” of a multicomponent invention under §271(f)(1). This is all that is required to resolve the question presented. C The history of §271(f) bolsters our conclusion. The Court has previously observed that Congress enacted §271(f) in response to our decision in Deepsouth Packing Co. v. Laitram Corp., 406 U. S. 518 (1972) . See Microsoft Corp., 550 U. S., at 444. In Deepsouth, the Court determined that, under patent law as it existed at the time, it was “not an infringement to make or use a patented product outside of the United States.” 406 U. S., at 527. The new §271(f) “expand[ed] the definition of infringement to include supplying from the United States a patented invention’s components,” as outlined in subsections (f)(1) and (f)(2). Microsoft, 550 U. S., at 444–445. The effect of this provision was to fill a gap in the enforceability of patent rights by reaching components that are manufactured in the United States but assembled overseas and that were beyond the reach of the statute in its prior formulation. Our ruling today comports with Congress’ intent. A supplier may be liable under§271(f)(1) for supplying from the United States all or a substantial portion of the components (plural) of the invention, even when those components are combined abroad. The same is true even for a single component under §271(f)(2) if it is especially made or especially adapted for use in the invention and not a staple article or commodity. We are persuaded, however, that when as in this case a product is made abroad and all components but a single commodity article are supplied from abroad, this activity is outside the scope of the statute. III We hold that the phrase “substantial portion” in 35 U. S. C. §271(f)(1) has a quantitative, not a qualitative, meaning. We hold further that §271(f)(1) does not cover the supply of a single component of a multicomponent invention. The judgment of the Court of Appeals for the Federal Circuit is therefore reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. The Chief Justice took no part in the decision of this case.Notes 1 The Tautz patent expired in 2015. The litigation thus concerns past acts of infringement only. 2 Because the parties here agree that the patented invention is made up of only these five components, we do not consider how to identify the “components” of a patent or whether and how that inquiry relates to the elements of a patent claim. 3 Applied Biosystems, LLC, and Invitrogen IP Holdings, Inc., are also petitioners in this proceeding and are wholly owned subsidiaries of Life Technologies Corporation. The agreement at issue here was originally between Promega and Applied Biosystems. 773 F. 3d 1338, 1344, n. 3 (CA Fed. 2014). 4 Chief Judge Prost dissented from the majority’s conclusion with respect to the “active inducement” element of 35 U. S. C. §271(f )(1). 773 F. 3d, at 1358–1360. Neither that question, nor any of the Federal Circuit’s conclusions regarding Life Technologies’ liability under §271(a) or infringement of four additional Promega patents, see id., at 1341, is before us. See 579 U. S. ___ (2016). 5 The examples Promega provides of other statutes’ use of the terms “substantial” or “significant” are inapposite. See Brief for Respondent 19–20. The text of these statutes, which arise in different statutory schemes with diverse purposes and structures, differs in material ways from the text of §271(f )(1). The Tax Code, for instance, refers to “a substantial portion of a return,” 26 U. S. C. §7701(a)(36)(A), not to “a substantial portion of the entries of a return.” 6 Life Technologies’ expert described the Taq polymerase as a “main” component. App. 160. The expert also described two other components the same way. Ibid. 7 This Court’s opinion in Microsoft Corp. v. AT&T Corp., 550 U. S. 437, 447 (2007) , is not to the contrary. The holding in that case turned not on the number of components involved, but rather on whether the software at issue was a component at all. 8 Promega argues that the important distinction between these provisions is that §271(f )(1), unlike §271(f )(2), requires a showing of specific intent for active inducement. Brief for Respondent 34–41. But cf. Global-Tech Appliances, Inc. v. SEB S. A., 563 U. S. 754 –766 (2011) (substantially equating the intent requirements for §§271(b) and 271(c), on which Promega asserts §§271(f )(1) and (f )(2) were modeled). But, to repeat, whatever intent subsection (f )(1) may require, it also imposes liability only on a party who supplies a “substantial portion of the components” of the invention. Thus, even assuming that subsection (f )(1)’s “active inducement” requirement is different from subsection (f )(2)’s “knowing” and “intending” element—a question we do not reach today—that difference between the two provisions does not read the “substantial portion” language out of the statute.

580.US.2016_14-1055

The Federal National Mortgage Association (Fannie Mae) is a federally chartered corporation that participates in the secondary mortgage market. By statute, Fannie Mae has the power “to sue and to be sued, and to complain and to defend, in any court of competent jurisdiction, State or Federal.” 12 U. S. C. §1723a(a). When petitioners Beverly Ann Hollis-Arrington and her daughter Crystal Lightfoot filed suit in state court alleging deficiencies in the refinancing, foreclosure, and sale of their home, Fannie Mae removed the case to federal court, relying on its sue-and-be-sued clause as the basis for jurisdiction. The District Court denied a motion to remand the case to state court and later entered judgment against petitioners. The Ninth Circuit affirmed. In concluding that the District Court had jurisdiction under Fannie Mae’s sue-and-be-sued clause, the court relied on American Nat. Red Cross v. S. G., 505 U. S. 247 , which it read as establishing a rule that when a sue-and-be-sued clause in a federal charter expressly authorizes suit in federal court, it confers jurisdiction on the federal courts. Held: Fannie Mae’s sue-and-be-sued clause does not grant federal courts jurisdiction over all cases involving Fannie Mae. Pp. 6–16. (a) This Court has addressed the jurisdictional reach of sue-and-be-sued clauses in five federal charters. Three clauses were held to grant jurisdiction—Osborn v. Bank of United States, 9 Wheat. 738; D’Oench, Duhme & Co. v. FDIC, 315 U. S. 447 ; American Nat. Red Cross v. S. G., 505 U. S. 247 —while two were found wanting—Bank of United States v. Deveaux, 5 Cranch 61; Bankers Trust Co. v. Texas & Pacific R. Co., 241 U. S. 295 . Describing the earlier decisions as this Court’s “best efforts at divining congressional intent retrospectively,” 505 U. S., at 252, the Court in Red Cross concluded that those decisions “support the rule that a congressional charter’s ‘sue and be sued’ provision may be read to confer federal court jurisdiction if, but only if, it specifically mentions the federal courts,” id., at 255. In specifically mentioning the federal courts, Fannie Mae’s sue-and-be-sued clause resembles the three clauses this Court has held confer jurisdiction. But unlike those clauses, Fannie Mae’s clause adds the qualification “any court of competent jurisdiction,” 12 U. S. C. §1723a(a). Thus, the outcome here turns on the meaning of “court of competent jurisdiction.” A court of competent jurisdiction is a court with the power to adjudicate the case before it, Black’s Law Dictionary 431, and a court’s subject-matter jurisdiction defines its power to hear cases, see Steel Co. v. Citizens for Better Environment, 523 U. S. 83 . It follows that a court of competent jurisdiction is a court with a grant of subject-matter jurisdiction covering the case before it. This Court has understood that phrase as a reference to a court with an existing source of subject-matter jurisdiction. See, e.g., Ex parte Phenix Ins. Co., 118 U. S. 610 . On this understanding, Fannie Mae’s sue-and-be-sued clause is most naturally read not to grant federal courts subject-matter jurisdiction over all cases involving Fannie Mae but to permit suit in any state or federal court already endowed with subject-matter jurisdiction. Red Cross does not require a different result. It did not set out a rule that an express reference to the federal courts suffices to make a sue-and-be-sued clause a grant of federal jurisdiction. Rather, it restated “the basic rule” of Deveaux and Osborn that a sue-and-be-sued clause conferring only a general right to sue does not grant jurisdiction to the federal courts. 505 U. S., at 253. Pp. 6–11. (b) Fannie Mae’s arguments against reading its sue-and-be-sued clause as merely capacity conferring are unpersuasive. Its alternative readings of “court of competent jurisdiction” are premised on the already rejected reading of Red Cross. The prior construction canon of statutory interpretation does not apply because none of the cases on which Fannie Mae relies suggest that Congress in 1954 would have surveyed the jurisprudential landscape and necessarily concluded that the courts had already settled the question whether a sue-and-be-sued clause containing the phrase “court of competent jurisdiction” confers jurisdiction on the federal courts. Finally, Fannie Mae’s appeals to congressional purpose do not call into question the plain text reading of its sue-and-be-sued clause. Pp. 11–16. 769 F. 3d 681, reversed. Sotomayor, J., delivered the opinion for a unanimous Court.

The corporate charter of the Federal National Mortgage Association, known as Fannie Mae, authorizes Fannie Mae “to sue and to be sued, and to complain and to defend, in any court of competent jurisdiction, State or Federal.” 12 U. S. C. §1723a(a). This case presents the question whether this sue-and-be-sued clause grants federal district courts jurisdiction over cases involving Fannie Mae. We hold that it does not. I A During the Great Depression, the Federal Government worked to stabilize and strengthen the residential mortgage market. Among other things, it took steps to increase liquidity (reasonably available funding) in the mortgage market. These efforts included the creation of the Federal Home Loan Banks, which provide credit to member institutions to finance affordable housing and economic development projects, and the Federal Housing Administration (FHA), which insures residential mort-gages. See Dept. of Housing and Urban Development, Back-ground and History of the Federal National Mortgage Association 1–7, A4 (1966). Also as part of these efforts, Title III of the National Housing Act (1934 Act) authorized the Administrator of the newly created FHA to establish “national mortgage associations” that could “purchase and sell [certain] first mortgages and such other first liens” and “borrow money for such purposes.” §301(a), 48Stat. 1252–1253. The associations were endowed with certain powers, including the power to “sue and be sued, complain and defend, in any court of law or equity, State or Federal.” §301(c), id., at 1253. In 1938, the FHA Administrator exercised that author-ity and chartered the Federal National Mortgage Association. Avoiding a mouthful of an acronym (FNMA), it went by Fannie Mae. See, e.g., Washington Post, July 14, 1940, p. P2 (“ ‘Fanny May’ ”); N. Y. Times, Mar. 23, 1950, p. 48 (“ ‘Fannie Mae’ ”). As originally chartered, Fannie Mae was wholly owned by the Federal Government and had three objectives: to “establish a market for [FHA-insured] first mortgages” covering new housing construction, to “facilitate the construction and financing of economically sound rental housing projects,” and to “make [the bonds it issued] available to . . . investors.” Fed. Nat. Mortgage Assn. Information Regarding the Activities of the Assn. 1 (Circular No. 1, 1938). Fannie Mae was rechartered in 1954. Housing Act of 1954 (1954 Act), §201, 68Stat. 613. No longer wholly Government owned, Fannie Mae had mixed ownership: Private shareholders held its common stock and the Department of the Treasury held its preferred stock. The 1954 Act required the Secretary of the Treasury to allow Fannie Mae to repurchase that stock. See id., at 613–615. It expected that Fannie Mae would repurchase all of its preferred stock and that legislation would then be enacted to turn Fannie Mae over to the private stockholders. From then on, Fannie Mae’s duties would “be carried out by a privately owned and privately financed corporation.” Id., at 615. Along with these structural changes, the 1954 Act replaced Fannie Mae’s initial set of powers with a more detailed list. In doing so, it revised the sue-and-be-sued clause to give Fannie Mae the power “to sue and to be sued, and to complain and to defend, in any court of competent jurisdiction, State or Federal.” Id., at 620. In 1968, Fannie Mae became fully privately owned and relinquished part of its portfolio to its new spinoff, the Government National Mortgage Association (known as Ginnie Mae). See Housing and Urban Development Act of 1968 (1968 Act), 82Stat. 536. Fannie Mae “continue[d] to operate the secondary market operations” but became “a Government-sponsored private corporation.” 12 U. S. C. §1716b. Ginnie Mae “remain[ed] in the Government” and took over “the special assistance functions and management and liquidating functions.” Ibid. Ginnie Mae received the same set of powers as Fannie Mae. See §1723(a); see also 1968 Act, §802(z), 82Stat. 540 (minor revisions to §1723a(a)). This general structure remains in place. Fannie Mae continues to participate in the secondary mortgage market. It purchases mortgages that meet its eligibility criteria, packages them into mortgage-backed securities, and sells those securities to investors, and it invests in mortgage-backed securities itself. One of those mortgage purchases led to Fannie Mae’s entanglement in this case. B Beverly Ann Hollis-Arrington refinanced her mortgage with Cendant Mortgage Corporation (Cendant) in the summer of 1999. Fannie Mae then bought the mortgage, while Cendant continued to service it. Unable to make her payments, Hollis-Arrington pursued a forbearance arrangement with Cendant. No agreement materialized, and the home entered foreclosure. Around this time, Cendant repurchased the mortgage from Fannie Mae because it did not meet Fannie Mae’s credit standards. To stave off the foreclosure, Hollis-Arrington and her daughter, Crystal Lightfoot, pursued bankruptcy and transferred the property between themselves. These ef-forts failed, and the home was sold at a trustee’s sale in 2001. The two then took to the courts to try to undo the foreclosure and sale. After two unsuccessful federal suits, the pair filed this suit in state court. They alleged that deficiencies in the refinancing, foreclosure, and sale of their home entitled them to relief against Fannie Mae. Their claims against other defendants are not relevant here. Fannie Mae removed the case to federal court under 28 U. S. C. §1441(a), which permits a defendant to remove from state to federal court “any civil action” over which the federal district courts “have original jurisdiction.” It relied on its sue-and-be-sued clause as the basis for jurisdiction. The District Court denied a motion to remand the case to state court. The District Court then dismissed the claims against Fannie Mae on claim preclusion grounds. After a series of motions, rulings, and appeals not related to the issue here, the District Court entered final judgment. Hollis-Arrington and Lightfoot immediately moved to set aside the judgment under Federal Rule of Civil Procedure 60(b), alleging “fraud upon the court.” App. 95–110. The District Court denied the motion. The Ninth Circuit affirmed the dismissal of the case and the denial of the Rule 60(b) motion. 465 Fed. Appx. 668 (2012). After Hollis-Arrington and Lightfoot sought rehearing, the Ninth Circuit withdrew its opinion and ordered briefing on the question whether the District Court had jurisdiction over the case under Fannie Mae’s sue-and-be-sued clause. 769 F. 3d 681, 682–683 (2014). A divided panel affirmed the District Court’s judgment. The majority relied on American Nat. Red Cross v. S. G., 505 U. S. 247 (1992) . It read that decision to have established a “rule [that] resolves this case”: When a sue-and-be-sued clause in a federal charter expressly authorizes suit in federal courts, it confers jurisdiction on the federal courts. 769 F. 3d, at 684. The dissent instead read Red Cross as setting out only a “ ‘default rule’ ” that provides a “starting point for [the] analysis.” 769 F. 3d, at 692 (opinion of Stein, J.). It read “any court of competent jurisdiction” in Fannie Mae’s sue-and-be-sued clause to overcome that default rule by requiring an independent source for jurisdiction in cases involving Fannie Mae. Ibid. Two Circuits have likewise concluded that the language in Fannie Mae’s sue-and-be-sued clause grants jurisdiction to federal courts. See Federal Home Loan Bank of Boston v. Moody’s Corp., 821 F. 3d 102 (CA1 2016) (Federal Home Loan Bank of Boston’s identical sue-and-be-sued clause); Pirelli Armstrong Tire Corp. Retiree Medical Benefits Trust ex rel. Fed. Nat. Mortgage Assn. v. Raines, 534 F. 3d 779 (CADC 2008) (Fannie Mae’s sue-and-be-sued clause). Four Circuits have disagreed, finding that similar language did not grant jurisdiction. See Western Securities Co. v. Derwinski, 937 F. 2d 1276 (CA7 1991) (Under 38 U. S. C. §1820(a)(1) (1988 ed.), Secretary of Veterans Affairs’ authority to “sue and be sued . . . in any court of competent jurisdiction, State or Federal”); C. H. Sanders Co. v. BHAP Housing Development Fund Co., 903 F. 2d 114 (CA2 1990) (Under 12 U. S. C. §1702 (1988 ed.), Secretary of Housing and Urban Development’s authority “in his official capacity, to sue and be sued in any court of competent jurisdiction, State or Federal”); Industrial Indemnity, Inc. v. Landrieu, 615 F. 2d 644 (CA5 1980) ( per curiam) (similar); Lindy v. Lynn, 501 F. 2d 1367 (CA3 1974) (similar). We granted certiorari, 579 U. S. ___ (2016), and now reverse. II Fannie Mae’s sue-and-be-sued clause authorizes it “to sue and to be sued, and to complain and to defend, in any court of competent jurisdiction, State or Federal.” 12 U. S. C. §1723a(a). As in other federal corporate charters, this language serves the uncontroversial function of clarifying Fannie Mae’s capacity to bring suit and to be sued. See Bank of United States v. Deveaux, 5 Cranch 61, 85–86 (1809). The question here is whether Fannie Mae’s sue-and-be-sued clause goes further and grants federal courts jurisdiction over all cases involving Fannie Mae. A In answering this question, “we do not face a clean slate.” Red Cross, 505 U. S., at 252. This Court has addressed the jurisdictional reach of sue-and-be-sued clauses in five federal charters. Three clauses were held to grant jurisdiction, while two were found wanting. The first discussion of sue-and-be-sued clauses came in a pair of opinions by Chief Justice Marshall. The charter of the first Bank of the United States allowed it “ ‘to sue and be sued, plead and be impleaded, answer and be answered, defend and be defended, in courts of record, or any other place whatsoever.’ ” Deveaux, 5 Cranch, at 85. Another provision allowed suits in federal court against certain bank officials, suggesting “the right to sue does not imply a right to sue in the courts of the union, unless it be expressed.” Id., at 86. In light of this language, the Court held that the first Bank of the United States had “no right . . . to sue in the federal courts.” Ibid. The Court con-cluded that the second Bank of the United States was not similarly disabled. Its charter allowed it “ ‘to sue and be sued, plead and be impleaded, answer and be answered, defend and be defended, in all State Courts having competent jurisdiction, and in any Circuit Court of the United States.’ ” Osborn v. Bank of United States, 9 Wheat. 738, 817 (1824). The Court took from Deveaux “that a general capacity in the Bank to sue, without mentioning the Courts of the Union, may not give a right to sue in those Courts.” 9 Wheat., at 818. By contrast, the second Bank’s charter did grant jurisdiction to the federal circuit courts because it used “words expressly conferring a right to sue in those Courts.” Ibid. A mortgage dispute between a railroad and its creditor led to the next consideration of this issue. The Texas and Pacific Railway Company’s federal charter authorized it “ ‘to sue and be sued, plead and be impleaded, defend and be defended, in all courts of law and equity within the United States.’ ” Bankers Trust Co. v. Texas & Pacific R. Co., 241 U. S. 295, 302 (1916) . This Court held that the clause had “the same generality and natural import as” the clause in Deveaux. 241 U. S., at 304. Thus, “all that was intended was to render this corporation capable of suing and being sued by its corporate name in any court . . . whose jurisdiction as otherwise competently defined was adequate to the occasion.” Id., at 303. Another lending dispute, involving defaulted bonds, led to the next statement on this issue. The Federal Deposit Insurance Corporation’s (FDIC) sue-and-be-sued clause authorized it “[t]o sue and be sued, complain and defend, in any court of law or equity, State or Federal.” 12 U. S. C. §264(j) (1940 ed.). In D’Oench, Duhme & Co. v. FDIC, 315 U. S. 447, 455 (1942) , this Court held that federal jurisdiction over the case was based on the FDIC’s sue-and-be-sued clause. See Red Cross, 505 U. S., at 254 (expressing no “doubt that the Court held federal jurisdiction to rest on the” sue-and-be-sued clause). This Court’s most recent discussion of a sue-and-be-sued clause came in Red Cross, which involved a state-law tort suit related to a contaminated blood transfusion. It described the previous quartet of decisions as reflecting this Court’s “best efforts at divining congressional intent retrospectively,” efforts that had put “Congress on prospective notice of the language necessary and sufficient to confer jurisdiction.” Id., at 252. Those decisions “support the rule that a congressional charter’s ‘sue and be sued’ provision may be read to confer federal court jurisdiction if, but only if, it specifically mentions the federal courts.” Id., at 255. Under that rule, the Court explained, the result was “clear.” Id., at 257. The Red Cross’ sue-and-be-sued clause, which permits it to “sue and be sued in courts of law and equity, State or Federal, within the jurisdiction of the United States,” 36 U. S. C. §300105(a)(5), confers jurisdiction. Red Cross, 505 U. S., at 257. “In expressly authorizing [suits] in federal courts, using language . . . in all relevant respects identical to [the clause in D’Oench] on which [the Court] based a holding of federal jurisdiction just five years before [its enactment], the provision extends beyond a mere grant of general corporate capacity to sue, and suffices to confer federal jurisdiction.” Ibid. Armed with these earlier cases, as synthesized by Red Cross, we turn to the sue-and-be-sued clause at issue here. B Fannie Mae’s sue-and-be-sued clause resembles the clauses this Court has held confer jurisdiction in one important respect. In authorizing Fannie Mae “to sue and to be sued, and to complain and to defend, in any court of competent jurisdiction, State or Federal,” 12 U. S. C. §1723a(a), it “specifically mentions the federal courts.” Red Cross, 505 U. S., at 255. This mention of the federal courts means that Fannie Mae’s charter clears a hurdle that the clauses in Deveaux and Bankers Trust did not. But Fannie Mae’s clause differs in a material respect from the three clauses the Court has held sufficient to grant federal jurisdiction. Those clauses referred to suits in the federal courts without qualification. In contrast, Fannie Mae’s sue-and-be-sued clause refers to “any court of competent jurisdiction, State or Federal.” §1723a(a) (emphasis added). Because this sue-and-be-sued clause is not “in all relevant respects identical” to a clause already held to grant federal jurisdiction, Red Cross, 505 U. S., at 257, this case cannot be resolved by a simple comparison. The outcome instead turns on the meaning of “court of competent jurisdiction” in Fannie Mae’s sue-and-be-sued clause. A court of competent jurisdiction is a court with the power to adjudicate the case before it. See Black’s Law Dictionary 431 (10th ed. 2014) (“[a] court that has the power and authority to do a particular act; one recognized by law as possessing the right to adjudicate a contro-versy”). And a court’s subject-matter jurisdiction defines its power to hear cases. See Steel Co. v. Citizens for Better Environment, 523 U. S. 83, 89 (1998) (Subject-matter jurisdiction is “the courts’ statutory or constitutional power to adjudicate the case” (emphasis deleted)); Wachovia Bank, N. A. v. Schmidt, 546 U. S. 303, 316 (2006) (“Subject-matter jurisdiction . . . concerns a court’s competence to adjudicate a particular category of cases”). It follows that a court of competent jurisdiction is a court with a grant of subject-matter jurisdiction covering the case before it. Cf. Pennoyer v. Neff, 95 U. S. 714, 733 (1878) (“[T]here must be a tribunal competent by its constitution—that is, by the law of its creation—to pass upon the subject-matter of the suit”). As a result, this Court has understood the phrase “court of competent jurisdiction” as a reference to a court with an existing source of subject-matter jurisdiction. Ex parte Phenix Ins. Co., 118 U. S. 610 (1886) , provides an example. There, the Court explained that a statute “providing for the transfer to a trustee of the interest of the owner in the vessel and freight, provides only that the trustee may ‘be appointed by any court of competent jurisdiction,’ leaving the question of such competency to depend on other provisions of law.” Id., at 617. See also Shoshone Mining Co. v. Rutter, 177 U. S. 505 –507 (1900) (statute authorizing suit “ ‘in a court of competent jurisdiction’ . . . unquestionably meant that the competency of the court should be determined by rules theretofore prescribed in respect to the jurisdiction of the Federal courts”). Califano v. Sanders, 430 U. S. 99 (1977) , provides another. It held that §10 of the Administrative Procedure Act, codified in 5 U. S. C. §§701–704, did not contain “an implied grant of subject-matter jurisdiction to review agency actions.” 430 U. S., at 105. In noting that “the actual text . . . nowhere contains an explicit grant of jurisdiction,” the Court pointed to two clauses requiring “judicial review . . . to proceed ‘in a court specified by statute’ or ‘in a court of competent jurisdiction’ ” and stated that both “seem to look to outside sources of jurisdictional authority.” Id., at 105–106, and n. 6. On this understanding, Fannie Mae’s sue-and-be-sued clause is most naturally read not to grant federal courts subject-matter jurisdiction over all cases involving Fannie Mae. In authorizing Fannie Mae to sue and be sued “in any court of competent jurisdiction, State or Federal,” it permits suit in any state or federal court already endowed with subject-matter jurisdiction over the suit. C Red Cross does not require a different result. Some, including the lower courts here, have understood it to set out a rule that an express reference to the federal courts suffices to make a sue-and-be-sued clause a grant of fed-eral jurisdiction. Red Cross contains no such rule. By its own terms, the rule Red Cross restates is “the basic rule” drawn in Deveaux and Osborn that a sue-and-be-sued clause conferring only a general right to sue does not grant jurisdiction to the federal courts. Red Cross, 505 U. S., at 253. Each mention of a “rule” refers back to this principle. See id., at 255 (reading this Court’s sue-and-be-sued clause cases to “support the rule that a . . . ‘sue and be sued’ provision may be read to confer federal court jurisdiction if, but only if, it specifically mentions the federal courts” (emphasis added)); id., at 256 (Bankers Trust applied “the rule thus established” to hold that the railroad’s sue-and-be-sued clause did not confer jurisdiction); 505 U. S., at 257 (finding the result “clear” under the “rule established in these cases” because the charter “expressly authoriz[es]” suits in federal courts in a clause “in all relevant respects identical” to one already found to confer jurisdiction). True enough, the dissent thought Red Cross established a broad rule. See 505 U. S., at 271–272 (opinion of Scalia, J.) (describing Red Cross as announcing a “rule . . . that any grant of a general capacity to sue with mention of federal courts will suffice to confer jurisdiction” (emphasis deleted)). The certainty of the dissent may explain the lower court decisions adopting a broader reading of Red Cross. But Red Cross itself establishes no such rule. And such a rule is hard to square with the opinion’s thorough consideration of the contrary arguments based in text, purpose, and legislative history. See id., at 258–263. Nothing in Red Cross suggests that courts should ignore “the ordinary sense of the language used,” id., at 263, when confronted with a federal charter’s sue-and-be-sued clause that expressly references the federal courts, but only those that are courts “of competent jurisdiction.” III Fannie Mae, preferring to be in federal court, raises several arguments against reading its sue-and-be-sued clause as merely capacity conferring. None are persuasive. A Fannie Mae first offers several alternative readings of “court of competent jurisdiction.” It suggests that the phrase might refer to a court with personal jurisdiction over the parties before it, a court of proper venue, or a court of general, rather than specialized, jurisdiction. Brief for Respondents 41–45. At bottom, Fannie Mae’s efforts on this front are premised on the reading of Red Cross rejected above. In its view, an express reference to the federal courts suffices to confer subject-matter jurisdiction on federal courts. It sees its only remaining task as explaining why that would not render “court of competent jurisdiction” superfluous. See Tr. of Oral Arg. 29–30. But the fact that a sue-and-be-sued clause references the federal courts does not resolve the jurisdictional question. Thus, arguments as to why the phrase “court of competent jurisdiction” could still have meaning if it does not carry its ordinary meaning are beside the point. Moreover, even if the phrase carries additional meaning, that would not further Fannie Mae’s argument. Take its suggestion that a “court of competent jurisdiction” is a court with personal jurisdiction. A court must have the power to decide the claim before it (subject-matter jurisdiction) and power over the parties before it (personal jurisdiction) before it can resolve a case. See Ruhrgas AG v. Marathon Oil Co., 526 U. S. 574 –585 (1999). Recognizing as much, this Court has stated that the phrase “court of competent jurisdiction,” while “usually used to refer to subject-matter jurisdiction, has also been used on occasion to refer to a court’s jurisdiction over the defendant’s person.” United States v. Morton, 467 U. S. 822, 828 (1984) (footnote omitted). See also Blackmar v. Guerre, 342 U. S. 512, 516 (1952) . But nothing in Fannie Mae’s sue-and-be-sued clause suggests that the reference to “court of competent jurisdiction” refers only to a court with personal jurisdiction over the parties before it. At most then, this point might support reading the phrase to refer to both subject-matter and personal jurisdiction. That does not help Fannie Mae. So long as the sue-and-be-sued clause refers to an outside source of subject-matter jurisdiction, it does not confer subject-matter jurisdiction. B Fannie Mae next claims that, by the time its sue-and-be-sued clause was enacted in 1954, courts had interpreted provisions containing the phrase “court of competent jurisdiction” to grant jurisdiction and that Congress was entitled to rely on those interpretations. This argument invokes the prior construction canon of statutory interpretation. The canon teaches that if courts have settled the meaning of an existing provision, the enactment of a new provision that mirrors the existing statutory text indicates, as a general matter, that the new provision has that same meaning. See Bragdon v. Abbott, 524 U. S. 624, 645 (1998) . Fannie Mae points to cases discussing three types of statutory provisions that, in its view, show that the phrase “court of competent jurisdiction” had acquired a settled meaning by 1954. The first pair addresses the FHA’s sue-and-be-sued clause. See 12 U. S. C. §1702 (“sue and be sued in any court of competent jurisdiction, State or Federal”). Two Court of Appeals decisions in the 1940’s concluded that the FHA sue-and-be-sued clause overrode the general rule, today found in 28 U. S. C. §§1346(a)(2), 1491, that monetary claims against the United States exceeding $10,000 must be brought in the Court of Federal Claims, rather than the federal district courts. See Ferguson v. Union Nat. Bank of Clarksburg, 126 F. 2d 753, 755–757 (CA4 1942); George H. Evans & Co. v. United States, 169 F. 2d 500, 502 (CA3 1948). These courts did not state that their jurisdiction was founded on the sue-and-be-sued clause, as opposed to statutes governing the original jurisdiction of the federal district courts. See, e.g., 28 U. S. C. §41(a) (1946 ed.). Thus, even assuming that two appellate court cases can “ ‘settle’ ” an issue, A. Scalia & B. Garner, Reading Law 325 (2012), these two cases did not because they did not speak to the question here. The second set of cases addresses provisions authorizing suit for a violation of a statute. One arose under the Fair Labor Standards Act of 1938, which authorizes employees to sue for violations of the Act in “any . . . court of competent jurisdiction.” §6(d)(1), 88Stat. 61, 29 U. S. C. §216(b). This Court, in its description of the facts, stated that “[j]urisdiction of the action was conferred by . . . 28 U. S. C. §41(8), and . . . 29 U. S. C. §216(b).” Williams v. Jacksonville Terminal Co., 315 U. S. 386, 390 (1942) . This brief, ambiguous statement did not settle the meaning of §216(b), and thus did not settle the meaning of the phrase “court of competent jurisdiction.” The other cases in this set dealt with the Housing and Rent Act of 1947. As enacted, the statute permitted suit in “any Federal, State, or Territorial court of competent jurisdiction.” §206(b), 61Stat. 199. Some courts read §206 not to confer jurisdiction and instead assessed their jurisdiction under the federal-question jurisdiction statute. See, e.g., Schuman v. Greenberg, 100 F. Supp. 187, 189 (NJ 1951) (collecting cases). At the time, that statute carried an amount-in-controversy requirement, 28 U. S. C. §41(1) (1946 ed.), and so some cases were dismissed or remanded to state court for lack of federal jurisdiction. Congress later amended §206 to permit suit “in any Federal court of competent jurisdiction regardless of the amount involved.” Defense Production Act Amendments of 1951, §204, 65Stat. 147. Congress’ elimination of the amount-in-controversy requirement suggests, if anything, it understood that “court of competent jurisdiction” could be read to require an outside source of jurisdiction. The third set of cases interpreted provisions making federal jurisdiction over certain causes of action exclusive. Brief for Respondents 36–37. Those cases confirm that the provisions require suit to be brought in federal courts but do not discuss the basis for federal jurisdiction. In sum, none of the cases on which Fannie Mae relies suggest that Congress in 1954 would have surveyed the jurisprudential landscape and necessarily concluded that the courts had already settled the question whether a sue-and-be-sued clause containing the phrase “court of competent jurisdiction” confers jurisdiction on the federal courts. C Fannie Mae ends with an appeal to congressional purpose, or, more accurately, a lack of congressional purpose. It argues that its original sue-and-be-sued clause, enacted in 1934, granted jurisdiction to federal courts and that there is no indication that Congress wanted to change the status quo in 1954. The addition in 1954 of “court of competent jurisdiction,” a phrase that, as discussed, carries a clear meaning, means that the current sue-and-be-sued clause does not confer jurisdiction. An indication whether that meaning was understood as a change from the 1934 Act is not required.[1]* Fannie Mae next points to its sibling rival, the Federal Home Loan Mortgage Corporation, known as Freddie Mac. The two share parallel authority to compete in the secondary mortgage market. Compare 12 U. S. C. §§1717(b)(2)–(6) (Fannie Mae) with §1454(a) (Freddie Mac). Suits involving Freddie Mac may be brought in federal court. See §1452(c) (“to sue and be sued, complain and defend, in any State, Federal, or other court”); §1452(f) (providing that Freddie Mac is a federal agency under 28 U. S. C. §§1345, 1442, that civil actions to which Freddie Mac is a party arise under federal law, and that Freddie Mac may remove cases to federal district court before trial). Fannie Mae argues there is no good reason to think that Congress gave Freddie Mac fuller access to the federal courts than it has. Leaving aside the clear textual indications suggesting Congress did just that, a plausible reason does exist. In 1970, when Freddie Mac’s sue-and-be-sued clause and related jurisdictional provisions were enacted, Freddie Mac was a Government-owned corporation. See Emergency Home Finance Act of 1970, §304(a), 84Stat. 454. Fannie Mae, on the other hand, had already transitioned into a privately owned corporation. Fannie Mae’s argument on this front, moreover, contains a deeper flaw. The doors to federal court remain open to Fannie Mae through diversity and federal-question jurisdiction. Fannie Mae provides no reason to think that in other cases, involving only state-law claims, access to the federal courts gives Freddie Mac an unintended competitive advantage over Fannie Mae that Congress would have wanted to avoid. Indeed, the usual assumption is that state courts are up to the task of adjudicating their own laws. Cf. Gulf Offshore Co. v. Mobil Oil Corp., 453 U. S. 473 –484 (1981). IV The judgment of the Ninth Circuit is reversed. It is so ordered.Notes 1 * The legislative history of the 1934 Act provides some reason to question Fannie Mae’s premise about Congress’ view of the status quo under the 1934 Act. During debate on this provision, Senator Logan asked Senator Bulkley, the chair of the subcommittee with authority over the bill, about the original sue-and-be-sued clause. Senator Bulkley explained that it merely conferred a capacity to sue and be sued “and [did] not confe[r] a right to go into a Federal court where it would not otherwise exist.” 78 Cong. Rec. 12008 (1934).

581.US.2016_15-7250

After federal agents found child pornography on petitioner’s computer, he pleaded guilty to possessing a visual depiction of a minor engaging in sexually explicit conduct, in violation of 18 U. S. C. §§2252(a)(4)(B) and (b)(2), an offense requiring a district court to “make restitution to the victim of the offense,” §3663A(a)(1). The District Court entered an initial judgment sentencing petitioner to a term of imprisonment. It also acknowledged that restitution was mandatory but deferred determination of the restitution amount. Petitioner filed a notice of appeal from this initial judgment. Months later, the District Court entered an amended judgment, ordering petitioner to pay restitution to one of his victims. Petitioner did not file a second notice of appeal from the amended judgment. When he nonetheless challenged the restitution amount before the Eleventh Circuit, the Government argued that he had forfeited his right to do so by failing to file a second notice of appeal. The Eleventh Circuit agreed, holding that petitioner could not challenge the restitution amount. Held: A defendant wishing to appeal an order imposing restitution in a deferred restitution case must file a notice of appeal from that order. If he fails to do so and the Government objects, he may not challenge the restitution order on appeal. Pp. 3–9. (a) Both 18 U. S. C. §3742(a), which governs criminal appeals, and Federal Rule of Appellate Procedure 3(a)(1) contemplate that a defendant will file a notice of appeal after the district court has decided the issue sought to be appealed. Here, petitioner filed only one notice of appeal, which preceded by many months the sentence and judgment imposing restitution. He therefore failed to properly appeal the amended judgment. Whether or not the requirement that a defendant file a timely notice of appeal from an amended judgment imposing restitution is a jurisdictional prerequisite, it is at least a mandatory claim-processing rule, which is “unalterable” if raised properly by the party asserting a violation of the rule. Eberhart v. United States, 546 U. S. 12 . Because the Government timely raised the issue, “the court’s duty to dismiss the appeal was mandatory.” Id., at 18. Pp. 3–5. (b) Petitioner’s argument that his single notice of appeal sufficed under the Federal Rules to appeal both judgments depends on two premises: First, in a deferred restitution case, there is only one “judgment,” as that term is used in Rules 4(b)(1) and (b)(2); and second, so long as a notice of appeal is filed after the initial judgment, it “springs forward” under Rule 4(b)(2) to appeal the amended judgment imposing restitution. Each premise is rejected. Pp. 5–7. (1) This Court’s analysis in Dolan v. United States, 560 U. S. 605 , makes clear that deferred restitution cases involve two appealable judgments, not one. The Dolan Court did not decide the question presented here, but the Court was not persuaded by the argument that “a sentencing judgment is not ‘final’ until it contains a definitive determination of the amount of restitution.” Id., at 617–618. Instead, the Court recognized, “strong arguments” supported the proposition that both the initial judgment and the restitution order were each immediately appealable final judgments. Ibid. Pp. 5–6. (2) Because petitioner’s notice of appeal was filed well before the District Court announced the sentence imposing restitution, the notice of appeal did not “spring forward” to become effective on the date the court entered its amended restitution judgment. By its own terms, Rule 4(b)(2) applies only to a notice of appeal filed after a sentence has been announced and before the judgment imposing the sentence is entered on the docket. Even if the District Court’s acknowledgment in the initial judgment that restitution was mandatory could qualify as a “sentence” that the District Court “announced” under Rule 4(b)(2), petitioner has never disputed that restitution is mandatory for his offense. Rather, he argued on appeal that the amount imposed is unlawful. Pp. 6–7. (c) Petitioner’s alternative argument that any defect in his notice of appeal should be overlooked as harmless error is rejected. Lemke v. United States, 346 U. S. 325 , on which he relies, has been superseded by the Federal Rules of Appellate Procedure in two ways. First, the Lemke petitioner’s notice of appeal would now be timely under Rule 4(b)(2). Petitioner in this case cannot take advantage of that Rule. Second, Rule 3(a)(2) now provides the consequences for litigant errors associated with filing a notice of appeal. The court of appeals may, in its discretion, overlook defects in a notice of appeal other than the failure to timely file a notice. It may not overlook the failure to file a notice of appeal at all. Pp. 8–9. 618 Fed. Appx. 579, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Alito, and Kagan, JJ., joined. Ginsburg, J., filed a dissenting opinion, in which Sotomayor, J., joined. Gorsuch, J., took no part in the consideration or decision of the case.

Sentencing courts are required to impose restitution as part of the sentence for specified crimes. But the amount to be imposed is not always known at the time of sentencing. When that is the case, the court may enter an initial judgment imposing certain aspects of a defendant’s sentence, such as a term of imprisonment, while deferring a determination of the amount of restitution until entry of a later, amended judgment. We must decide whether a single notice of appeal, filed between the initial judgment and the amended judgment, is sufficient to invoke appellate review of the later-determined restitution amount. We hold that it is not, at least where, as here, the Government objects to the defendant’s failure to file a notice of appeal following the amended judgment. I After federal agents found more than 300 files containing child pornography on his computer, petitioner Marcelo Manrique pleaded guilty to possessing a visual depiction of a minor engaging in sexually explicit conduct, in violation of 18 U. S. C. §§2252(a)(4)(B) and (b)(2). Under the Mandatory Victims Restitution Act of 1996 (MVRA), the District Court was required to order petitioner to “make restitution to the victim of the offense.” §3663A(a)(1); see §§2259(a), (b)(2) (“An order of restitution under this section shall be issued and enforced in accordance with [§]3664 in the same manner as an order under [§]3663A”). On June 24, 2014, the District Court entered an initial judgment sentencing petitioner to 72 months of imprisonment and a life term of supervised release. At the sentencing hearing, the court acknowledged that restitution was mandatory. But, consistent with the MVRA, the court postponed determining the victims’ damages, which had not yet been ascertained. See, e.g., §3664(d)(5); Dolan v. United States, 560 U. S. 605 –608 (2010). Accordingly, the judgment expressly deferred “determination of restitution” and noted that an “Amended Judgment . . . w[ould] be entered after such determination.” App. 39. On July 8, petitioner filed a notice of appeal “from the final judgment and sentence entered in this action on the 24th day of June, 2014.” Id., at 42. The District Court held a restitution hearing on September 17, 2014. Only one of the victims sought restitution. The court ordered petitioner to pay $4,500 in restitution to her and entered an amended judgment the next day imposing that sentence. Petitioner did not file a second notice of appeal from the court’s order imposing restitution or from the amended judgment. Notwithstanding his failure to file a second notice of appeal, petitioner challenged the restitution amount before the Eleventh Circuit, arguing in his brief that the Government had not shown he was the proximate cause of the victim’s injuries and that the restitution amount bore no rational relationship to the damages she claimed. The Government countered that petitioner had forfeited his right to challenge the restitution amount by failing to file a second notice of appeal. The Court of Appeals agreed that petitioner could not challenge the restitution amount and declined to consider his challenge. 618 Fed. Appx. 579, 583–584 (CA11 2015) ( per curiam). We granted certiorari, 578 U. S. ___ (2016), and now affirm. II A To secure appellate review of a judgment or order, a party must file a notice of appeal from that judgment or order. Filing a notice of appeal transfers adjudicatory authority from the district court to the court of appeals. The statute that governs appeals of criminal sentences, 18 U. S. C. §3742(a), provides that a “defendant may file a notice of appeal in the district court for review of an otherwise final sentence” in certain specified circumstances. See United States v. Ruiz, 536 U. S. 622 −628 (2002). And Federal Rule of Appellate Procedure 3(a)(1) specifies that “[a]n appeal permitted by law as of right . . . may be taken only by filing a notice of appeal with the district clerk within the time allowed by Rule 4.” (Emphasis added.) Both §3742(a) and Rule 4 contemplate that the defendant will file the notice of appeal after the district court has decided the issue sought to be appealed. Section 3742(a)(1) permits the defendant to file a notice of appeal of a sentence that “was imposed in violation of law.” (Emphasis added.) And Rule 4(b)(1)(A)(i) provides generally that, “[i]n a criminal case, a defendant’s notice of appeal must be filed in the district court within 14 days after . . . the entry of either the judgment or the order being appealed.” (Emphasis added.) Petitioner filed only one notice of appeal, which preceded by many months the sentence and judgment imposing restitution. His notice of appeal could not have been “for review” of the restitution order, §3742(a), and it was not filed within the timeframe allowed by Rule 4. He thus failed to properly appeal under the statute and the Rules the amended judgment imposing restitution. The Government contends that filing a notice of appeal from the judgment imposing restitution is a jurisdictional prerequisite to securing appellate review of the restitution amount. See, e.g., Brief for United States 28–31. This position follows, according to the Government, from many of our cases emphasizing the “jurisdictional significance” of a notice of appeal. E.g., Griggs v. Provident Consumer Discount Co., 459 U. S. 56, 58 (1982) ( per curiam). Because the notice of appeal is jurisdictional, the Government explains, the Court of Appeals was required to dismiss petitioner’s appeal regardless of whether the Government raised the issue. We do not need to decide in this case whether the Government is correct. The requirement that a defendant file a timely notice of appeal from an amended judgment imposing restitution is at least a mandatory claim-processing rule. See Greenlaw v. United States, 554 U. S. 237 –253 (2008); see also Rule 3(a)(2) (“An appellant’s failure to take any step other than the timely filing of a notice of appeal does not affect the validity of the appeal, but is ground only for the court of appeals to act as it considers appropriate, including dismissing the appeal” (emphasis added)). Mandatory claim-processing rules “seek to promote the orderly progress of litigation by requiring that the parties take certain procedural steps at certain specified times.” Henderson v. Shinseki, 562 U. S. 428, 435 (2011) . Unlike jurisdictional rules, mandatory claim-processing rules may be forfeited “if the party asserting the rule waits too long to raise the point.” Eberhart v. United States, 546 U. S. 12, 15 (2005) ( per curiam) (internal quotation marks omitted). If a party “properly raise[s] them,” however, they are “unalterable.” Id., at 15, 19. The Government timely raised petitioner’s failure to file a notice of appeal from the amended judgment imposing restitution before the Court of Appeals. See Brief for United States in No. 14–13029 (CA11), pp. 22–25 (arguing that petitioner “waived his right to appeal the district court’s order of restitution by failing to file a notice of appeal from that order” (capitalization omitted)). Accordingly, “the court’s duty to dismiss the appeal was mandatory.” Eberhart, supra, at 18. B Petitioner disputes this conclusion, arguing that his single notice of appeal sufficed under the Rules to appeal both the initial judgment and the amended judgment imposing restitution. As we understand it, his argument depends on two premises: First, in a deferred restitution case, there is only one “judgment,” as that term is used in Rules 4(b)(1) and (b)(2); and second, so long as a notice of appeal is filed after the initial judgment, it “springs forward” under Rule 4(b)(2) to appeal the amended judgment imposing restitution. We reject each of these premises. 1 Petitioner argues that the initial judgment deferring restitution and the amended judgment imposing a specific restitution amount merge to become “the judgment” referenced in the Federal Rules. See Rule 4(b)(1)(A)(i) (notice of appeal must be filed within 14 days after “the entry of . . . the judgment . . . being appealed”); Rule 4(b)(2) (“Filing Before Entry of Judgment”). He argues that his notice of appeal, which was filed within 14 days of the initial judgment, was therefore sufficient to invoke appellate review of the merged judgment. Petitioner’s approach is inconsistent with our reasoning in Dolan, 560 U. S. 605 . The petitioner in that case argued that the amended judgment imposing restitution is the only final, appealable judgment in a deferred restitution case. See id., at 616. Although we did not decide “whether or when a party can, or must, appeal”—the question presented here—we were not persuaded by the argument that “a sentencing judgment is not ‘final’ until it contains a definitive determination of the amount of restitution.” Id., at 617–618. To the contrary, we recognized “strong arguments” supporting the proposition that both the “initial judgment [that] imposed a sentence of imprisonment and supervised release” and the subsequent “ ‘sentence that impose[d] an order of restitution’ ” were each immediately appealable final judgments. Ibid. (citing 18 U. S. C. §§3582(b) (imprisonment), 3583(a) (supervised release), and 3664(o) (restitution)). Consequently, we were not surprised “to find instances where a defendant ha[d] appealed from the entry of a judgment containing an initial sentence that includes a term of imprisonment” and “subsequently appealed from a later order setting forth the final amount of restitution.” 560 U. S., at 618. Our analysis in Dolan thus makes clear that deferred restitution cases involve two appealable judgments, not one.[1] 2 Petitioner’s reliance on Rule 4(b)(2) is also misplaced. That Rule provides that a “notice of appeal filed after the court announces a decision, sentence, or order—but before the entry of the judgment or order—is treated as filed on the date of and after the entry.” A prematurely filed notice of appeal will become effective under the Rule to challenge a later-entered judgment in some circumstances. As this Court explained in construing Rule 4(a)(2)’s parallel provision for civil cases, the Rule “was intended to protect the unskilled litigant who files a notice of appeal from a decision that he reasonably but mistakenly believes to be a final judgment, while failing to file a notice of appeal from the actual final judgment.” FirsTier Mortgage Co. v. Investors Mortgage Ins. Co., 498 U. S. 269, 276 (1991) . By its own terms, however, Rule 4(b)(2) applies only to a notice of appeal filed after a sentence has been “announce[d]” and before the judgment imposing the sentence is entered on the docket. See Rule 4(b)(6) (“A judgment or order is entered for purposes of this Rule 4(b) when it is entered on the criminal docket”). If the court has not yet decided the issue that the appellant seeks to appeal, then the Rule does not come into play. Accordingly, it does not apply where a district court enters an initial judgment deferring restitution and subsequently amends the judgment to include the sentence of restitution. By deferring restitution, the court is declining to announce a sentence. When petitioner filed his notice of appeal in this case, the District Court had observed only that restitution was “mandatory.” App. 27. The court did not announce the restitution amount (or even hold a hearing on the issue) until months later. Even if describing restitution as mandatory could qualify as a “sentence” that the District Court “announced” under Rule 4(b)(2), petitioner has never disputed that restitution is mandatory for his offense. Rather, he argued on appeal that the amount of the restitution imposed—an issue the court did not consider until months later—is unlawful. Because petitioner’s notice of appeal was filed well before the District Court announced the sentence imposing $4,500 in restitution, the notice of appeal did not “spring forward” to become effective on the date the court entered its amended judgment imposing that sentence. C Finally, petitioner argues in the alternative that any defect in his notice of appeal should be overlooked as harmless error, citing Lemke v. United States, 346 U. S. 325 (1953) (per curiam). In that case, the petitioner filed a notice of appeal the day after his sentence was announced but three days before the judgment was entered. Id., at 326. His notice of appeal was dismissed as premature under Federal Rule of Criminal Procedure 37(a)(2), which then governed notices of appeal in criminal cases. This Court reversed on the ground that the premature filing was harmless error under Rule 52(a). Ibid. The Court’s holding in Lemke does not apply to petitioner’s failure to file a notice of appeal from the amended judgment. Lemke has been superseded by the Federal Rules of Appellate Procedure in two ways. First, the Lemke petitioner’s notice of appeal would now be timely under Rule 4(b)(2). As discussed in Part II–B–2, supra, petitioner here cannot take advantage of that rule. Second, Rule 3(a)(2) now provides the consequences for litigant errors associated with filing a notice of appeal. The court of appeals may, in its discretion, overlook defects in a notice of appeal other than the failure to timely file a notice. It may not overlook the failure to file a notice of appeal at all. The filing of a notice of appeal from an amended judgment imposing restitution is at least a mandatory claim-processing rule, Part II–A, supra, meaning that the requirement to file such a notice is unalterable, so long as the opposing party raises the issue. By definition, mandatory claim-processing rules, although subject to forfeiture, are not subject to harmless-error analysis. Petitioner in this case did not file a defective notice of appeal from the amended judgment imposing restitution, but rather failed altogether to file a notice of appeal from the amended judgment. Courts do not have discretion to overlook such an error, at least where it is called to their attention. * * * We hold that a defendant who wishes to appeal an order imposing restitution in a deferred restitution case must file a notice of appeal from that order. Because petitioner failed to do so, and the Government objected, the Court of Appeals properly declined to consider his challenge to the amount of restitution imposed. The judgment of the Court of Appeals, accordingly, is affirmed. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 We do not intend to call into question this Court’s decision in Corey v. United States, 375 U. S. 169, 176 (1963) (holding that a defendant may challenge his conviction after a single notice of appeal filed from a final sentence imposed under §4208(b)).

580.US.2016_14-9496

During a traffic stop, police officers in Joliet, Illinois, searched petitioner Elijah Manuel and found a vitamin bottle containing pills. Suspecting the pills to be illegal drugs, the officers conducted a field test, which came back negative for any controlled substance. Still, they arrested Manuel and took him to the police station. There, an evidence technician tested the pills and got the same negative result, but claimed in his report that one of the pills tested “positive for the probable presence of ecstasy.” App. 92. An arresting officer also reported that, based on his “training and experience,” he “knew the pills to be ecstasy.” Id., at 91. On the basis of those false statements, another officer filed a sworn complaint charging Manuel with unlawful possession of a controlled substance. Relying exclusively on that complaint, a county court judge found probable cause to detain Manuel pending trial. While Manuel was in jail, the Illinois police laboratory tested the seized pills and reported that they contained no controlled substances. But Manuel remained in custody, spending a total of 48 days in pretrial detention. More than two years after his arrest, but less than two years after his criminal case was dismissed, Manuel filed a 42 U. S. C. §1983 lawsuit against Joliet and several of its police officers (collectively, the City), alleging that his arrest and detention violated the Fourth Amendment. The District Court dismissed Manuel’s suit, holding, first, that the applicable two-year statute of limitations barred his unlawful arrest claim, and, second, that under binding Circuit precedent, pretrial detention following the start of legal process (here, the judge’s probable-cause determination) could not give rise to a Fourth Amendment claim. Manuel appealed the dismissal of his unlawful detention claim; the Seventh Circuit affirmed. Held: 1. Manuel may challenge his pretrial detention on Fourth Amendment grounds. This conclusion follows from the Court’s settled precedent. In Gerstein v. Pugh, 420 U. S. 103 , the Court decided that a pretrial detention challenge was governed by the Fourth Amendment, noting that the Fourth Amendment establishes the minimum constitutional “standards and procedures” not just for arrest but also for “detention,” id., at 111, and “always has been thought to define” the appropriate process “for seizures of person[s] . . . in criminal cases, including the detention of suspects pending trial,” id., at 125, n. 27. And in Albright v. Oliver, 510 U. S. 266 , a majority of the Court again looked to the Fourth Amendment to assess pretrial restraints on liberty. Relying on Gerstein, the plurality reiterated that the Fourth Amendment is the “relevan[t]” constitutional provision to assess the “deprivations of liberty that go hand in hand with criminal prosecutions.” Id., at 274; see id., at 290 (Souter, J., concurring in judgment) (“[R]ules of recovery for such harms have naturally coalesced under the Fourth Amendment”). That the pretrial restraints in Albright arose pursuant to legal process made no difference, given that they were allegedly unsupported by probable cause. As reflected in those cases, pretrial detention can violate the Fourth Amendment not only when it precedes, but also when it follows, the start of legal process. The Fourth Amendment prohibits government officials from detaining a person absent probable cause. And where legal process has gone forward, but has done nothing to satisfy the probable-cause requirement, it cannot extinguish a detainee’s Fourth Amendment claim. That was the case here: Because the judge’s determination of probable cause was based solely on fabricated evidence, it did not expunge Manuel’s Fourth Amendment claim. For that reason, Manuel stated a Fourth Amendment claim when he sought relief not merely for his arrest, but also for his pretrial detention. Pp. 6–10. 2. On remand, the Seventh Circuit should determine the claim’s accrual date, unless it finds that the City has previously waived its timeliness argument. In doing so, the court should look to the common law of torts for guidance, Carey v. Piphus, 435 U. S. 247 –258, while also closely attending to the values and purposes of the constitutional right at issue. The court may also consider any other still-live issues relating to the elements of and rules applicable to Manuel’s Fourth Amendment claim. Pp. 11–15. 590 Fed. Appx. 641, reversed and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined. Thomas, J., filed a dissenting opinion. Alito, J., filed a dissenting opinion, in which Thomas, J., joined.

Petitioner Elijah Manuel was held in jail for some seven weeks after a judge relied on allegedly fabricated evidence to find probable cause that he had committed a crime. The primary question in this case is whether Manuel may bring a claim based on the Fourth Amendment to contest the legality of his pretrial confinement. Our answer follows from settled precedent. The Fourth Amendment, this Court has recognized, establishes “the standards and procedures” governing pretrial detention. See, e.g., Gerstein v. Pugh, 420 U. S. 103, 111 (1975) . And those constitutional protections apply even after the start of “legal process” in a criminal case—here, that is, after the judge’s determination of probable cause. See Albright v. Oliver, 510 U. S. 266, 274 (1994) (plurality opinion); id., at 290 (Souter, J., concurring in judgment). Accordingly, we hold today that Manuel may challenge his pretrial detention on the ground that it violated the Fourth Amendment (while we leave all other issues, including one about that claim’s timeliness, to the court below). I Shortly after midnight on March 18, 2011, Manuel was riding through Joliet, Illinois, in the passenger seat of a Dodge Charger, with his brother at the wheel. A pair of Joliet police officers pulled the car over when the driver failed to signal a turn. See App. 90. According to the complaint in this case, one of the officers dragged Manuel from the car, called him a racial slur, and kicked and punched him as he lay on the ground. See id., at 31–32, 63.[1] The policeman then searched Manuel and found a vitamin bottle containing pills. See id., at 64. Suspecting that the pills were actually illegal drugs, the officers conducted a field test of the bottle’s contents. The test came back negative for any controlled substance, leaving the officers with no evidence that Manuel had committed a crime. See id., at 69. Still, the officers arrested Manuel and took him to the Joliet police station. See id., at 70. There, an evidence technician tested the pills once again, and got the same (negative) result. See ibid. But the technician lied in his report, claiming that one of the pills was “found to be . . . positive for the probable presence of ecstasy.” Id., at 92. Similarly, one of the arresting officers wrote in his report that “[f ]rom [ his] training and experience, [ he] knew the pills to be ecstasy.” Id., at 91. On the basis of those statements, another officer swore out a criminal complaint against Manuel, charging him with unlawful possession of a controlled substance. See id., at 52–53. Manuel was brought before a county court judge later that day for a determination of whether there was probable cause for the charge, as necessary for further detention. See Gerstein, 420 U. S., at 114 (requiring a judicial finding of probable cause following a warrantless arrest to impose any significant pretrial restraint on liberty); Ill. Comp. Stat., ch. 725, §5/109–1 (West 2010) (implementing that constitutional rule). The judge relied exclusively on the criminal complaint—which in turn relied exclusively on the police department’s fabrications—to support a finding of probable cause. Based on that determination, he sent Manuel to the county jail to await trial. In the somewhat obscure legal lingo of this case, Manuel’s subsequent detention was thus pursuant to “legal process”—because it followed from, and was authorized by, the judge’s probable-cause determination.[2] While Manuel sat in jail, the Illinois police laboratory reexamined the seized pills, and on April 1, it issued a report concluding (just as the prior two tests had) that they contained no controlled substances. See App. 51. But for unknown reasons, the prosecution—and, critically for this case, Manuel’s detention—continued for more than another month. Only on May 4 did an Assistant State’s Attorney seek dismissal of the drug charge. See id., at 48, 101. The County Court immediately granted the request, and Manuel was released the next day. In all, he had spent 48 days in pretrial detention. On April 22, 2013, Manuel brought this lawsuit under 42 U. S. C. §1983 against the City of Joliet and several of its police officers (collectively, the City). Section 1983 creates a “species of tort liability,” Imbler v. Pachtman, 424 U. S. 409, 417 (1976) , for “the deprivation of any rights, privileges, or immunities secured by the Constitution,” §1983. Manuel’s complaint alleged that the City violated his Fourth Amendment rights in two ways—first by arresting him at the roadside without any reason, and next by “detaining him in police custody” for almostseven weeks based entirely on made-up evidence. See App. 79–80.[3] The District Court dismissed Manuel’s suit. See 2014 WL 551626 (ND Ill., Feb. 12, 2014). The court first held that the applicable two-year statute of limitations barred Manuel’s claim for unlawful arrest, because more than two years had elapsed between the date of his arrest (March 18, 2011) and the filing of his complaint (April 22, 2013). But the court relied on another basis in rejecting Manuel’s challenge to his subsequent detention (which stretched from March 18 to May 5, 2011). Binding Circuit precedent, the District Court explained, made clear that pretrial detention following the start of legal process could not give rise to a Fourth Amendment claim. See id., at *1 (citing, e.g., Newsome v. McCabe, 256 F. 3d 747, 750 (CA7 2001)). According to that line of decisions, a §1983 plaintiff challenging such detention must allege a breach of the Due Process Clause—and must show, to recover on that theory, that state law fails to provide an adequate remedy. See 2014 WL 551626, at *1–*2. Because Manuel’s complaint rested solely on the Fourth Amendment—and because, in any event, Illinois’s remedies were robust enough to preclude the due process avenue—the District Court found that Manuel had no way to proceed. See ibid. The Court of Appeals for the Seventh Circuit affirmed the dismissal of Manuel’s claim for unlawful detention (the only part of the District Court’s decision Manuel appealed). See 590 Fed. Appx. 641 (2015). Invoking its prior caselaw, the Court of Appeals reiterated that such claims could not be brought under the Fourth Amendment. Once a person is detained pursuant to legal process, the court stated, “the Fourth Amendment falls out of the picture and the detainee’s claim that the detention is improper becomes [one of] due process.” Id., at 643–644 (quoting Llovet v. Chicago, 761 F. 3d 759, 763 (CA7 2014)). And again: “When, after the arrest[,] a person is not let go when he should be, the Fourth Amendment gives way to the due process clause as a basis for challenging his detention.” 590 Fed. Appx., at 643 (quoting Llovet, 761 F. 3d, at 764). So the Seventh Circuit held that Manuel’s complaint, in alleging only a Fourth Amendment violation, rested on the wrong part of the Constitution: A person detained following the onset of legal process could at most (although, the court agreed, not in Illinois) challenge his pretrial confinement via the Due Process Clause. See 590 Fed. Appx., at 643–644. The Seventh Circuit recognized that its position makes it an outlier among the Courts of Appeals, with ten others taking the opposite view. See id., at 643; Hernandez-Cuevas v. Taylor, 723 F. 3d 91, 99 (CA1 2013) (“[T]here is now broad consensus among the circuits that the Fourth Amendment right to be free from seizure but upon probable cause extends through the pretrial period”).[4] Still, the court decided, Manuel had failed to offer a sufficient reason for overturning settled Circuit precedent; his argument, albeit “strong,” was “better left for the Supreme Court.” 590 Fed. Appx., at 643. On cue, we granted certiorari. 577 U. S. ___ (2016). II The Fourth Amendment protects “[t]he right of the people to be secure in their persons . . . against unreasonable . . . seizures.” Manuel’s complaint seeks just that protection. Government officials, it recounts, detained—which is to say, “seiz[ed]”—Manuel for 48 days following his arrest. See App. 79–80; Brendlin v. California, 551 U. S. 249, 254 (2007) (“A person is seized” whenever officials “restrain[ ] his freedom of movement” such that he is “not free to leave”). And that detention was “unreason-able,” the complaint continues, because it was based solely on false evidence, rather than supported by probable cause. See App. 79–80; Bailey v. United States, 568 U. S. 186, 192 (2013) (“[T]he general rule [is] that Fourth Amendment seizures are ‘reasonable’ only if based on probable cause to believe that the individual has committed a crime”). By their respective terms, then, Manuel’s claim fits the Fourth Amendment, and the Fourth Amendment fits Manuel’s claim, as hand in glove. This Court decided some four decades ago that a claim challenging pretrial detention fell within the scope of the Fourth Amendment. In Gerstein, two persons arrested without a warrant brought a §1983 suit complaining that they had been held in custody for “a substantial period solely on the decision of a prosecutor.” 420 U. S., at 106. The Court looked to the Fourth Amendment to analyze—and uphold—their claim that such a pretrial restraint on liberty is unlawful unless a judge (or grand jury) first makes a reliable finding of probable cause. See id., at 114, 117, n. 19. The Fourth Amendment, we began, establishes the minimum constitutional “standards and procedures” not just for arrest but also for ensuing “detention.” Id., at 111. In choosing that Amendment “as the rationale for decision,” the Court responded to a concurring Justice’s view that the Due Process Clause offered the better framework: The Fourth Amendment, the majority countered, was “tailored explicitly for the criminal justice system, and it[ ] always has been thought to define” the appropriate process “for seizures of person[s] . . . in criminal cases, including the detention of suspects pending trial.” Id., at 125, n. 27. That Amendment, standing alone, guaranteed “a fair and reliable determination of probable cause as a condition for any significant pretrial restraint.” Id., at 125. Accordingly, those detained prior to trial without such a finding could appeal to “the Fourth Amendment’s protection against unfounded invasions of liberty.” Id., at 112; see id., at 114.[5] And so too, a later decision indicates, those objecting to a pretrial deprivation of liberty may invoke the Fourth Amendment when (as here) that deprivation occurs after legal process commences. The §1983 plaintiff in Albright complained of various pretrial restraints imposed after a court found probable cause to issue an arrest warrant, and then bind him over for trial, based on a policeman’s unfounded charges. See 510 U. S., at 268–269 (plurality opinion). For uncertain reasons, Albright ignored the Fourth Amendment in drafting his complaint; instead, he alleged that the defendant officer had infringed his substantive due process rights. This Court rejected that claim, with five Justices in two opinions remitting Albright to the Fourth Amendment. See id., at 271 (plurality opinion) (“We hold that it is the Fourth Amendment . . . under which [ his] claim must be judged”); id., at 290 (Souter, J., concurring in judgment) (“[I]njuries like those [ he] alleges are cognizable in §1983 claims founded upon . . . the Fourth Amendment”). “The Framers,” the plurality wrote, “considered the matter of pretrial deprivations of liberty and drafted the Fourth Amendment to address it.” Id., at 274. That the deprivations at issue were pursuant to legal process made no difference, given that they were (allegedly) unsupported by probable cause; indeed, neither of the two opinions so much as mentioned that procedural circumstance. Relying on Gerstein, the plurality stated that the Fourth Amendment remained the “relevan[t]” constitutional provision to assess the “deprivations of liberty”—most notably, pretrial detention—“that go hand in hand with criminal prosecutions.” 510 U. S., at 274; see id., at 290 (Souter, J., concurring in judgment) (“[R]ules of recovery for such harms have naturally coalesced under the Fourth Amendment”). As reflected in Albright’s tracking of Gerstein’s analysis, pretrial detention can violate the Fourth Amendment not only when it precedes, but also when it follows, the start of legal process in a criminal case. The Fourth Amendment prohibits government officials from detaining a person in the absence of probable cause. See supra, at 6. That can happen when the police hold someone without any reason before the formal onset of a criminal proceeding. But it also can occur when legal process itself goes wrong—when, for example, a judge’s probable-cause determination is predicated solely on a police officer’s false statements. Then, too, a person is confined without constitutionally adequate justification. Legal process has gone forward, but it has done nothing to satisfy the Fourth Amendment’s probable-cause requirement. And for that reason, it cannot extinguish the detainee’s Fourth Amendment claim—or somehow, as the Seventh Circuit has held, convert that claim into one founded on the Due Process Clause. See 590 Fed. Appx., at 643–644. If the complaint is that a form of legal process resulted in pretrial detention unsupported by probable cause, then the right allegedly infringed lies in the Fourth Amendment.[6] For that reason, and contrary to the Seventh Circuit’s view, Manuel stated a Fourth Amendment claim when he sought relief not merely for his (pre-legal-process) arrest, but also for his (post-legal-process) pretrial detention.[7] Consider again the facts alleged in this case. Police officers initially arrested Manuel without probable cause, based solely on his possession of pills that had field tested negative for an illegal substance. So (putting timeliness issues aside) Manuel could bring a claim for wrongful arrest under the Fourth Amendment. And the same is true (again, disregarding timeliness) as to a claim for wrongful detention—because Manuel’s subsequent weeks in custody were also unsupported by probable cause, and so also constitutionally unreasonable. No evidence of Manuel’s criminality had come to light in between the roadside arrest and the County Court proceeding initiating legal process; to the contrary, yet another test of Man-uel’s pills had come back negative in that period. Allthat the judge had before him were police fabrications about the pills’ content. The judge’s order holding Manuel for trial therefore lacked any proper basis. And that means Manuel’s ensuing pretrial detention, no less than his original arrest, violated his Fourth Amendment rights. Or put just a bit differently: Legal process did not expunge Manuel’s Fourth Amendment claim because the process he received failed to establish what that Amendment makes essential for pretrial detention—probable cause to believe he committed a crime.[8] III Our holding—that the Fourth Amendment governs a claim for unlawful pretrial detention even beyond the start of legal process—does not exhaust the disputed legal issues in this case. It addresses only the threshold inquiry in a §1983 suit, which requires courts to “identify the specific constitutional right” at issue. Albright, 510 U. S., at 271. After pinpointing that right, courts still must determine the elements of, and rules associated with, an action seeking damages for its violation. See, e.g., Carey v. Piphus, 435 U. S. 247 –258 (1978). Here, the parties particularly disagree over the accrual date of Manuel’s Fourth Amendment claim—that is, the date on which the applicable two-year statute of limitations began to run. The timeliness of Manuel’s suit hinges on the choice between their proposed dates. But with the following brief comments, we remand that issue to the court below. In defining the contours and prerequisites of a §1983 claim, including its rule of accrual, courts are to look first to the common law of torts. See ibid. (explaining that tort principles “provide the appropriate starting point” in specifying the conditions for recovery under §1983); Wallace v. Kato, 549 U. S. 384 –390 (2007) (same for accrual dates in particular). Sometimes, that review of common law will lead a court to adopt wholesale the rules that would apply in a suit involving the most analogous tort. See id., at 388–390; Heck v. Humphrey, 512 U. S. 477 –487 (1994). But not always. Common-law principles are meant to guide rather than to control the definition of §1983 claims, serving “more as a source of inspired examples than of prefabricated components.” Hartman v. Moore, 547 U. S. 250, 258 (2006) ; see Rehberg v. Paulk, 566 U. S. 356, 366 (2012) (noting that “§1983 is [not] simply a federalized amalgamation of pre-existing common-law claims”). In applying, selecting among, or adjust-ing common-law approaches, courts must closely attend to the values and purposes of the constitutional right at issue. With these precepts as backdrop, Manuel and the City offer competing views about what accrual rule should govern a §1983 suit challenging post-legal-process pretrial detention. According to Manuel, that Fourth Amendment claim accrues only upon the dismissal of criminal charges—here, on May 4, 2011, less than two years before he brought his suit. See Reply Brief 2; Brief for United States as Amicus Curiae 24–25, n. 16 (taking the same position). Relying on this Court’s caselaw, Manuel analogizes his claim to the common-law tort of malicious prosecution. See Reply Brief 9; Wallace, 549 U. S., at 389–390. An element of that tort is the “termination of the . . . proceeding in favor of the accused”; and accordingly, the statute of limitations does not start to run until that termination takes place. Heck, 512 U. S., at 484, 489. Man-uel argues that following the same rule in suits like his will avoid “conflicting resolutions” in §1983 litigation and criminal proceedings by “preclud[ing] the possibility of the claimant succeeding in the tort action after having been convicted in the underlying criminal prosecution.” Id., at 484, 486; see Reply Brief 10–11; Brief for United States as Amicus Curiae 24–25, n. 16. In support of Manuel’s position, all but two of the ten Courts of Appeals that have recognized a Fourth Amendment claim like his have incorporated a “favorable termination” element and so pegged the statute of limitations to the dismissal of the criminal case. See n. 4, supra.[9] That means in the great majority of Circuits, Manuel’s claim would be timely. The City, however, contends that any such Fourth Amendment claim accrues (and the limitations period starts to run) on the date of the initiation of legal process—here, on March 18, 2011, more than two years before Manuel filed suit. See Brief for Respondents 33. According to the City, the most analogous tort to Manuel’s constitutional claim is not malicious prosecution but false arrest, which accrues when legal process commences. See Tr. of Oral Arg. 47; Wallace, 549 U. S., at 389 (noting accrual rule for false arrest suits). And even if malicious prosecution were the better comparison, the City continues, a court should decline to adopt that tort’s favorable-termination element and associated accrual rule in adjudicating a §1983 claim involving pretrial detention. That element, the City argues, “make[s] little sense” in this context because “the Fourth Amendment is concerned not with the outcome of a prosecution, but with the legality of searches and seizures.” Brief for Respondents 16. And finally, the City contends that Manuel forfeited an alternative theory for treating his date of release as the date of accrual: to wit, that his pretrial detention “constitute[d] a continuing Fourth Amendment violation,” each day of which triggered the statute of limitations anew. Id., at 29, and n. 6; see Tr. of Oral Arg. 36; see also Albright, 510 U. S., at 280 (Ginsburg, J., concurring) (propounding a similar view). So Manuel, the City concludes, lost the opportunity to recover for his pretrial detention by waiting too long to file suit. We leave consideration of this dispute to the Court of Appeals. “[W]e are a court of review, not of first view.” Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005). Because the Seventh Circuit wrongly held that Manuel lacked any Fourth Amendment claim once legal process began, the court never addressed the elements of, or rules applicable to, such a claim. And in particular, the court never confronted the accrual issue that the parties contest here.[10] On remand, the Court of Appeals should decide that question, unless it finds that the City has previously waived its timeliness argument. See Reply to Brief in Opposition 1–2 (addressing the possibility of waiver); Tr. of Oral Arg. 40–44 (same). And so too, the court may consider any other still-live issues relating to the contours of Manuel’s Fourth Amendment claim for unlawful pretrial detention. * * * For the reasons stated, we reverse the judgment of the Seventh Circuit and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 Because we here review an order dismissing Manuel’s suit, we accept as true all the factual allegations in his complaint. See, e.g., Leatherman v. Tarrant County Narcotics Intelligence and Coordination Unit, 507 U. S. 163, 164 (1993) . 2 Although not addressed in Manuel’s complaint, the police department’s alleged fabrications did not stop at this initial hearing on probable cause. About two weeks later, on March 30, a grand jury indicted Manuel based on similar false evidence: testimony from one of the arresting officers that “[t]he pills field tested positive” for ecstasy. App. 96 (grand jury minutes). 3 Manuel’s allegation of unlawful detention concerns only the period after the onset of legal process—here meaning, again, after the County Court found probable cause that he had committed a crime. See supra, at 3. The police also held Manuel in custody for several hours between his warrantless arrest and his first appearance in court. But throughout this litigation, Manuel has treated that short period as part and parcel of the initial unlawful arrest. See, e.g., Reply Brief 1. 4 See also Singer v. Fulton County Sheriff, 63 F. 3d 110, 114–118 (CA2 1995); McKenna v. Philadelphia, 582 F. 3d 447, 461 (CA3 2009); Lambert v. Williams, 223 F. 3d 257, 260–262 (CA4 2000); Castellano v. Fragozo, 352 F. 3d 939, 953–954, 959–960 (CA5 2003) (en banc); Sykes v. Anderson, 625 F. 3d 294, 308–309 (CA6 2010); Galbraith v. County of Santa Clara, 307 F. 3d 1119, 1126–1127 (CA9 2002); Wilkins v. De-Reyes, 528 F. 3d 790, 797–799 (CA10 2008); Whiting v. Traylor, 85 F. 3d 581, 584–586 (CA11 1996); Pitt v. District of Columbia, 491 F. 3d 494, 510–511 (CADC 2007). 5 The Court repeated the same idea in a follow-on decision to Gerstein. In County of Riverside v. McLaughlin, 500 U. S. 44, 47 (1991) , we considered how quickly a jurisdiction must provide the probable-cause determination that Gerstein demanded “as a prerequisite to an extended pretrial detention.” In holding that the decision should occur within 48 hours of an arrest, the majority understood its “task [as] articulat[ing] more clearly the boundaries of what is permissible under the Fourth Amendment.” 500 U. S., at 56. In arguing for still greater speed, the principal dissent invoked the original meaning of “the Fourth Amendment’s prohibition of ‘unreasonable seizures,’ insofar as it applies to seizure of the person.” Id., at 60 (Scalia, J., dissenting). The difference between the two opinions was significant, but the commonality still more so: All Justices agreed that the Fourth Amendment provides the appropriate lens through which to view a claim involving pretrial detention. 6 The opposite view would suggest an untenable result: that a person arrested pursuant to a warrant could not bring a Fourth Amendment claim challenging the reasonableness of even his arrest, let alone any subsequent detention. An arrest warrant, after all, is a way of initiating legal process, in which a magistrate finds probable cause that a person committed a crime. See Wallace v. Kato, 549 U. S. 384, 389 (2007) (explaining that the seizure of a person was “without legal process” because police officers “did not have a warrant for his arrest”); W. Keeton, D. Dobbs, R. Keeton, & D. Owen, Prosser and Keeton on Law of Torts §119, pp. 871, 886 (5th ed. 1984) (similar). If legal process is the cut-off point for the Fourth Amendment, then someone arrested (as well as later held) under a warrant procured through false testimony would have to look to the Due Process Clause for relief. But that runs counter to our caselaw. See, e.g., Whiteley v. Warden, Wyo. State Penitentiary, 401 U. S. 560 –569 (1971) (holding that an arrest violated the Fourth Amendment because a magistrate’s warrant was not backed by probable cause). And if the Seventh Circuit would reply that arrest warrants are somehow different—that there is legal process and then again there is legal process—the next (and in our view unanswerable) question would be why. 7 Even the City no longer appears to contest that conclusion. On multiple occasions during oral argument in this Court, the City agreed that “a Fourth Amendment right . . . survive[d] the initiation of process” at the hearing in which the county judge found probable cause and ordered detention. Tr. of Oral Arg. 31; see id., at 33 (concurring with the statement that “once [an] individual is brought . . . before a magistrate, and the magistrate using the same bad evidence says, stay here in jail . . . until we get to trial, that that period is a violation of the Fourth Amendment”); id., at 51 (stating that a detainee has “a Fourth Amendment claim” if “misstatements at [such a probable-cause hearing] led to ongoing pretrial seizure”). 8 The dissent goes some way toward claiming that a different kind of pretrial legal process—a grand jury indictment or preliminary examination—does expunge such a Fourth Amendment claim. See post, at 9, n. 4 (opinion of Alito, J.) (raising but “not decid[ing] that question”); post, at 10 (suggesting an answer nonetheless). The effect of that view would be to cut off Manuel’s claim on the date of his grand jury indictment (March 30)—even though that indictment (like the County Court’s probable-cause proceeding) was entirely based on false testi-mony and even though Manuel remained in detention for 36 days longer.See n. 2, supra. Or said otherwise—even though the legal process he received failed to establish the probable cause necessary for his continued confinement. We can see no principled reason to draw that line. Nothing in the nature of the legal proceeding establishing probable cause makes a difference for purposes of the Fourth Amendment: Whatever its precise form, if the proceeding is tainted—as here, by fabricated evidence—and the result is that probable cause is lacking, then the ensuing pretrial detention violates the confined person’s Fourth Amendment rights, for all the reasons we have stated. By contrast (and contrary to the dissent’s suggestion, see post, at 9, n. 3), once a trial has occurred, the Fourth Amendment drops out: A person challenging the sufficiency of the evidence to support both a conviction and any ensuing incarceration does so under the Due Process Clause of the Fourteenth Amendment. See Jackson v. Virginia, 443 U. S. 307, 318 (1979) (invalidating a conviction under the Due Process Clause when “the record evidence could [not] reasonably support a finding of guilt beyond a reasonable doubt”); Thompson v. Louisville, 362 U. S. 199, 204 (1960) (striking a conviction under the same provision when “the record [wa]s entirely lacking in evidence” of guilt—such that it could not even establish probable cause). Gerstein and Albright, as already suggested, both reflected and recognized that constitutional division of labor. See supra, at 6–8. In their words, the Framers “drafted the Fourth Amendment” to address “the matter of pretrial deprivations of liberty,” Albright, 510 U. S., at 274 (emphasis added), and the Amendment thus provides “standards and procedures” for “the detention of suspects pending trial,” Gerstein, 420 U. S., at 125, n. 27 (emphasis added). 9 The two exceptions—the Ninth and D. C. Circuits—have not yet weighed in on whether a Fourth Amendment claim like Manuel’s includes a “favorable termination” element. 10 The dissent would have us address these questions anyway, on the ground that “the conflict on the malicious prosecution question was the centerpiece of Manuel’s argument in favor of certiorari.” Post, at 2. But the decision below did not implicate a “conflict on the malicious prosecution question”—because the Seventh Circuit, in holding that detainees like Manuel could not bring a Fourth Amendment claim at all, never considered whether (and, if so, how) that claim should resemble the malicious prosecution tort. Nor did Manuel’s petition for certiorari suggest otherwise. The principal part of his question presented—mirroring the one and only Circuit split involving the decision below—reads as follows: “[W]hether an individual’s Fourth Amendment right to be free from unreasonable seizure continues beyond legal process.” Pet. for Cert. i. That is exactly the issue we have resolved. The rest of Manuel’s question did indeed express a view as to what would follow from an affirmative answer (“so as to allow a malicious prosecution claim”). Ibid. (And as the dissent notes, the Seventh Circuit recounted that he made the same argument in that court. See post, at 2, n. 1.) But as to that secondary issue, we think (for all the reasons just stated) that Manuel jumped the gun. See supra, at 11–14. And contra the dissent, his doing so provides no warrant for our doing so too.

582.US.2016_16-309

Petitioner Divna Maslenjak is an ethnic Serb who resided in Bosnia during the 1990’s, when a civil war divided the new country. In 1998, she and her family sought refugee status in the United States. Interviewed under oath, Maslenjak explained that the family feared persecution from both sides of the national rift: Muslims would mistreat them because of their ethnicity, and Serbs would abuse them because Maslenjak’s husband had evaded service in the Bosnian Serb Army by absconding to Serbia. Persuaded of the Maslenjaks’ plight, American officials granted them refugee status. Years later, Maslenjak applied for U. S. citizenship. In the application process, she swore that she had never given false information to a government official while applying for an immigration benefit or lied to an official to gain entry into the United States. She was naturalized as a U. S. citizen. But it soon emerged that her professions of honesty were false: Maslenjak had known all along that her husband spent the war years not secreted in Serbia, but serving as an officer in the Bosnian Serb Army. The Government charged Maslenjak with knowingly “procur[ing], contrary to law, [her] naturalization,” in violation of 18 U. S. C. §1425(a). According to the Government’s theory, Maslenjak violated §1425(a) because, in the course of procuring her naturalization, she broke another law: 18 U. S. C. §1015(a), which prohibits knowingly making a false statement under oath in a naturalization proceeding. The District Court instructed the jury that, to secure a conviction under §1425(a), the Government need not prove that Maslenjak’s false statements were material to, or influenced, the decision to approve her citizenship application. The Sixth Circuit affirmed the conviction, holding that if Maslenjak made false statements violating §1015(a) and procured naturalization, then she also violated §1425(a). Held: 1. The text of §1425(a) makes clear that, to secure a conviction, the Government must establish that the defendant’s illegal act played a role in her acquisition of citizenship. To “procure . . . naturalization” means to obtain it. And the adverbial phrase “contrary to law” specifies how a person must procure naturalization so as to run afoul of the statute: illegally. Thus, someone “procure[s], contrary to law, naturalization” when she obtains citizenship illegally. As ordinary usage demonstrates, the most natural understanding of that phrase is that the illegal act must have somehow contributed to the obtaining of citizenship. To get citizenship unlawfully is to get it through an unlawful means—and that is just to say that an illegality played some role in its acquisition. The Government’s contrary view—that §1425(a) requires only a violation in the course of procuring naturalization—falters on the way language naturally works. Suppose that an applicant for citizenship fills out the paperwork in a government office with a knife tucked away in her handbag. She has violated the law against possessing a weapon in a federal building, and she has done so in the course of procuring citizenship, but nobody would say she has “procure[d]” her citizenship “contrary to law.” That is because the violation of law and the acquisition of citizenship in that example are merely coincidental: The one has no causal relation to the other. Although the Government attempts to define such examples out of the statute, that effort falls short for multiple reasons. Most important, the Government’s attempted carve-out does nothing to alter the linguistic understanding that gives force to the examples the Government would exclude. Under ordinary rules of language usage, §1425(a) demands a causal or means-end connection between a legal violation and naturalization. The broader statutory context reinforces the point, because the Government’s reading would create a profound mismatch between the requirements for naturalization and those for denaturalization: Some legal violations that do not justify denying citizenship would nonetheless justify revoking it later. For example, lies told out of “embarrassment, fear, or a desire for privacy” (rather than “for the purpose of obtaining [immigration] benefits”) are not generally disqualifying under the statutory requirement of “good moral character.” Kungys v. United States, 485 U. S. 759 ; 8 U. S. C. §1101(f)(6). But under the Government’s reading of §1425(a), any lie told in the naturalization process would provide a basis for rescinding citizenship. The Government could thus take away on one day what it was required to give the day before. And by so unmooring the revocation of citizenship from its award, the Government opens the door to a world of disquieting consequences—which this Court would need far stronger textual support to believe Congress intended. The statute Congress passed, most naturally read, strips a person of citizenship not when she committed any illegal act during the naturalization process, but only when that act played some role in her naturalization. Pp. 4–9. 2. When the underlying illegality alleged in a §1425(a) prosecution is a false statement to government officials, a jury must decide whether the false statement so altered the naturalization process as to have influenced an award of citizenship. Because the entire naturalization process is set up to provide little room for subjective preferences or personal whims, that inquiry is properly framed in objective terms: To decide whether a defendant acquired citizenship by means of a lie, a jury must evaluate how knowledge of the real facts would have affected a reasonable government official properly applying naturalization law. If the facts the defendant misrepresented are themselves legally disqualifying for citizenship, the jury can make quick work of that inquiry. In such a case, the defendant’s lie must have played a role in her naturalization. But that is not the only time a jury can find that a defendant’s lies had the requisite bearing on a naturalization decision, because lies can also throw investigators off a trail leading to disqualifying facts. When relying on such an investigation-based theory, the Government must make a two-part showing. Initially, the Government must prove that the misrepresented fact was sufficiently relevant to a naturalization criterion that it would have prompted reasonable officials, “seeking only evidence concerning citizenship qualifications,” to undertake further investigation. Kungys, 485 U. S., at 774, n. 9. If that much is true, the inquiry turns to the prospect that such an investigation would have borne disqualifying fruit. The Government need not show definitively that its investigation would have unearthed a disqualifying fact. It need only establish that the investigation “would predictably have disclosed” some legal disqualification. Id., at 774. If that is so, the defendant’s misrepresentation contributed to the citizenship award in the way §1425(a) requires. This demanding but still practicable causal standard reflects the real-world attributes of cases premised on what an unhindered investigation would have found. When the Government can make its two-part showing, the defendant may overcome it by establishing that she was qualified for citizenship (even though she misrepresented facts that suggested the opposite). Thus, whatever the Government shows with respect to a thwarted investigation, qualification for citizenship is a complete defense to a prosecution under §1425(a). Pp. 10–15. 3. Measured against this analysis, the jury instructions in this case were in error. The jury needed to find more than an unlawful false statement. However, it was not asked to—and so did not—make any of the necessary determinations. The Government’s assertion that any instructional error was harmless is left for resolution on remand. Pp. 15–16. 821 F. 3d 675, vacated and remanded. Kagan, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Ginsburg, Breyer, and Sotomayor, JJ., joined. Gorsuch, J., filed an opinion concurring in part and concurring in the judgment, in which Thomas, J., joined. Alito, J., filed an opinion concurring in the judgment.

A federal statute, 18 U. S. C. §1425(a), makes it a crime to “knowingly procure[ ], contrary to law, the naturalization of any person.” And when someone is convicted under §1425(a) of unlawfully procuring her own naturalization, her citizenship is automatically revoked. See 8 U. S. C. §1451(e). In this case, we consider what the Government must prove to obtain such a conviction. We hold that the Government must establish that an illegal act by the defendant played some role in her acquisition of citizenship. When the illegal act is a false statement, that means demonstrating that the defendant lied about facts that would have mattered to an immigration official, because they would have justified denying naturalization or would predictably have led to other facts warranting that result. I Petitioner Divna Maslenjak is an ethnic Serb who re-sided in Bosnia during the 1990’s, when a civil war between Serbs and Muslims divided the new country. In 1998, she and her family (her husband Ratko Maslenjak and their two children) met with an American immigration official to seek refugee status in the United States. Interviewed under oath, Maslenjak explained that the family feared persecution in Bosnia from both sides of the national rift. Muslims, she said, would mistreat them because of their ethnicity. And Serbs, she testified, would abuse them because her husband had evaded service in the Bosnian Serb Army by absconding to Serbia—where he remained hidden, apart from the family, for some five years. See App. to Pet. for Cert. 58a–60a. Persuaded of the Maslenjaks’ plight, American officials granted them refugee status, and they immigrated to the United States in 2000. Six years later, Maslenjak applied for naturalization. Question 23 on the application form asked whether she had ever given “false or misleading information” to a government official while applying for an immigration benefit; question 24 similarly asked whether she had ever “lied to a[ ] government official to gain entry or admission into the United States.” Id., at 72a. Maslenjak answered “no” to both questions, while swearing under oath that her replies were true. Id., at 72a, 74a. She also swore that all her written answers were true during a subsequent interview with an immigration official. In August 2007, Maslenjak was naturalized as a U. S. citizen. But Maslenjak’s professions of honesty were false: In fact, she had made up much of the story she told to immigration officials when seeking refuge in this country. Her fiction began to unravel at around the same time she applied for citizenship. In 2006, immigration officials confronted Maslenjak’s husband Ratko with records showing that he had not fled conscription during the Bosnian civil war; rather, he had served as an officer in the Bos-nian Serb Army. And not only that: He had served in a brigade that participated in the Srebrenica massacre—a slaughter of some 8,000 Bosnian Muslim civilians. Within a year, the Government convicted Ratko on charges of making false statements on immigration documents. The newly naturalized Maslenjak attempted to prevent Ratko’s deportation. During proceedings on that matter, Maslenjak admitted she had known all along that Ratko spent the war years not secreted in Serbia but fighting in Bosnia. As a result, the Government charged Maslenjak with knowingly “procur[ing], contrary to law, [her] naturalization,” in violation of 18 U. S. C. §1425(a). According to the Government’s theory, Maslenjak violated §1425(a) because, in the course of procuring her naturalization, she broke another law: 18 U. S. C. §1015(a), which prohibits knowingly making a false statement under oath in a naturalization proceeding. The false statements the Government invoked were Maslenjak’s answers to questions 23 and 24 on the citizenship application (stating that she had not lied in seeking refugee status) and her corresponding statements in the citizenship interview. Those statements, the Government argued to the District Court, need not have affected the naturalization decision to support a conviction under §1425(a). The court agreed: Over Maslenjak’s objection, it instructed the jury that a conviction was proper so long as the Government “prove[d] that one of the defendant’s statements was false”—even if the statement was not “material” and “did not influence the decision to approve [her] naturalization.” App. to Pet. for Cert. 86a. The jury returned a guilty verdict; and the District Court, based on that finding, stripped Maslenjak of her citizenship. See 8 U. S. C. §1451(e). The United States Court of Appeals for the Sixth Circuit affirmed the conviction. As relevant here, the Sixth Circuit upheld the District Court’s instructions that Maslenjak’s false statements need not have influenced the naturalization decision. If, the Court of Appeals held, Maslenjak made false statements violating §1015(a) and she procured naturalization, then she also violated §1425(a)—irrespective of whether the false statements played any role in her obtaining citizenship. See 821 F. 3d 675, 685–686 (2016). That decision created a conflict in the Circuit Courts.[1] We granted certiorari to resolve it, 580 U. S. ___ (2017), and we now vacate the Sixth Circuit’s judgment. II A Section 1425(a), the parties agree, makes it a crime to commit some other illegal act in connection with naturalization. But the parties dispute the nature of the required connection. Maslenjak argues that the relationship must be “causal” in kind: A person “procures” her naturalization “contrary to law,” she contends, only if a predicate crime in some way “contribut[ed]” to her gaining citizenship. Brief for Petitioner 21. By contrast, the Government proposes a basically chronological link: Section 1425(a), it urges, “punishes the commission of other violations of law in the course of procuring naturalization”—even if the illegality could not have had any effect on the naturalization decision. Brief for United States 14 (emphasis added). We conclude that Maslenjak has the better of this argument. We begin, as usual, with the statutory text. In ordinary usage, “to procure” something is “to get possession of” it. Webster’s Third New International Dictionary 1809 (2002); accord, Black’s Law Dictionary 1401 (10th ed. 2014) (defining “procure” as “[t]o obtain (something), esp. by special effort or means”). So to “procure . . . naturalization” means to obtain naturalization (or, to use another word, citizenship). The adverbial phrase “contrary to law,” wedged in between “procure” and “naturalization,” then specifies how a person must procure naturalization so as to run afoul of the statute: in contravention of the law—or, in a word, illegally. Putting the pieces together, someone “procure[s], contrary to law, naturalization” when she obtains citizenship illegally. What, then, does that whole phrase mean? The most natural understanding is that the illegal act must have somehow contributed to the obtaining of citizenship. Consider if someone said to you: “John obtained that painting illegally.” You might imagine that he stole it off the walls of a museum. Or that he paid for it with a forged check. Or that he impersonated the true buyer when the auction house delivered it. But in all events, you would imagine illegal acts in some kind of means-end relation—or otherwise said, in some kind of causal relation—to the painting’s acquisition. If someone said to you, “John obtained that painting illegally, but his unlawful acts did not play any role in his obtaining it,” you would not have a clue what the statement meant. You would think it nonsense—or perhaps the opening of a riddle. That is because if no illegal act contributed at all to getting the painting, then the painting would not have been gotten illegally. And the same goes for naturalization. If whatever illegal conduct occurring within the naturalization process was a causal dead-end—if, so to speak, the ripples from that act could not have reached the decision to award citizenship—then the act cannot support a charge that the applicant obtained naturalization illegally. The conduct, though itself illegal, would not also make the obtaining of citizenship so. To get citizenship unlawfully, we understand, is to get it through an unlawful means—and that is just to say that an illegality played some role in its acquisition.[2] The Government’s contrary view—that §1425(a) requires only a “violation[ ] of law in the course of procuring naturalization”—falters on the way language naturally works. Brief for United States 14. Return for a moment to our artwork example. Imagine this time that John made an illegal turn while driving to the auction house to purchase a painting. Would you say that he had “procured the painting illegally” because he happened to violate the law in the course of obtaining it? Not likely. And again, the same is true with respect to naturalization. Suppose that an applicant for citizenship fills out the necessary paperwork in a government office with a knife tucked away in her handbag (but never mentioned or used). She has violated the law—specifically, a statute criminalizing the possession of a weapon in a federal building. See 18 U. S. C. §930. And she has surely done so “in the course of” procuring citizenship. But would you say, using English as you ordinarily would, that she has “procure[d]” her citizenship “contrary to law” (or, as you would really speak, “illegally”)? Once again, no. That is because the violation of law and the acquisition of citizenship are in that example merely coincidental: The one has no causal relation to the other. The Government responds to such examples by seeking to define them out of the statute, but that effort falls short for multiple reasons. According to the Government, the laws to which §1425(a) speaks are only laws “pertaining to naturalization.” Brief for United States 20. But to begin with, that claim fails on its own terms. The Government’s proposed limitation has no basis in §1425(a)’s text (which refers to “law” generally); it is a deus ex machina—rationalized only by calling it “necessary,” Tr. of Oral Arg. 39, and serving only to get the Government out of a tight interpretive spot. Indeed, the Government does not really buy its own argument: At another point, it asserts that an applicant for citizenship can violate §1425(a) by bribing a government official, see Brief for United States 16—even though the law against that conduct has nothing in particular to do with naturalization. See 18 U. S. C. §201(b)(1). And still more important, the Government’s (sometime) carve-out does nothing to alter the linguistic understanding that gives force to the examples the Government would exclude—and that applies just as well to every application that would remain. Laws pertaining to naturalization, in other words, are subject to the same rules of language usage as laws concerning other subjects. And under those rules, as we have shown, §1425(a) demands a means-end connection between a legal violation and naturalization. See supra, at 5–6. Take §1015(a)’s bar on making false statements in connection with naturalization—the prototypical §1425(a) predicate, and the one at issue here. If such a statement (in an interview, say) has no bearing at all on the decision to award citizenship, then it cannot render that award—as §1425(a) requires—illegally gained. The broader statutory context reinforces that point, because the Government’s reading would create a profound mismatch between the requirements for naturalization on the one hand and those for denaturalization on the other. See West Virginia Univ. Hospitals, Inc. v. Casey, 499 U. S. 83, 101 (1991) (“[I]t is our role to make sense rather than nonsense out of the corpus juris”). The immigration statute requires all applicants for citizenship to have “good moral character,” and largely defines that term through a list of unlawful or unethical behaviors. 8 U. S. C. §§1427(a)(3), 1101(f ).[3] On the Government’s theory, some legal violations that do not justify denying citizenship under that definition would nonetheless justify revoking it later. Again, false statements under §1015(a) offer an apt illustration. The statute’s description of “good moral character” singles out a specific class of lies—“false testimony for the purpose of obtaining [immigration] benefits”—as a reason to deny naturalization. 8 U. S. C. §1101(f )(6). By contrast, “[w]illful misrepresentations made for other reasons, such as embarrassment, fear, or a desire for privacy, were not deemed sufficiently culpable to brand the applicant as someone who lacks good moral character”—and so are not generally disqualifying. Kungys v. United States, 485 U. S. 759, 780 (1988) (quoting Supplemental Brief for United States 12). But under the Government’s reading of §1425(a), a lie told in the naturalization process—even out of embarrassment, fear, or a desire for privacy—would always provide a basis for rescinding citizenship. The Government could thus take away on one day what it was required to give the day before. And by so wholly unmooring the revocation of citizenship from its award, the Government opens the door to a world of disquieting consequences—which we would need far stronger textual support to believe Congress intended. Consider the kinds of questions a person seeking citizenship confronts on the standard application form. Says one: “Have you EVER been . . . in any way associated with[ ] any organization, association, fund, foundation, party, club, society, or similar group[?]” Form N–400, Application for Naturalization 12 (2016), online at http://www.uscis.gov/n-400 (as last visited June 20, 2017) (bold in original). Asks another: “Have you EVER committed . . . a crime or offense for which you were NOT arrested?” Id., at 14. Suppose, for reasons of embarrassment or what-have-you, a person concealed her membership in an online support group or failed to disclose a prior speeding violation. Under the Government’s view, a prosecutor could scour her paperwork and bring a §1425(a) charge on that meager basis, even many years after she became a citizen. That would give prosecutors nearly limitless leverage—and afford newly naturalized Americans precious little security. Small wonder that Congress, in enacting §1425(a), did not go so far as the Government claims. The statute it passed, most naturally read, strips a person of citizenship not when she committed any illegal act during the naturalization process, but only when that act played some role in her naturalization. B That conclusion leaves us with a more operational question: How should §1425(a)’s requirement of causal influence apply in practice, when charges are brought under that law?[4] Because the proper analysis may vary with the nature of the predicate crime, we confine our discussion of that issue to the kind of underlying illegality alleged here: a false statement made to government officials. Such conduct can affect a naturalization decision in a single, significant way—by distorting the Government’s understanding of the facts when it investigates, and then adjudicates, an application. So the issue a jury must decide in a case like this one is whether a false statement sufficiently altered those processes as to have influenced an award of citizenship. The answer to that question, like the naturalization decision itself, turns on objective legal criteria. Congress has prescribed specific eligibility standards for new citizens, respecting such matters as length of residency and “physical[ ] presen[ce],” understanding of English and American government, and (as previously mentioned) “good moral character,” with all its many specific components. See 8 U. S. C. §§1423(a), 1427(a); supra, at 8. Government officials are obligated to apply that body of law faithfully—granting naturalization when the appli-cable criteria are satisfied, and denying it when they are not. See Kungys, 485 U. S., at 774, n. 9 (opinion of Scalia, J.); id., at 787 (Stevens, J., concurring in judgment). And to ensure right results are reached, a court can reverse such a determination, at an applicant’s request, based on its “own findings of fact and conclusions of law.” 8 U. S. C. §1421(c). The entire system, in other words, is set up to provide little or no room for subjective preferences or personal whims. Because that is so, the question of what any individual decisionmaker might have done with accurate information is beside the point: The defendant in a §1425(a) case should neither benefit nor suffer from a wayward official’s deviations from legal requirements. Accordingly, the proper causal inquiry under §1425(a) is framed in objective terms: To decide whether a defendant acquired citizenship by means of a lie, a jury must evaluate how knowledge of the real facts would have affected a reasonable government official properly applying naturalization law. If the facts the defendant misrepresented are themselves disqualifying, the jury can make quick work of that inquiry. In such a case, there is an obvious causal link between the defendant’s lie and her procurement of citizenship. To take an example: An applicant for citizenship must be physically present in the United States for more than half of the five-year period preceding her application. See 8 U. S. C. §1427(a)(1). Suppose a defendant misrepresented her travel history to convey she had met that requirement, when in fact she had not. The Government need only expose that lie to establish that she obtained naturalization illegally—for had she told the truth instead, the official would have promptly denied her application. Or consider another, perhaps more common case stemming from the “good moral character” criterion. See §1427(a)(3); supra, at 8. That phrase is defined to exclude any person who has been convicted of an aggravated fel-ony. See §1101(f )(8). If a defendant falsely denied such a conviction, she too would have gotten her citizenship by means of a lie—for otherwise the outcome would have been different. In short, when the defendant misrepresents facts that the law deems incompatible with citizenship, her lie must have played a role in her naturalization. But that is not the only time a jury can find that a defendant’s lie had the requisite bearing on a naturalization decision. For even if the true facts lying behind a false statement would not “in and of themselves justify denial of citizenship,” they could have “led to the discovery of other facts which would” do so. Chaunt v. United States, 364 U. S. 350 –353 (1960). We previously addressed that possibility when considering the civil statute that authorizes the Government to revoke naturalization. See Kungys, 485 U. S., at 774–777 (opinion of Scalia, J.) (interpreting 8 U. S. C. §1451(a)).[5] As we explained in that context, a person whose lies throw investigators off a trail leading to disqualifying facts gets her citizenship by means of those lies—no less than if she had denied the damning facts at the very end of the trail. See ibid. When relying on such an investigation-based theory, the Government must make a two-part showing to meet its burden. As an initial matter, the Government has to prove that the misrepresented fact was sufficiently relevant to one or another naturalization criterion that it would have prompted reasonable officials, “seeking only evidence concerning citizenship qualifications,” to undertake further investigation. Id., at 774, n. 9. If that much is true, the inquiry turns to the prospect that such an investigation would have borne disqualifying fruit. As to that second link in the causal chain, the Government need not show definitively that its investigation would have unearthed a disqualifying fact (though, of course, it may). Rather, the Government need only establish that the investigation “would predictably have disclosed” some legal disqualification. Id., at 774; see id., at 783 (Brennan, J., concurring). If that is so, the defendant’s misrepresentation contributed to the citizenship award in the way we think §1425(a) requires. That standard reflects two real-world attributes of cases premised on what an unhindered investigation would have found. First is the difficulty of proving that a hypothetical inquiry would have led to some disqualifying discovery, often several years after the defendant told her lies. As witnesses and other evidence disappear, the Government’s effort to reconstruct the course of a “could have been” investigation confronts ever-mounting obstacles. See id., at 779 (opinion of Scalia, J.). Second, and critical to our analysis, is that the defendant—not the Government—bears the blame for that evidentiary predicament. After all, the inquiry cannot get this far unless the defendant made an unlawful false statement and, by so doing, obstructed the normal course of an investigation. See id., at 783 (Brennan, J., concurring) (emphasizing that “the citizen’s misrepresentation [in a naturalization proceeding] necessarily frustrated the Government’s investigative efforts”); see also Bigelow v. RKO Radio Pictures, Inc., 327 U. S. 251, 265 (1946) (“The most elementary conceptions of justice and public policy require that the wrongdoer shall bear the risk of the uncertainty which his own wrong has created”). Section 1425(a) is best read to take those exigencies and equities into account, by enabling the Government (as just described) to rest on disqualifications that a thwarted investigation predictably would have uncovered. A yet-stricter causal requirement, demanding proof positive that a disqualifying fact would have been found, sets the bar so high that “we cannot conceive that Congress intended” that result. Kungys, 485 U. S., at 777 (opinion of Scalia, J.). And nothing in the statutory text requires that approach. While §1425(a) clearly imports some kind of causal or means-end relation, see supra, at 5–9, Congress left that relation’s precise character unspecified. Cf. Burrage v. United States, 571 U. S. ___, ___ (2014) (slip op., at 10) (noting that courts have not always construed criminal statutes to “require[ ] strict but-for causality,” and have greater reason to reject such a reading when the laws do not use language like “results from” or “because of”). The open-endedness of the statutory language allows, indeed supports, our adoption of a demanding but still practicable causal standard. Even when the Government can make its two-part showing, however, the defendant may be able to overcome it. Section 1425(a) is not a tool for denaturalizing people who, the available evidence indicates, were actually qualified for the citizenship they obtained. When addressing the civil denaturalization statute, this Court insisted on a similar point: We provided the defendant with an opportunity to rebut the Government’s case “by showing, through a preponderance of the evidence, that the statutory requirement as to which [a lie] had a natural tendency to produce a favorable decision was in fact met.” Kungys, 485 U. S., at 777 (opinion of Scalia, J.) (emphasis deleted); accord, id., at 783–784 (Brennan, J., concurring). Or said otherwise, we gave the defendant a chance to establish that she was qualified for citizenship, and held that she could not be denaturalized if she did so—even though she concealed or misrepresented facts that suggested the opposite. And indeed, all our denaturalization decisions share this crucial feature: We have never read a statute to strip citizenship from someone who met the legal criteria for acquiring it. See, e.g., Fedorenko v. United States, 449 U. S. 490 –507 (1981); Costello v. United States, 365 U. S. 265 –272 (1961); Schneiderman v. United States, 320 U. S. 118 –123 (1943). We will not start now. Whatever the Government shows with respect to a thwarted investigation, qualification for citizenship is a complete defense to a prosecution brought under §1425(a). III Measured against all we have said, the jury instructions in this case were in error. As earlier noted, the District Court told the jury that it could convict based on any false statement in the naturalization process (i.e., any violation of §1015(a)), no matter how inconsequential to the ultimate decision. See App. to Pet. for Cert. 86a; supra, at 3. But as we have shown, the jury needed to find more than an unlawful false statement. Recall that Maslenjak’s lie in the naturalization process concerned her prior statements to immigration officials: She swore that she had been honest when applying for admission as a refugee, but in fact she had not. See supra, at 2–3. The jury could have convicted if that earlier dishonesty (i.e., the thing she misrepresented when seeking citizenship) were itself a reason to deny naturalization—say, because it counted as “false testimony for the purpose of obtaining [immigration] benefits” and thus demonstrated bad moral character. See supra, at 11–12. Or else, the jury could have convicted if (1) knowledge of that prior dishonesty would have led a reasonable official to make some further investigation (say, into the circumstances of her admission), (2) that inquiry would predictably have yielded a legal basis for rejecting her citizenship application, and (3) Maslenjak failed to show that (notwithstanding such an objective likelihood) she was in fact qualified to become a U. S. citizen. See supra, at 12–15. This jury, however, was not asked to—and so did not—make any of those determinations. Accordingly, Maslenjak was not convicted by a properly instructed jury of “procur[ing], contrary to law, [her] naturalization.” The Government asserts that any instructional error in this case was harmless. “Had officials known the truth,” the Government asserts, “it would have affected their decision to grant [Maslenjak] citizenship.” Brief for United States 12. Unsurprisingly, Maslenjak disagrees. See Tr. of Oral Arg. 6–8; Reply to Brief in Opposition 9–10. In keeping with our usual practice, we leave that dispute for resolution on remand. See, e.g., Skilling v. United States, 561 U. S. 358, 414 (2010) . For the reasons stated, we vacate the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered.Notes 1 Compare 821 F. 3d 675, 685–686 (CA6 2016) (case below), with United States v. Munyenyezi, 781 F. 3d 532, 536 (CA1 2015) (requiring the Government to make some showing that a misrepresentation mattered to the naturalization decision); United States v. Latchin, 554 F. 3d 709, 712–715 (CA7 2009) (same); United States v. Alferahin, 433 F. 3d 1148, 1154–1156 (CA9 2006) (same); United States v. Aladekoba, 61 Fed. Appx. 27, 28 (CA4 2003) (same). 2 To be fair, the idea of “obtaining citizenship illegally” has one other possible meaning, but no one defends it here because it does not fit with the rest of §1425. On this alternative reading, a person would violate §1425(a) by obtaining citizenship without the requisite legal qualifications—regardless of whether she committed another illegal act in the naturalization process. To vary our earlier example, suppose someone told you that John procured a gun illegally. You might think that meant John got the gun through independently unlawful conduct (e.g., he held up a gun store), as in the case of the painting. But you might instead think that John was just not legally qualified to take possession of a gun—because, for example, he once committed a felony. That alternative interpretation is plausible with respect to goods that not everyone is eligible to obtain, like guns—or like naturalization. And indeed, we have interpreted a civil statute closely resembling §1425(a)—which authorizes denaturalization when, inter alia, citizenship is “illegally procured,” 8 U. S. C. §1451(a)—to cover that qualifications-based species of illegality. See Fedorenko v. United States, 449 U. S. 490, 506 (1981). But neither party urges that reading here, and for good reason. Unlike its civil analogue, §1425(a) has a companion provision—§1425(b)—that makes it a crime to “procure or obtain naturalization” for “[one]self or another person not entitled thereto.” If obtaining citizenship without legal entitlement were enough to violate §1425(a), then that highly specific language in §1425(b) would be superfluous. Rather than reading those words to do no work, in violation of ordinary canons of statutory construction, we understand Congress to have defined two separate crimes in §1425: Assuming the appropriate mens rea, subsection (a) covers illegal means of procurement, as described above, while subsection (b) covers simple lack of qualifications. As we will explain, however, questions relating to citizenship qualifications play a significant role when applying §1425(a)’s causal standard in cases (like this one) predicated on false statements. See infra, at 10–11. 3 The list of disqualifying conduct is wide-ranging. See, e.g., 8 U. S. C. §1101(f)(4) (illegal gambling); §1101(f)(8) (aggravated felony conviction); §1101(f)(9) (participation in genocide). 4 Justice Gorsuch would stop before answering that question, see post, at 2 (opinion concurring in part and concurring in judgment), but we think that such a halfway-decision would fail to fulfill our responsibility to both parties and courts. The Government needs to know what prosecutions to bring; defendants need to know what defenses to offer; and district courts need to know how to instruct juries. Telling them only “§1425(a) has something to do with causation” would not much help them make those decisions. And we are well-positioned to provide further guidance. The parties have had every opportunity to address the nature of the statute’s causal standard, and both gave us considered views about how the law should work in practice. See, e.g., Brief for Petitioner 23–24, 30; Brief for United States 17–18, 48; Tr. of Oral Arg. 14–16, 23–25, 39–46. Moreover, many lower courts have already addressed those same issues—including one that has called this Court’s failure to provide clear guidance “maddening[ ].” Latchin, 554 F. 3d, at 713; see, e.g., id., at 713–714; Munyenyezi, 781 F. 3d, at 536–538; Alferahin, 433 F. 3d, at 1155; Aladekoba, 61 Fed. Appx., at 27–28; United States v. Acheampong, 2015 WL 926113, *2–*3 (D Kan., Mar. 3, 2015); United States v. Odeh, 2014 WL 5473042, *7–*8 (ED Mich., Oct. 27, 2014). 5 Kungys concerned the part of that statute providing for the revocation of citizenship “procured by concealment of a material fact or by willful misrepresentation.” §1451(a). As noted earlier, the same statute includes a prong covering citizenship that is “illegally procured.” See n. 2, supra.

581.US.2016_15-1248

Damiana Ochoa worked for eight years in a physically demanding job for petitioner McLane Co., a supply-chain services company. McLane requires employees in those positions—both new employees and those returning from medical leave—to take a physical evaluation. When Ochoa returned from three months of maternity leave, she failed the evaluation three times and was fired. She then filed a sex discrimination charge under Title VII of the Civil Rights Act of 1964. The Equal Employment Opportunity (EEOC) began an investigation, but McLane declined its request for so-called “pedigree information”: names, Social Security numbers, addresses, and telephone numbers of employees asked to take the evaluation. After the EEOC expanded the investigation’s scope both geographically (to cover McLane’s national operations) and substantively (to investigate possible age discrimination), it issued subpoenas, as authorized by 42 U. S. C. §2000e–9, requesting pedigree information relating to its new investigation. When McLane refused to provide the information, the EEOC filed two actions in Federal District Court—one arising out of Ochoa’s charge and one arising out of the EEOC’s own age-discrimination charge—seeking enforcement of its subpoenas. The District Judge declined to enforce the subpoenas, finding that the pedigree information was not relevant to the charges, but the Ninth Circuit reversed. Reviewing the District Court’s decision to quash the subpoena de novo, the court concluded that the lower court erred in finding the pedigree information irrelevant. Held: A district court’s decision whether to enforce or quash an EEOC subpoena should be reviewed for abuse of discretion, not de novo. Pp. 6–12. (a) Both factors that this Court examines when considering whether such decision should be subject to searching or deferential appellate review point toward abuse-of-discretion review. First, the longstanding practice of the courts of appeals is to review a district court’s decision to enforce or quash an administrative subpoena for abuse of discretion. Title VII confers on the EEOC the same authority to issue subpoenas that the National Labor Relations Act (NLRA) confers on the National Labor Relations Board (NLRB). During the three decades between the NLRA’s enactment and the incorporation of its subpoena-enforcement provisions into Title VII, every Circuit to consider the question had held that a district court’s decision on enforcement of an NLRB subpoena is subject to abuse-of-discretion review. Congress amended Title VII to authorize EEOC subpoenas against this uniform backdrop of deferential appellate review, and today, nearly every Court of Appeals reviews a district court’s decision whether to enforce an EEOC subpoena for abuse of discretion. This “long history of appellate practice,” Pierce v. Underwood, 487 U. S. 552 , carries significant persuasive weight. Second, basic principles of institutional capacity counsel in favor of deferential review. In most cases, the district court’s enforcement decision will turn either on whether the evidence sought is relevant to the specific charge or whether the subpoena is unduly burdensome in light of the circumstances. Both of these tasks are well suited to a district judge’s expertise. The first requires the district court to evaluate the relationship between the particular materials sought and the particular matter under investigation—an analysis “variable in relation to the nature, purposes and scope of the inquiry.” Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186 . And whether a subpoena is overly burdensome turns on the nature of the materials sought and the difficulty the employer will face in producing them—“ ‘fact-intensive, close calls’ ” better suited to resolution by the district court than the court of appeals. Cooter & Gell v. Hartmarx Corp., 496 U. S. 384 . Other functional considerations also show the appropriateness of abuse-of-discretion review. For one, the district courts’ considerable experience in making similar decisions in other contexts, see Buford v. United States, 532 U. S. 59 , gives them the “institutional advantag[e],” id., at 64, that comes with greater experience. Deferential review also “streamline[s] the litigation process by freeing appellate courts from the duty of reweighing evidence and reconsidering facts already weighed and considered by the district court,” Cooter & Gell, 496 U. S., at 404, something particularly important in a proceeding designed only to facilitate the EEOC’s investigation. Pp. 6–9. (b) Court-appointed amicus’ arguments in support of de novo review are not persuasive. Amicus claims that the district court’s primary task is to test a subpoena’s legal sufficiency and thus requires no exercise of discretion. But that characterization is not inconsistent with abuse-of-discretion review, which may be employed to insulate the trial judge’s decision from appellate review for the same kind of functional concerns that underpin the Court’s conclusion that abuse of discretion is the appropriate standard. It is also unlikely that affording deferential review to a district court’s subpoena decision would clash with Court of Appeals decisions that instructed district courts to defer to the EEOC’s determination about the relevance of evidence to the charge at issue. Such decisions are better read as resting on the established rule that the term “relevant” be understood “generously” to permit the EEOC “access to virtually any material that might cast light on the allegations against the employer.” EEOC v. Shell Oil Co., 466 U. S. 54 –69. Nor do the constitutional underpinnings of the Shell Oil standard require a different result. While this Court has described a subpoena as a “ ‘constructive’ search,” Oklahoma Press, 327 U. S., at 202, and implied that the Fourth Amendment is the source of the requirement that a subpoena not be “too indefinite,” United States v. Morton Salt Co., 338 U. S. 632 , not every decision touching on the Fourth Amendment is subject to searching review. See, e.g., United States v. Nixon, 418 U. S. 683 . Cf. Illinois v. Gates, 462 U. S. 213 ; Ornelas v. United States, 517 U. S. 690 , distinguished. Pp. 9–11. (c) The case is remanded so that the Court of Appeals can review the District Court’s decision under the appropriate standard in the first instance. In doing so, the Court of Appeals may consider, as and to the extent it deems appropriate, any of McLane’s arguments regarding the burdens imposed by the subpoena. Pp. 11–12. 804 F. 3d 1051, vacated and remanded. Sotomayor, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Breyer, Alito, and Kagan, JJ., joined. Ginsburg, J., filed an opinion concurring in part and dissenting in part.

Title VII of the Civil Rights Act of 1964 permits the Equal Employment Opportunity Commission (EEOC) to issue a subpoena to obtain evidence from an employer that is relevant to a pending investigation. The statute autho-rizes a district court to issue an order enforcing such a subpoena. The question presented here is whether a court of appeals should review a district court’s decision to enforce or quash an EEOC subpoena de novo or for abuse of discretion. This decision should be reviewed for abuse of discretion. I A Title VII of the Civil Rights Act of 1964 prohibits employment discrimination on the basis of “race, color, religion, sex, or national origin.” §703(a), 78Stat. 255, 42 U. S. C. §2000e–2(a). The statute entrusts the enforcement of that prohibition to the EEOC. See §2000e–5(a); EEOC v. Shell Oil Co., 466 U. S. 54 –62 (1984). The EEOC’s responsibilities “are triggered by the filing of a specific sworn charge of discrimination,” University of Pa. v. EEOC, 493 U. S. 182, 190 (1990) , which can be filed either by the person alleging discrimination or by the EEOC itself, see §2000e–5(b). When it receives a charge, the EEOC must first notify the employer, ibid., and must then investigate “to determine whether there is reasonable cause to believe that the charge is true,” University of Pa., 493 U. S., at 190 (internal quotation marks omitted). This case is about one of the tools the EEOC has at its disposal in conducting its investigation: a subpoena. In order “[t]o enable the [EEOC] to make informed decisions at each stage of the enforcement process,” Title VII “confers a broad right of access to relevant evidence.” Id., at 191. It provides that the EEOC “shall . . . have access to, for the purposes of examination, . . . any evidence of any person being investigated or proceeded against that relates to unlawful employment practices covered by” Title VII and “is relevant to the charge under investigation.” 42 U. S. C. §2000e–8(a). And the statute enables the EEOC to obtain that evidence by “authoriz[ing] [it] to issue a subpoena and to seek an order enforcing [the subpoena].” University of Pa., 493 U. S., at 191; see §2000e–9.[1] Under that authority, the EEOC may issue “subp[o]enas requiring the attendance and testimony of witnesses or the production of any evidence.” 29 U. S. C. §161(1). An employer may petition the EEOC to revoke the subpoena, see ibid., but if the EEOC rejects the petition and the employer still “refuse[s] to obey [the] subp[o]ena,” the EEOC may ask a district court to issue an order enforcing it, see §161(2). A district court’s role in an EEOC subpoena enforcement proceeding, we have twice explained, is a straightforward one. See University of Pa., 493 U. S., at 191; Shell Oil, 466 U. S., at 72, n. 26. A district court is not to use an enforcement proceeding as an opportunity to test the strength of the underlying complaint. Ibid. Rather, a district court should “ ‘satisfy itself that the charge is valid and that the material requested is “relevant” to the charge.’ ” University of Pa., 493 U. S., at 191. It should do so cognizant of the “generou[s]” construction that courts have given the term “relevant.” Shell Oil, 466 U. S., at 68–69 (“virtually any material that might cast light on the allegations against the employer”). If the charge is proper and the material requested is relevant, the district court should enforce the subpoena unless the employer establishes that the subpoena is “too indefinite,” has been issued for an “illegitimate purpose,” or is unduly burdensome. Id., at 72, n. 26. See United States v. Morton Salt Co., 338 U. S. 632 –653 (1950) (“The gist of the protection is in the requirement . . . that the disclosure sought shall not be unreasonable” (internal quotation marks omitted)). B This case arises out of a Title VII suit filed by a woman named Damiana Ochoa. Ochoa worked for eight years as a “cigarette selector” for petitioner McLane Co., a supply-chain services company. According to McLane, the job is a demanding one: Cigarette selectors work in distribution centers, where they are required to lift, pack, and move large bins containing products. McLane requires employees taking physically demanding jobs—both new employees and employees returning from medical leave—to take a physical evaluation. According to McLane, the evaluation “tests . . . range of motion, resistance, and speed”and “is designed, administered, and validated by a third party.” Brief for Petitioner 6. In 2007, Ochoa took threemonths of maternity leave. When she attempted to return to work, McLane asked her to take the evaluation. Ochoa attempted to pass the evaluation three times, but failed. McLane fired her. Ochoa filed a charge of discrimination, alleging (among other things) that she had been fired on the basis of her gender. The EEOC began an investigation, and—at its request—McLane provided it with basic information about the evaluation, as well as a list of anonymous employees that McLane had asked to take the evaluation. McLane’s list included each employee’s gender, role at the company, and evaluation score, as well as the reason each employee had been asked to take the evaluation. But the company refused to provide what the parties call “pedigree information”: the names, Social Security numbers, last known addresses, and telephone numbers of the employees who had been asked to take the evaluation. Upon learning that McLane used the evaluation nationwide, the EEOC expanded the scope of its investigation, both geographi-cally (to focus on McLane’s nationwide operations) and sub-stantively (to investigate whether McLane had discriminated against its employees on the basis of age). It issued subpoenas requesting pedigree information as it related to its new investigation. But McLane refused to providethe pedigree information, and so the EEOC filed two actions in Federal District Court—one arising out of Ochoa’s charge and one arising out of a separate age-discrimination charge the EEOC itself had filed—seeking enforcement of its subpoenas. The enforcement actions were assigned to the same District Judge, who, after a hearing, declined to enforce the subpoenas to the extent that they sought the pedigree information. See EEOC v. McLane Co., 2012 WL 1132758, *5 (D Ariz., Apr. 4, 2012) (age discrimination charge); Civ. No. 12–2469 (D Ariz., Nov. 19, 2012), App. to Pet. for Cert. 28–30 (Title VII charge).[2] In the District Court’s view, the pedigree information was not “relevant” to the charges because “ ‘an individual’s name, or even an interview he or she could provide if contacted, simply could not shed light on whether the [evaluation] represents a tool of . . . discrimination.’ ” App. to Pet. for Cert. 29 (quoting 2012 WL 1132758, at *5; some internal quotation marks omitted). The Ninth Circuit reversed. See 804 F. 3d 1051 (2015). Consistent with Circuit precedent, the panel reviewed the District Court’s decision to quash the subpoena de novo, and concluded that the District Court had erred in finding the pedigree information irrelevant. Id., at 1057. But the panel questioned in a footnote why de novo review applied, observing that its sister Circuits “appear[ed] to review issues related to enforcement of administrative subpoenas for abuse of discretion.” Id., at 1056, n. 3; see infra, at 7 (reviewing Court of Appeals authority). This Court granted certiorari to resolve the disagreement between the Courts of Appeals over the appropriate standard of review for the decision whether to enforce an EEOC subpoena. 579 U. S. ___ (2016). Because the United States agrees with McLane that such a decision shouldbe reviewed for abuse of discretion, Stephen B. Kinnard was appointed as amicus curiae to defend the judgment below. 580 U. S. ___ (2016). He has ably discharged his duties. II A When considering whether a district court’s decision should be subject to searching or deferential appellate review—at least absent “explicit statutory command”—we traditionally look to two factors. Pierce v. Underwood, 487 U. S. 552, 558 (1988) . First, we ask whether the “history of appellate practice” yields an answer. Ibid. Second, at least where “neither a clear statutory prescription nor a historical tradition exists,” we ask whether, “ ‘as a matter of the sound administration of justice, one judicial actor is better positioned than another to decide the issue in question.’ ” Id., at 558, 559–560 (quoting Miller v. Fenton, 474 U. S. 104, 114 (1985) ). Both factors point toward abuse-of-discretion review here. First, the longstanding practice of the courts of appeals in reviewing a district court’s decision to enforce or quash an administrative subpoena is to review that decision for abuse of discretion. That practice predates even Title VII itself. As noted, Title VII confers on the EEOC the same authority to issue subpoenas that the National Labor Relations Act (NLRA) confers on the National Labor Relations Board (NLRB). See n. 1, supra. During the three decades between the enactment of the NLRA and the incorporation of the NLRA’s subpoena-enforcement provisions into Title VII, every Circuit to consider the question had held that a district court’s decision whether to enforce an NLRB subpoena should be reviewed for abuse of discretion. See NLRB v. Consolidated Vacuum Corp., 395 F. 2d 416, 419–420 (CA2 1968); NLRB v. Friedman, 352 F. 2d 545, 547 (CA3 1965); NLRB v. Northern Trust Co., 148 F. 2d 24, 29 (CA7 1945); Goodyear Tire & Rubber Co. v. NLRB, 122 F. 2d 450, 453–454 (CA6 1941). By the time Congress amended Title VII to authorize EEOC subpoenas in 1972, it did so against this uniform backdrop of deferential appellate review. Today, nearly as uniformly, the Courts of Appeals apply the same deferential review to a district court’s decision as to whether to enforce an EEOC subpoena. Almost every Court of Appeals reviews such a decision for abuse of discretion. See, e.g., EEOC v. Kronos Inc., 620 F. 3d 287, 295–296 (CA3 2010); EEOC v. Randstad, 685 F. 3d 433, 442 (CA4 2012); EEOC v. Roadway Express, Inc., 261 F. 3d 634, 638 (CA6 2001); EEOC v. United Air Lines, Inc., 287 F. 3d 643, 649 (CA7 2002); EEOC v. Technocrest Systems, Inc., 448 F. 3d 1035, 1038 (CA8 2006); EEOC v. Dillon Companies, Inc., 310 F. 3d 1271, 1274 (CA10 2002); EEOC v. Royal Caribbean Cruises, Ltd., 771 F. 3d 757, 760 (CA11 2014) (per curiam). As Judge Watford—writing for the panel below—recognized, the Ninth Circuit alone applies a more searching form of review. See 804 F. 3d, at 1056, n. 3 (“Why we review questions of relevance and undue burden de novo is unclear”); see also EPA v. Alyeska Pipeline Serv. Co., 836 F. 2d 443, 445–446 (CA9 1988) (holding that de novo review applies). To be sure, the inquiry into the appropriate standard of review cannot be resolved by a head-counting exercise. But the “long his-tory of appellate practice” here, Pierce, 487 U. S., at 558, carries significant persuasive weight. Second, basic principles of institutional capacity counsel in favor of deferential review. The decision whether to enforce an EEOC subpoena is a case-specific one that turns not on “a neat set of legal rules,” Illinois v. Gates, 462 U. S. 213, 232 (1983) , but instead on the application of broad standards to “multifarious, fleeting, special, narrow facts that utterly resist generalization,” Pierce, 487 U. S., at 561–562 (internal quotation marks omitted). In the mine run of cases, the district court’s decision whether to enforce a subpoena will turn either on whether the evidence sought is relevant to the specific charge before it or whether the subpoena is unduly burdensome in light of the circumstances. Both tasks are well suited to a district judge’s expertise. The decision whether evidence sought is relevant requires the district court to evaluate the relationship between the particular materials sought and the particular matter under investigation—an analysis “variable in relation to the nature, purposes and scope of the inquiry.” Oklahoma Press Publishing Co. v. Walling, 327 U. S. 186, 209 (1946) . Similarly, the decision whether a subpoena is overly burdensome turns on the nature of the materials sought and the difficulty the employer will face in producing them. These inquiries are “generally not amenable to broad per se rules,” Sprint/United Management Co. v. Mendelsohn, 552 U. S. 379, 387 (2008) ; rather, they are the kind of “fact-intensive, close calls” better suited to resolution by the district court than the court of appeals, Cooter & Gell v. Hartmarx Corp., 496 U. S. 384, 404 (1990) (internal quotation marks omitted).[3] Other functional considerations also show that abuse-of-discretion review is appropriate here. For one, district courts have considerable experience in other contexts making decisions similar—though not identical—to those they must make in this one. See Buford v. United States, 532 U. S. 59, 66 (2001) (“[T]he comparatively greater expertise” of the district court may counsel in favor of deferential review). District courts decide, for instance, whether evidence is relevant at trial, Fed. Rule Evid. 401; whether pretrial criminal subpoenas are unreasonable in scope, Fed. Rule Crim. Proc. 16(c)(2); and more. These decisions are not the same as the decisions a district court must make in enforcing an administrative subpoena. But they are similar enough to give the district court the “institutional advantag[e],” Buford, 532 U. S., at 64, that comes with greater experience. For another, as we noted in Cooter & Gell, deferential review “streamline[s] the litigation process by freeing appellate courts from the duty of reweighing evidence and reconsidering facts already weighed and considered by the district court,” 496 U. S., at 404—a particularly important consideration in a “satellite” proceeding like this one, ibid., designed only to facilitate the EEOC’s investigation. B Amicus’ arguments to the contrary have aided our consideration of this case. But they do not persuade us that de novo review is appropriate. Amicus’ central argument is that the decision whether a subpoena should be enforced does not require the exercise of discretion on the part of the district court, and so it should not be reviewed for abuse of discretion. On amicus’ view, the district court’s primary role is to test the legal sufficiency of the subpoena, not to weigh whether it should be enforced as a substantive matter. Cf. Shell Oil, 466 U. S., at 72, n. 26 (rejecting the argument that the district court should assess the validity of the underlying claim in a proceeding to enforce a subpoena). Even accepting amicus’ view of the district court’s task, however, this understanding of abuse-of-discretion review is too narrow. As commentators have observed, abuse-of-discretion review is employed not only where a decisionmaker has “a wide range of choice as to what he decides, free from the constraints which characteristically attach whenever legal rules enter the decision[making] process”; it is also employed where the trial judge’s decision is given “an unu-sual amount of insulation from appellate revision” for func-tional reasons. Rosenberg, Judicial Discretion of the Trial Court, Viewed From Above, 22 Syracuse L. Rev. 635, 637 (1971); see also 22 C. Wright & K. Graham, Federal Practice and Procedure §5166.1 (2d ed. 2012). And as we have explained, it is in large part due to functional concerns that we conclude the district court’s decision should be reviewed for abuse of discretion. Even if the district court’s decision can be characterized in the way that amicus suggests, that characterization would not be inconsistent with abuse-of-discretion review. Nor are we persuaded by amicus’ remaining arguments. Amicus argues that affording deferential review to a district court’s decision would clash with Court of Appeals decisions instructing district courts to defer themselves to the EEOC’s determination that evidence is relevant to the charge at issue. See Director, Office of Thrift Supervision, v. Vinson & Elkins, LLP, 124 F. 3d 1304, 1307 (CADC 1997) (district courts should defer to agency appraisals of relevance unless they are “obviously wrong”); EEOC v. Lockheed Martin Corp., Aero & Naval Systems, 116 F. 3d 110, 113 (CA4 1997) (same). In amicus’ view, it is “analytically impossible” for the court of appeals to defer to the district court if the district court must itself defer to the agency. Tr. of Oral Arg. 29. We think the better reading of those cases is that they rest on the established rule that the term “relevant” be understood “generously” to permit the EEOC “access to virtually any material that might cast light on the allegations against the employer.” Shell Oil, 466 U. S., at 68–69. A district court deciding whether evidence is “relevant” under Title VII need not defer to the EEOC’s decision on that score; it must simply answer the question cognizant of the agency’s broad authority to seek and obtain evidence. Because the statute does not set up any scheme of double deference, amicus’ arguments as to the infirmities of such a scheme are misplaced. Nor do we agree that, as amicus suggests, the constitutional underpinnings of the Shell Oil standard require a different result. To be sure, we have described a subpoena as a “ ‘constructive’ search,” Oklahoma Press, 327 U. S., at 202, and implied that the Fourth Amendment is the source of the requirement that a subpoena not be “too indefinite,” Morton Salt, 338 U. S., at 652. But not every decision that touches on the Fourth Amendment is subject to searching review. Subpoenas in a wide variety of other contexts also implicate the privacy interests protected by the Fourth Amendment, but courts routinely review the enforcement of such subpoenas for abuse of discretion. See, e.g., United States v. Nixon, 418 U. S. 683, 702 (1974) (pretrial subpoenas duces tecum); In re Grand Jury Subpoena, 696 F. 3d 428, 432 (CA5 2012) (grand jury subpoenas); In re Grand Jury Proceedings, 616 F. 3d 1186, 1201 (CA10 2010) (same). And this Court has emphasized that courts should pay “great deference” to a magistrate judge’s determination of probable cause, Gates, 462 U. S., at 236 (internal quotation marks omitted)—a decision more akin to a district court’s preenforcement review of a subpoena than the warrantless searches and seizures we considered in Ornelas v. United States, 517 U. S. 690 (1996) , on which amicus places great weight. The constitutional pedigree of Shell Oil does not change our view of the correct standard of review. III For these reasons, a district court’s decision to enforce an EEOC subpoena should be reviewed for abuse of discretion, not de novo. The United States also argues that the judgment below can be affirmed because it is clear that the District Court abused its discretion. But “we are a court of review, not of first view,” Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005), and the Court of Appeals has not had the chance to review the District Court’s decision under the appropriate standard. That task is for the Court of Appeals in the first instance. As part of its analysis, the Court of Appeals may also consider, as and to the extent it deems appropriate, any arguments made by McLane regarding the burdens imposed by the subpoena. The judgment of the Court of Appeals is hereby vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The statute does so by conferring on the EEOC the same authority given to the National Labor Relations Board to conduct investigations. See 42 U. S. C. §2000e–9 (“For the purpose of all . . . investigations conducted by the Commission . . . section 161 of title 29 shall apply”). 2 The District Court also refused to enforce the subpoena to the extent that it sought a second category of evidence: information about when and why those employees who had been fired after taking the test had been fired. The District Court provided no explanation for not enforcing the subpoena to the extent it sought this information, and the Court of Appeals reversed on that ground. 804 F. 3d 1051, 1059 (CA9 2015). McLane does not challenge this aspect of the Court of Appeals’ decision. See Tr. of Oral Arg. 8. 3 To be sure, there are pure questions of law embedded in a district court’s decision to enforce or quash a subpoena. Whether a charge is “valid,” EEOC v. Shell Oil Co., 466 U. S. 54, 72, n. 26 (1984) —that is, legally sufficient—is a pure question of law. And the question whether a district court employed the correct standard of relevance, see id., at 68–69—as opposed to how it applied that standard to the facts of a given case—is a question of law. But “applying a unitary abuse-of-discretion standard” does not shelter a district court that makes an error of law, because “[a] district court would necessarily abuse its discretion if it based its ruling on an erroneous view of the law.” Cooter & Gell v. Hartmarx Corp., 496 U. S. 384, 403, 405 (1990) .

582.US.2016_16-5294

Ake v. Oklahoma, 470 U. S. 68 , clearly established that when an indigent “defendant demonstrates . . . that his sanity at the time of the offense is to be a significant fact at trial, the State must” provide the defendant with “access to a competent psychiatrist who will conduct an appropriate examination and assist in evaluation, preparation, and presentation of the defense.” One month after Ake was decided, Alabama charged petitioner McWilliams with rape and murder. Finding him indigent, the trial court appointed counsel, who requested a psychiatric evaluation of McWilliams. The court granted the motion and the State convened a commission, which concluded that McWilliams was competent to stand trial and had not been suffering from mental illness at the time of the alleged offense. A jury convicted McWilliams of capital murder and recommended a death sentence. Later, while the parties awaited McWilliams’ judicial sentencing hearing, McWilliams’ counsel asked for neurological and neuropsychological testing of McWilliams. The court agreed and McWilliams was examined by Dr. Goff. Dr. Goff filed a report two days before the judicial sentencing hearing. He concluded that McWilliams was likely exaggerating his symptoms, but nonetheless appeared to have some genuine neuropsychological problems. Just before the hearing, counsel also received updated records from the commission’s evaluation and previously subpoenaed mental health records from the Alabama Department of Corrections. At the hearing, defense counsel requested a continuance in order to evaluate all the new material, and asked for the assistance of someone with expertise in psychological matters to review the findings. The trial court denied defense counsel’s requests. At the conclusion of the hearing, the court sentenced McWilliams to death. On appeal, McWilliams argued that the trial court denied him the right to meaningful expert assistance guarantee by Ake. The Alabama Court of Criminal Appeals affirmed McWilliams’ conviction and sentence, holding that Dr. Goff’s examination satisfied Ake’s requirements. The State Supreme Court affirmed, and McWilliams failed to obtain state postconviction relief. On federal habeas review, a Magistrate Judge also found that the Goff examination satisfied Ake and, therefore, that the State Court of Criminal Appeals’ decision was not contrary to, or an unreasonable application of, clearly established federal law. See 28 U. S. C. §2254(d)(1). Adopting the Magistrate Judge’s report and recommendation, the District Court denied relief. The Eleventh Circuit affirmed. Held: 1. Ake clearly established that when certain threshold criteria are met, the state must provide a defendant with access to a mental health expert who is sufficiently available to the defense and independent from the prosecution to effectively “conduct an appropriate examination and assist in evaluation, preparation, and presentation of the defense.” 470 U. S., at 83. The Alabama courts’ determination that McWilliams received all the assistance to which Ake entitled him was contrary to, or an unreasonable application of, clearly established federal law. Pp. 11–16. (a) Three preliminary issues require resolution. First, the conditions that trigger Ake’s application are present. McWilliams is and was an “indigent defendant,” 470 U. S., at 70, and his “mental condition” was both “relevant to . . . the punishment he might suffer,” id., at 80, and “seriously in question,” id., at 70. Second, this Court rejects Alabama’s claim the State was relieved of its Ake obligations because McWilliams received brief assistance from a volunteer psychologist at the University of Alabama. Even if the episodic help of an outside volunteer could satisfy Ake, the State does not refer to any specific record facts that indicate that the volunteer psychologist was available to the defense at the judicial sentencing proceeding. Third, contrary to Alabama’s suggestion, the record indicates that McWilliams did not get all the mental health assistance that he requested. Rather, he asked for additional help at the judicial sentencing hearing, but was rebuffed. Pp. 11–13. (b) This Court does not have to decide whether Ake requires a State to provide an indigent defendant with a qualified mental health expert retained specifically for the defense team. That is because Alabama did not meet even Ake’s most basic requirements in this case. Ake requires more than just an examination. It requires that the State provide the defense with “access to a competent psychiatrist who will conduct an appropriate [1] examination and assist in [2] evaluation, [3] preparation, and [4] presentation of the defense.” 470 U. S., at 83. Even assuming that Alabama met the examination requirement, it did not meet any of the other three. No expert helped the defense evaluate the Goff report or McWilliams’ extensive medical records and translate these data into a legal strategy. No expert helped the defense prepare and present arguments that might, e.g., have explained that McWilliams’ purported malingering was not necessarily inconsistent with mental illness. No expert helped the defense prepare direct or cross-examination of any witnesses, or testified at the judicial sentencing hearing. Since Alabama’s provision of mental health assistance fell so dramatically short of Ake’s requirements, the Alabama courts’ decision affirming McWilliams’ sentence was “contrary to, or involved an unreasonable application of, clearly established Federal law.” 28 U. S. C. §2254(d)(1). Pp. 13–16. 2. The Eleventh Circuit should determine on remand whether the Alabama courts’ error had the “substantial and injurious effect or influence” required to warrant a grant of habeas relief, Davis v. Ayala, 576 U. S. ___, ___, specifically considering whether access to the type of meaningful assistance in evaluating, preparing, and presenting the defense that Ake requires could have made a difference. P. 16. 634 Fed. Appx. 698, reversed and remanded. Breyer, J., delivered the opinion of the Court, in which Kennedy, Ginsburg, Sotomayor, and Kagan, JJ., joined. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas and Gorsuch, JJ., joined.

Thirty-one years ago, petitioner James Edmond McWilliams, Jr., was convicted of capital murder by an Alabama jury and sentenced to death. McWilliams challenged his sentence on appeal, arguing that the State had failed to provide him with the expert mental health assistance the Constitution requires, but the Alabama courts refused to grant relief. We now consider, in this habeas corpus case, whether the Alabama courts’ refusal was “contrary to, or involved an unreasonable application of, clearly established Federal law.” 28 U. S. C. §2254(d)(1). We hold that it was. Our decision in Ake v. Oklahoma, 470 U. S. 68 (1985) , clearly established that, when certain threshold criteria are met, the State must provide an indigent defendant with access to a mental health expert who is sufficiently available to the defense and independent from the prosecution to effectively “assist in evaluation, preparation, and presentation of the defense.” Id., at 83. Petitioner in this case did not receive that assistance. I McWilliams and the State of Alabama agree that Ake (which this Court decided in February 1985) sets forth the applicable constitutional standards. Before turning to the circumstances of McWilliams’ case, we describe what the Court held in Ake. We put in italics language that we find particularly pertinent here. The Court began by stating that the “issue in this case is whether the Constitution requires that an indigent defendant have access to the psychiatric examination and assistance necessary to prepare an effective defense based on his mental condition, when his sanity at the time of the offense is seriously in question.” Id., at 70 (emphasis added). The Court said it would consider that issue within the framework of earlier cases granting “an indigent defendant . . . a fair opportunity to present his defense” and “to participate meaningfully in a judicial proceeding in which his liberty is at stake.” Id., at 76. “Meaningful access to justice,” the Court added, “has been the consistent theme of these cases.” Id., at 77. The Court then wrote that “when the State has made the defendant’s mental condition relevant to his criminal culpability and to the punishment he might suffer, the assistance of a psychiatrist may well be crucial to the defendant’s ability to marshal his defense.” Id., at 80. A psychiatrist may, among other things, “gather facts,” “analyze the information gathered and from it draw plausible conclusions,” and “know the probative questions to ask of the opposing party’s psychiatrists and how to interpret their answers.” Ibid. These and related considerations “lea[d] inexorably to the conclusion that, without the assistance of a psychiatrist to conduct a professional examination on issues relevant to the defense, to help determine whether the insanity defense is viable, to present testimony, and to assist in preparing the cross-examination of a State’s psychiatric witnesses, the risk of an inaccurate resolution of sanity issues is extremely high. With such assistance, the defendant is fairly able to present at least enough information to the jury, in a meaningful manner, as to permit it to make a sensible determination.” Id., at 82 (emphasis added). The Court concluded: “We therefore hold that when a defendant demonstrates to the trial judge that his sanity at the time of the offense is to be a significant factor at trial, the State must, at a minimum, assure the defendant access to a competent psychiatrist who will conduct an appropriate examination and assist in evaluation, preparation, and presentation of the defense. . . . Our concern is that the indigent defendant have access to a competent psychiatrist for the[se] purpose[s].” Id., at 83 (emphasis added). Ake thus clearly establishes that when its threshold criteria are met, a State must provide a mental health professional capable of performing a certain role: “conduct[ing] an appropriate examination and assist[ing] in evaluation, preparation, and presentation of the defense.” Ibid. Unless a defendant is “assure[d]” the assistance of someone who can effectively perform these functions, he has not received the “minimum” to which Ake entitles him. Ibid. II A One month after this Court decided Ake, the State of Alabama charged McWilliams with rape and murder. The trial court found McWilliams indigent and provided him with counsel. It also granted counsel’s pretrial motion for a psychiatric evaluation of McWilliams’ sanity, including aspects of his mental condition relevant to “mitigating circumstances to be considered in a capital case in the sentencing stage.” T. 1526. (“T.” refers to the certified trial record; “P. C. T.” refers to the certified court reporter’s state postconviction proceedings transcript.) The court ordered the State to convene a “Lunacy Commission,” which would examine McWilliams and file a report with the court. See id., at 1528–1529. Subsequently a three-member Lunacy Commission examined McWilliams at a state hospital, the Taylor Hardin Secure Medical Facility. The three members, all psychiatrists, concluded that McWilliams was competent to stand trial and that he had not been suffering from mental illness at the time of the alleged offense. Id., at 1544–1546. One of them, Dr. Kamal Nagi, wrote that “Mr. McWilliams is grossly exaggerating his psychological symptoms to mimic mental illness.” Id., at 1546. Dr. Nagi noted that McWilliams’ performance on one of the tests “suggested that [McWilliams] had exaggerated his endorsement of symptoms of illness and the profile was considered a ‘fake bad.’ ” Ibid. McWilliams’ trial took place in late August 1986. On August 26 the jury convicted him of capital murder. The prosecution sought the death penalty, which under then-applicable Alabama law required both a jury recommendation (with at least 10 affirmative votes) and a later determination by the judge. See Ala. Code §13A–5–46(f) (1986). The jury-related portion of the sentencing proceeding took place the next day. The prosecution reintroduced evidence from the guilt phase and called a police officer to testify that McWilliams had a prior conviction. T. 1297, 1299–1303. The defense called McWilliams and his mother. Both testified that McWilliams, when a child, had suffered multiple serious head injuries. Id., at 1303–1318, 1320–1335. McWilliams also described his history of psychiatric and psychological evaluations, reading from the prearrest report of one psychologist, who concluded that McWilliams had a “blatantly psychotic thought disorder” and needed inpatient treatment. Id., at 1329–1332. When the prosecutor, cross-examining McWilliams, asked about the neurological effects of his head injuries, McWilliams replied, “I am not a psychiatrist.” Id., at 1328. Similarly, when the prosecutor asked McWilliams’ mother whether her son was “crazy,” she answered, “I am no expert: I don’t know whether my son is crazy or not. All I know, that my son do need help.” Id., at 1317. The prosecution then called two of the mental health professionals who had signed the Lunacy Commission’s report, Dr. Kamal Nagi and Dr. Norman Poythress. Dr. Nagi testified that he had found no evidence of psychosis, but did not appear to be aware of McWilliams’ history of head trauma. See id., at 1351–1352. Dr. Poythress testified that one of the tests that McWilliams took was “clinically invalid” because the test’s “validity scales” indicated that McWilliams had exaggerated or faked his symptoms. Id., at 1361–1363. Although McWilliams’ counsel had subpoenaed further mental health records from Holman State Prison, where McWilliams was being held, the jury did not have the opportunity to consider them, for, though subpoenaed on August 13, the records had not arrived by August 27, the day of the jury hearing. After the hearing, the jury recommended the death penalty by a vote of 10 to 2, the minimum required by Alabama law. The court scheduled its judicial sentencing hearing for October 9, about six weeks later. B Five weeks before that hearing, the trial court ordered the Alabama Department of Corrections to respond to McWilliams’s subpoena for mental health records. Id., at 1619. The court also granted McWilliams’ motion for neurological and neuropsychological exams. Id., at 1615–1617. That motion (apparently filed at the suggestion of a University of Alabama psychologist who had “volunteer[ed]” to help counsel “in her spare time,” P. C. T. 251–252) asked the court to “issue an order requiring the State of Alabama to do complete neurological and neuropsychological testing on the Defendant in order to have the test results available for his sentencing hearing.” T. 1615. Consequently, Dr. John Goff, a neuropsychologist employed by the State’s Department of Mental Health, examined McWilliams. On October 7, two days before the judicial sentencing hearing, Dr. Goff filed his report. The report concluded that McWilliams presented “some diagnostic dilemmas.” Id., at 1635. On the one hand, he was “obviously attempting to appear emotionally disturbed” and “exaggerating his neuropsychological problems.” Ibid. But on the other hand, it was “quite apparent that he ha[d] some genuine neuropsychological problems.” Ibid. Tests revealed “cortical dysfunction attributable to right cerebral hemisphere dysfunction,” shown by “left hand weakness, poor motor coordination of the left hand, sensory deficits including suppressions of the left hand and very poor visual search skills.” Id., at 1636. These deficiencies were “suggestive of a right hemisphere lesion” and “compatible with the injuries [McWilliams] sa[id] he sustained as a child.” Id., at 1635. The report added that McWilliams’ “obvious neuropsychological deficit” could be related to his “low frustration tolerance and impulsivity,” and suggested a diagnosis of “organic personality syndrome.” Ibid. The day before the sentencing hearing defense counsel also received updated records from Taylor Hardin hospital, and on the morning of the hearing he received the records (subpoenaed in mid-August) from Holman Prison. The prison records indicated that McWilliams was taking an assortment of psychotropic medications including Desyrel, Librium, and an antipsychotic, Mellaril. See App. 190a–193a. C The judicial sentencing hearing began on the morning of October 9. Defense counsel told the trial court that the eleventh-hour arrival of the Goff report and the mental health records left him “unable to present any evidence today.” Id., at 194a. He said he needed more time to go over the new information. Furthermore, since he was “not a psychologist or a psychiatrist,” he needed “to have someone else review these findings” and offer “a second opinion as to the severity of the organic problems discovered.” Id., at 192a–196a. The trial judge responded, “All right. Well, let’s proceed.” Id., at 197a. The prosecution then presented its case. Once it had finished, defense counsel moved for a continuance in order “to allow us to go through the material that has been provided to us in the last 2 days.” Id., at 204a. The judge offered to give defense counsel until 2 p.m. that afternoon. He also stated that “[a]t that time, The Court will entertain any motion that you may have with some other person to review” the new material. Id., at 205a. Defense counsel protested that “there is no way that I can go through this material,” but the judge immediately added, “Well, I will give you the opportunity. . . . If you do not want to try, then you may not.” Id., at 206a. The court then adjourned until 2 p.m. During the recess, defense counsel moved to withdraw. He said that “the abritrary [sic] position taken by this Court regarding the Defendant’s right to present mitigating circumstances is unconscionable resulting in this proceeding being a mockery.” T. 1644. He added that “further participation would be tantamount to exceptance [sic] of the Court’s ruling.” Ibid. The trial court denied the motion to withdraw. When the proceedings resumed, defense counsel renewed his motion for a continuance, explaining, “It is the position of the Defense that we have received these records at such a late date, such a late time that it has put us in a position as laymen, with regard to psychological matters, that we cannot adequately make a determination as what to present to The Court with regards to the particular deficiencies that the Defendant has. We believe that he has the type of diagnosed illness that we pointed out earlier for The Court and have mentioned for The Court. But we cannot determine ourselves from the records that we have received and the lack of receiving the test and the lack of our own expertise, whether or not such a condition exists; whether the reports and tests that have been run by Taylor Hardin, and the Lunacy Commission, and at Holman are tests that should be challenged in some type of way or the results should be challenged, we really need an opportunity to have the right type of experts in this field, take a look at all of those records and tell us what is happening with him. And that is why we renew the Motion for a Continuance.” App. 207a. The trial court denied the motion. The prosecutor then offered his closing statement, in which he argued that there were “no mitigating circumstances.” Id., at 209a. Defense counsel replied that he “would be pleased to respond to [the prosecutor’s] remarks that there are no mitigating circumstances in this case if I were able to have time to produce . . . any mitigating circumstances.” Id., at 210a. But, he said, since neither he nor his co-counsel were “doctors,” neither was “really capable of going through those records on our own.” Ibid. The court had thus “foreclosed by structuring this hearing as it has, the Defendant from presenting any evidence of mitigation in psychological—psychiatric terms.” Id., at 211a. The trial judge then said that he had reviewed the records himself and found evidence that McWilliams was faking and manipulative. Ibid. Defense counsel attempted to contest that point, which led to the following exchange: “MR. SOGOL: I told Your Honor that my looking at those records was not of any value to me; that I needed to have somebody look at those records who understood them, who could interpret them for me. Did I not tell Your Honor that? THE COURT: As I said, on the record earlier, Mr. Sogol, and I don’t want to argue or belabor this, but I would have given you the opportunity to make a motion to present someone to evaluate that. MR. SOGOL: Your Honor gave me no time in which to do that. Your Honor told me to be here at 2 o’clock this afternoon. Would Your Honor have wanted me to file a Motion for Extraordinary Expenses to get someone? THE COURT: I want you to approach with your client, please.” Id., at 211a–212a. The court then sentenced McWilliams to death. The court later issued a written sentencing order. It found three aggravating circumstances and no mitigating circumstances. It found that McWilliams “was not and is not psychotic,” and that “the preponderance of the evidence from these tests and reports show [McWilliams] to be feigning, faking, and manipulative.” Id., at 188a. The court wrote that even if McWilliams’ mental health issues “did rise to the level of a mitigating circumstance, the aggravating circumstances would far outweigh this as a mitigating circumstance.” Ibid. D McWilliams appealed, arguing that the trial court had denied him the right to meaningful expert assistance guaranteed by Ake. The Alabama Court of Criminal Appeals rejected his argument. It wrote that Ake’s requirements “are met when the State provides the [defendant] with a competent psychiatrist.” McWilliams v. State, 640 So. 2d 982, 991 (1991). And Alabama, by “allowing Dr. Goff to examine” McWilliams, had satisfied those requirements. Ibid. The court added that “[t]here is no indication in the record that [McWilliams] could not have called Dr. Goff as a witness to explain his findings or that he even tried to contact the psychiatrist to discuss his findings,” ibid.; that “the trial court indicated that it would have considered a motion to present an expert to evaluate this report” had one been made, ibid.; and that there was “no prejudice by the trial court’s denial of [McWilliams’] motion for continuance,” id., at 993. The appeals court therefore affirmed McWilliams’ conviction and sentence. The Alabama Supreme Court, in turn, affirmed the appeals court (without addressing the Ake issue). Ex parte McWilliams, 640 So. 2d 1015 (1993). After McWilliams failed to obtain postconviction relief from the state courts, he sought a federal writ of habeas corpus. See 28 U. S. C. §2254. E In federal habeas court McWilliams argued before a Magistrate Judge that he had not received the expert assistance that Ake required. The Magistrate Judge recommended against issuing the writ. He wrote that McWilliams had “received the assistance required by Ake” because Dr. Goff “completed the testing” that McWilliams requested. App. 88a. Hence, the decision of the Alabama Court of Criminal Appeals was not contrary to, or an unreasonable application of, clearly established federal law. See 28 U. S. C. §2254(d)(1). The District Court adopted the Magistrate Judge’s report and recommendation and denied relief. A divided panel of the Eleventh Circuit Court of Appeals affirmed. See McWilliams v. Commissioner, Ala. Dept. of Corrections, 634 Fed. Appx. 698 (2015) ( per curiam); id., at 711 (Jordan, J., concurring); id., at 712 (Wilson, J., dissenting). McWilliams filed a petition for certiorari. We granted the petition. III A The question before us is whether the Alabama Court of Criminal Appeals’ determination that McWilliams got all the assistance to which Ake entitled him was “contrary to, or involved an unreasonable application of, clearly established Federal law.” 28 U. S. C. §2254(d)(1). Before turning to the heart of that question, we resolve three preliminary issues. First, no one denies that the conditions that trigger application of Ake are present. McWilliams is and was an “indigent defendant,” 470 U. S., at 70. See supra, at 3. His “mental condition” was “relevant to . . . the punishment he might suffer,” 470 U. S., at 80. See supra, at 4–5. And, that “mental condition,” i.e., his “sanity at the time of the offense,” was “seriously in question,” 470 U. S., at 70. See supra, at 4–5. Consequently, the Constitution, as interpreted in Ake, required the State to provide McWilliams with “access to a competent psychiatrist who will conduct an appropriate examination and assist in evaluation, preparation, and presentation of the defense.” 470 U. S., at 83. Second, we reject Alabama’s claim that the State was exempted from its obligations because McWilliams already had the assistance of Dr. Rosenszweig, the psychologist at the University of Alabama who “volunteer[ed]” to help defense counsel “in her spare time” and suggested the defense ask for further testing, P. C. T. 251–252. Even if the episodic assistance of an outside volunteer could relieve the State of its constitutional duty to ensure an indigent defendant access to meaningful expert assistance, no lower court has held or suggested that Dr. Rosenszweig was available to help, or might have helped, McWilliams at the judicial sentencing proceeding, the proceeding here at issue. Alabama does not refer to any specific record facts that indicate that she was available to the defense at this time. Third, Alabama argues that Ake’s requirements are irrelevant because McWilliams “never asked for more expert assistance” than he got, “even though the trial court gave him the opportunity to do so.” Brief for Respondent 50–51. The record does not support this contention. When defense counsel requested a continuance at the sentencing hearing, he repeatedly told the court that he needed “to have someone else review” the Goff report and medical records. App. 193a. See, e.g., id., at 196a (“[I]t is just incumbent upon me to have a second opinion as to the severity of the organic problems discovered”); id., at 207a (“[W]e really need an opportunity to have the right type of experts in this field, take a look at all of these records and tell us what is happening with him”); id., at 211a (“I told Your Honor that my looking at these records was not of any value to me; that I needed to have somebody look at those records who understood them, who could interpret them for me”). Counsel also explicitly asked the trial court what else he was supposed to ask for to obtain an expert: “Would Your Honor have wanted me to file a Motion for Extraordinary Expenses to get someone?” Id., at 212a. We have reproduced a lengthier account of the exchanges, supra, at 7–9. They make clear that counsel wanted additional expert assistance to review the report and records—that was the point of asking for a continuance. In response, the court told counsel to approach the bench and sentenced McWilliams to death. Thus the record, in our view, indicates that McWilliams did request additional help from mental health experts. B We turn to the main question before us: whether the Alabama Court of Criminal Appeals’ determination that McWilliams got all the assistance that Ake requires was “contrary to, or involved an unreasonable application of, clearly established Federal law.” 28 U. S. C. §2254(d)(1). McWilliams would have us answer “yes” on the ground that Ake clearly established that a State must provide an indigent defendant with a qualified mental health expert retained specifically for the defense team, not a neutral expert available to both parties. He points to language in Ake that seems to foresee that consequence. See, e.g., 470 U. S., at 81 (“By organizing a defendant’s mental history, examination results and behavior, and other information, interpreting it in light of their expertise, and then laying out their investigative and analytic process to the jury, the psychiatrists for each party enable the jury to make its most accurate determination of the truth on the issue before them” (emphasis added)). We need not, and do not, decide, however, whether this particular McWilliams claim is correct. As discussed above, Ake clearly established that a defendant must receive the assistance of a mental health expert who is sufficiently available to the defense and independent from the prosecution to effectively “assist in evaluation, preparation, and presentation of the defense.” Id., at 83. As a practical matter, the simplest way for a State to meet this standard may be to provide a qualified expert retained specifically for the defense team. This appears to be the approach that the overwhelming majority of jurisdictions have adopted. See Brief for National Association of Criminal Defense Lawyers et al. as Amici Curiae 8–35 (describing practice in capital-active jurisdictions); Tr. of Oral Arg. 40 (respondent conceding that “this issue really has been mooted over the last 30-some-odd years because of statutory changes”). It is not necessary, however, for us to decide whether the Constitution requires States to satisfy Ake’s demands in this way. That is because Alabama here did not meet even Ake’s most basic requirements. The dissent calls our unwillingness to resolve the broader question whether Ake clearly established a right to an ex-pert independent from the prosecution a “most unseemly maneuver.” Post, at 1–2 (opinion of Alito, J.). We do not agree. We recognize that we granted petitioner’s first question presented—which addressed whether Ake clearly established a right to an independent expert—and not his second, which raised more case-specific concerns. See Pet. for Cert. i. Yet that does not bind us to issue a sweeping ruling when a narrow one will do. As we explain below, our determination that Ake clearly established that a defendant must receive the assistance of a mental health expert who is sufficiently available to the defense and independent from the prosecution to effectively “assist in evaluation, preparation, and presentation of the defense,” 470 U. S., at 83, is sufficient to resolve the case. We therefore need not decide whether Ake clearly established more. (Nor do we agree with the dissent that our approach is “acutely unfair to Alabama” by not “giv[ing] the State a fair chance to respond.” Post, at 12. In fact, the State devoted an entire section of its merits brief to explaining why it thought that “[n]o matter how the Court resolves the [independent expert] question, the court of appeals correctly denied the habeas petition.” Brief for Respondent 50. See also id., at 14, 52 (referring to the lower courts’ case-specific determinations that McWilliams got all the assistance Ake requires).) The Alabama appeals court held that “the requirements of Ake v. Oklahoma . . . are met when the State provides the [defendant] with a competent psychiatrist. The State met this requirement in allowing Dr. Goff to examine [McWilliams].” McWilliams, 640 So. 2d, at 991. This was plainly incorrect. Ake does not require just an examination. Rather, it requires the State to provide the defense with “access to a competent psychiatrist who will conduct an appropriate [1] examination and assist in [2] evaluation, [3] preparation, and [4] presentation of the defense.” Ake, supra, at 83 (emphasis added). We are willing to assume that Alabama met the examination portion of this requirement by providing for Dr. Goff’s examination of McWilliams. See supra, at 6. But what about the other three parts? Neither Dr. Goff nor any other expert helped the defense evaluate Goff’s report or McWilliams’ extensive medical records and translate these data into a legal strategy. Neither Dr. Goff nor any other expert helped the defense prepare and present arguments that might, for example, have explained that McWilliams’ purported malingering was not necessarily inconsistent with mental illness (as an expert later testified in postconviction proceedings, see P. C. T. 936–943). Neither Dr. Goff nor any other expert helped the defense prepare direct or cross-examination of any witnesses, or testified at the judicial sentencing hearing himself. The dissent emphasizes that Dr. Goff was never ordered to do any of these things by the trial court. See post, at 13, n. 5. But that is precisely the point. The relevant court order did not ask Dr. Goff or anyone else to provide the defense with help in evaluating, preparing, and presenting its case. It only required “the Department of Corrections” to “complete neurological and neuropsychological testing on the Defendant . . . and send all test materials, results and evaluations to the Clerk of the Court.” T. 1612. Nor did the short time frame allow for more expert assistance. (Indeed, given that timeframe, we do not see how Dr. Goff or any other expert could have satisfied the latter three portions of Ake’s requirements even had he been instructed to do so.) Then, when McWilliams asked for the addi-tional assistance to which he was constitutionally entitled at the sentencing hearing, the judge rebuffed his requests. See supra, at 7–9. Since Alabama’s provision of mental health assistance fell so dramatically short of what Ake requires, we must conclude that the Alabama court decision affirming McWilliams’s conviction and sentence was “contrary to, or involved an unreasonable application of, clearly established Federal law.” 28 U. S. C. §2254(d)(1). IV The Eleventh Circuit held in the alternative that, even if the Alabama courts clearly erred in their application of federal law, their “error” nonetheless did not have the “substantial and injurious effect or influence” required to warrant a grant of habeas relief, Davis v. Ayala, 576 U. S. ___, ___ (2015) (slip op., at 10) (internal quotation marks omitted). See 634 Fed. Appx., at 707. In reaching this conclusion, however, the Eleventh Circuit only considered whether “[a] few additional days to review Dr. Goff’s findings” would have made a difference. Ibid. It did not specifically consider whether access to the type of meaningful assistance in evaluating, preparing, and presenting the defense that Ake requires would have mattered. There is reason to think that it could have. For example, the trial judge relied heavily on his belief that McWilliams was malingering. See App. 188a, 211a. If McWilliams had the assistance of an expert to explain that “[m]alingering is not inconsistent with serious mental illness,” Brief for American Psychiatric Association et al. as Amici Curiae 20, he might have been able to alter the judge’s perception of the case. Since “we are a court of review, not of first view,” Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005), we do not now resolve this question. Rather we leave it to the lower courts to decide in the first instance. The judgment of the Court of Appeals is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

582.US.2016_15-457

Orders granting or denying class certification are inherently interlocutory, hence not immediately reviewable under 28 U. S. C. §1291, which empowers federal courts of appeals to review only “final decisions of the district courts.” In Coopers & Lybrand v. Livesay, 437 U. S. 463 , a 1978 decision, this Court held that the death-knell doctrine—which rested on courts’ recognition that a denial of class certification would sometimes end a lawsuit for all practical purposes—did not warrant mandatory appellate jurisdiction of certification orders. Id., at 470, 477. Although the death-knell theory likely “enhanced the quality of justice afforded a few litigants,” it did so at a heavy cost to §1291’s finality requirement. Id., at 473. First, the potential for multiple interlocutory appeals inhered in the doctrine. See id., at 474. Second, the death-knell theory forced appellate courts indiscriminately into the trial process, circumventing the two-tiered “screening procedure” Congress established for interlocutory appeals in 28 U. S. C. §1292(b). Id., at 474, 476. Finally, the doctrine “operat[ed] only in favor of plaintiffs,” even though the class-certification question may be critically important to defendants as well. Id., at 476. Two decades later, in 1998, after Congress amended the Rules Enabling Act, 28 U. S. C. §2071 et seq., to empower this Court to promulgate rules providing for interlocutory appeal of orders “not otherwise provided for [in §1292],” §1292(e), this Court approved Federal Rule of Civil Procedure 23(f). Rule 23(f) authorizes “permissive interlocutory appeal” from adverse class-certification orders in “the sole discretion of the court of appeals.” 28 U. S. C. App., p. 815. This discretionary arrangement was the product of careful calibration on the part of the rulemakers. Respondents, owners of Microsoft’s videogame console, the Xbox 360, filed this putative class action alleging a design defect in the device. The District Court struck respondents’ class allegations from the complaint, and the Court of Appeals denied respondents permission to appeal that order under Rule 23(f). Instead of pursuing their individual claims to final judgment on the merits, respondents stipulated to a voluntary dismissal of their claims with prejudice, but reserved the right to revive their claims should the Court of Appeals reverse the District Court’s certification denial. Respondents then appealed, challenging only the interlocutory order striking their class allegations. The Ninth Circuit held it had jurisdiction to entertain the appeal under §1291. It then held that the District Court’s rationale for striking respondents’ class allegations was an impermissible one, but refused to opine on whether class certification was inappropriate for a different reason, leaving that question for the District Court on remand. Held: Federal courts of appeals lack jurisdiction under §1291 to review an order denying class certification (or, as here, an order striking class allegations) after the named plaintiffs have voluntarily dismissed their claims with prejudice. Pp. 11–17. (a) Section 1291’s final-judgment rule preserves the proper balance between trial and appellate courts, minimizes the harassment and delay that would result from repeated interlocutory appeals, and promotes the efficient administration of justice. This Court has resisted efforts to stretch §1291 to permit appeals of right that would erode the finality principle and disserve its objectives. See, e.g., Mohawk Industries, Inc. v. Carpenter, 558 U. S. 100 . Attempts to secure appeal as of right from adverse class certification orders fit that bill. Pp. 11–12. (b) Respondents’ voluntary-dismissal tactic, even more than the death-knell theory, invites protracted litigation and piecemeal appeals. Under the death-knell doctrine, a court of appeals could decline to hear an appeal if it determined that the plaintiff “ha[d] adequate incentive to continue” despite the denial of class certification. Coopers & Lybrand, 437 U. S., at 471. Under respondents’ theory, however, the decision whether an immediate appeal will lie resides exclusively with the plaintiff, who need only dismiss her claims with prejudice in order to appeal the district court’s order denying class certification. And she may exercise that option more than once, interrupting district court proceedings with an interlocutory appeal again, should the court deny class certification on a different ground. Respondents contend that their position promotes efficiency, observing that after dismissal with prejudice the case is over if the plaintiff loses on appeal. But plaintiffs with weak merits claims may readily assume that risk, mindful that class certification often leads to a hefty settlement. And the same argument was evident in the death-knell context, yet this Court determined that the potential for piecemeal litigation was “apparent and serious.” Id., at 474. That potential is greater still under respondents’ theory, where plaintiffs alone determine whether and when to appeal an adverse certification ruling. Pp. 12–14. (c) Also like the death-knell doctrine, respondents’ theory allows indiscriminate appellate review of interlocutory orders. Beyond disturbing the “ ‘appropriate relationship between the respective courts,’ ” Coopers & Lybrand, 437 U. S., at 476, respondents’ dismissal tactic undercuts Rule 23(f)’s discretionary regime. This consideration is “[o]f prime significance to the jurisdictional issue” in this case, Swint v. Chambers County Comm’n, 514 U. S. 35 , because Congress has established rulemaking as the means for determining when a decision is final for purposes of §1291 and for providing for appellate review of interlocutory orders not covered by statute, see §§2072(c) and 1292(e). Respondents maintain that Rule 23(f) is irrelevant, for it concerns interlocutory orders, whereas this case involves an actual final judgment. Yet permitting respondents’ voluntary-dismissal tactic to yield an appeal of right would seriously undermine Rule 23(f)’s careful calibration, as well as Congress’ designation of rulemaking “as the preferred means for determining whether and when prejudgment orders should be immediately appealable,” Mohawk Industries, 558 U. S., at 113. Plaintiffs in putative class actions cannot transform a tentative interlocutory order into a final judgment within the meaning of §1291 simply by dismissing their claims with prejudice. Finality “is not a technical concept of temporal or physical termination.” Cobbledick v. United States, 309 U. S. 323 . It is one “means [geared to] achieving a healthy legal system,” ibid., and its contours are determined accordingly. Pp. 14–16. (d) The one-sidedness of respondents’ voluntary-dismissal device reinforces the conclusion that it does not support mandatory appellate jurisdiction of refusals to grant class certification. The tactic permits only plaintiffs, never defendants, to force an immediate appeal of an adverse certification ruling. Yet the “class issue” may be just as important to defendants, Coopers & Lybrand, 437 U. S., at 476, for class certification may force a defendant to settle rather than run the risk of ruinous liability. P. 17. 797 F. 3d 607, reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed an opinion concurring in the judgment, in which Roberts, C. J., and Alito, J., joined. Gorsuch, J., took no part in the consideration or decision of the case.

This case concerns options open to plaintiffs, when denied class-action certification by a district court, to gain appellate review of the district court’s order. Orders granting or denying class certification, this Court has held, are “inherently interlocutory,” Coopers & Lybrand v. Livesay, 437 U. S. 463, 470 (1978) , hence not immediately reviewable under 28 U. S. C. §1291, which provides for appeals from “final decisions.” Pursuant to Federal Rule of Civil Procedure 23(f), promulgated in 1998, however, orders denying or granting class certification may be appealed immediately if the court of appeals so permits. Absent such permission, plaintiffs may pursue their individual claims on the merits to final judgment, at which point the denial of class-action certification becomes ripe for review. The plaintiffs in the instant case, respondents here, were denied Rule 23(f) permission to appeal the District Court’s refusal to grant class certification. Instead of pursuing their individual claims to final judgment on the merits, respondents stipulated to a voluntary dismissal of their claims “with prejudice,” but reserved the right to revive their claims should the Court of Appeals reverse the District Court’s certification denial. We hold that the voluntary dismissal essayed by respondents does not qualify as a “final decision” within the compass of §1291. The tactic would undermine §1291’s firm finality principle, designed to guard against piecemeal appeals, and subvert the balanced solution Rule 23(f ) put in place for immediate review of class-action orders. I A Under §1291 of the Judicial Code, federal courts of appeals are empowered to review only “final decisions of the district courts.” 28 U. S. C. §1291.[1] Two guides, our decision in Coopers & Lybrand v. Livesay, 437 U. S. 463 (1978) , and Federal Rule of Civil Procedure 23(f), control our application of that finality rule here. 1 In Coopers & Lybrand, this Court considered whether a plaintiff in a putative class action may, under certain circumstances, appeal as of right a district court order striking class allegations or denying a motion for class certification. We held unanimously that the so-called “death-knell” doctrine did not warrant mandatory appellate jurisdiction of such “inherently interlocutory” orders. 437 U. S., at 470, 477. Courts of Appeals employing the doctrine “regarded [their] jurisdiction as depending on whether [rejection of class-action status] had sounded the ‘death knell’ of the action.” Id., at 466. These courts asked whether the refusal to certify a class would end a lawsuit for all practical purposes because the value of the named plaintiff’s individual claims made it “economically imprudent to pursue his lawsuit to a final judgment and [only] then seek appellate review of [the] adverse class determination.” Id., at 469–470. If, in the court of appeals’ view, the order would terminate the litigation, the court deemed the order an appealable final decision under §1291. Id., at 471. If, instead, the court determined that the plaintiff had “adequate incentive to continue [litigating], the order [was] considered interlocutory.” Ibid. Consequently, immediate appeal would be denied. The death-knell theory likely “enhance[d] the quality of justice afforded a few litigants,” we recognized. Id., at 473. But the theory did so, we observed, at a heavy cost to §1291’s finality requirement, and therefore to “the judicial system’s overall capacity to administer justice.” Id., at 473; see id., at 471 (Section 1291 “evinces a legislative judgment that ‘restricting appellate review to final decisions prevents the debilitating effect on judicial administration caused by piecemeal appeal disposition.’ ” (quoting Eisen v. Carlisle & Jacquelin, 417 U. S. 156, 170 (1974) (alterations and internal quotation marks omitted))). First, the potential for multiple interlocutory appeals inhered in the doctrine: When a ruling denying class certification on one ground was reversed on appeal, a death-knell plaintiff might again claim “entitle[ment] to an appeal as a matter of right” if, on remand, the district court denied class certification on a different ground. Coopers & Lybrand, 437 U. S., at 474. Second, the doctrine forced appellate courts indiscriminately into the trial process, thereby defeating a “vital purpose of the final-judgment rule—that of maintaining the appropriate relationship between the respective courts.” Id., at 476 (internal quotation marks omitted); see id., at 474. The Interlocutory Appeals Act of 1958, 28 U. S. C. §1292(b), we explained, had created a two-tiered “screening procedure” to preserve this relationship and to restrict the availability of interlocutory review to “appropriate cases.” 437 U. S., at 474. For a party to obtain review under §1292(b), the district court must certify that the interlocutory order “involves a controlling question of law as to which there is substantial ground for difference of opinion and that an immediate appeal from the order may materially advance the ultimate termination of the litigation.” The court of appeals may then, “in its discretion, permit an appeal to be taken from such order.” The death-knell doctrine, we stressed, “circumvent[ed] [§1292(b)’s] restrictions.” Id., at 475. Finally, we observed, the doctrine was one sided: It “operate[d] only in favor of plaintiffs,” even though the class-certification question is often “of critical importance to defendants as well.” Id., at 476. Just as a denial of class certification may sound the death knell for plaintiffs, “[c]ertification of a large class may so increase the defendant’s potential damages liability and litigation costs that he may find it economically prudent to settle and to abandon a meritorious defense.” Ibid.[2] In view of these concerns, the Court reached this conclusion in Coopers & Lybrand: “The fact that an interlocutory order may induce a party to abandon his claim before final judgment is not a sufficient reason for considering [the order] a ‘final decision’ within the meaning of §1291.” Id., at 477.[3] 2 After Coopers & Lybrand, a party seeking immediate review of an adverse class-certification order had no easy recourse. The Federal Rules of Civil Procedure did not then “contain any unique provisions governing appeals” in class actions, id., at 470, so parties had to survive §1292(b)’s two-level inspection, see id., at 474–475, and n. 27; supra, at 3–4, or satisfy the extraordinary-circumstances test applicable to writs of mandamus, see Will v. United States, 389 U. S. 90, 108 (1967) (Black, J., concurring) (“[In] extraordinary circumstances, mandamus may be used to review an interlocutory order which is by no means ‘final’ and thus appealable under federal statutes.”); cf. Coopers & Lybrand, 437 U. S., at 466, n. 6. Another avenue opened in 1998 when this Court approved Federal Rule of Civil Procedure 23(f). Seen as a response to Coopers & Lybrand, see, e.g., Blair v. Equifax Check Services, Inc., 181 F. 3d 832, 834 (CA7 1999); Solimine & Hines, Deciding To Decide: Class Action Certification and Interlocutory Review by the United States Courts of Appeals Under Rule 23(f), 41 Wm. & Mary L. Rev. 1531, 1568 (2000), Rule 23(f) authorizes “permissive interlocutory appeal” from adverse class-certification orders in the discretion of the court of appeals, Advisory Committee’s 1998 Note on subd. (f) of Fed. Rule Civ. Proc. 23, 28 U. S. C. App., p. 815 (hereinafter Committee Note on Rule 23(f)). The Rule was adopted pursuant to §1292(e), see Committee Note on Rule 23(f), which empowers this Court, in accordance with the Rules Enabling Act, 28 U. S. C. §2072, to promulgate rules “to provide for an appeal of an interlocutory decision to the courts of appeals that is not otherwise provided for [in §1292].” §1292(e).[4] Rule 23(f) reads: “A court of appeals may permit an appeal from an order granting or denying class-action certification . . . if a petition for permission to appeal is filed with the circuit clerk within 14 days after the order is entered. An appeal does not stay proceedings in the district court unless the district judge or the court of appeals so orders.”[5] Courts of appeals wield “unfettered discretion” under Rule 23(f), akin to the discretion afforded circuit courts under §1292(b). Committee Note on Rule 23(f). But Rule 23(f) otherwise “departs from the §1292(b) model,” for it requires neither district court certification nor adherence to §1292(b)’s other “limiting requirements.” Committee Note on Rule 23(f); see supra, at 3–4. This resolution was the product of careful calibration. By “[r]emoving the power of the district court to defeat any opportunity to appeal,” the drafters of Rule 23(f) sought to provide “significantly greater protection against improvident certification decisions than §1292(b)” alone offered. Judicial Conference of the United States, Advisory Committee on Civil Rules, Minutes of November 9–10, 1995. But the drafters declined to go further and provide for appeal as a matter of right. “[A] right to appeal would lead to abuse” on the part of plaintiffs and defendants alike, the drafters apprehended, “increas[ing] delay and expense” over “routine class certification decisions” unworthy of immediate appeal. Ibid. (internal quotation marks omitted). See also Brief for Civil Procedure Scholars as Amici Curiae 6–7, 11–14 (“Rule 23(f) was crafted to balance the benefits of immediate review against the costs of interlocutory appeals.” (capitalization omitted)). Rule 23(f ) therefore commits the decision whether to permit interlocutory appeal from an adverse certification decision to “the sole discretion of the court of appeals.” Committee Note on Rule 23(f); see Federal Judicial Center, T. Willging, L. Hooper, & R. Niemic, Empirical Study of Class Actions in Four Federal District Courts: Final Report to the Advisory Committee on Civil Rules 86 (1996) (hereinafter Federal Judicial Center Study) (“The discretionary nature of the proposed rule . . . is designed to be a guard against abuse of the appellate process.”).[6] The Rules Committee offered some guidance to courts of appeals considering whether to authorize appeal under Rule 23(f). “Permission is most likely to be granted,” the Committee Note states, “when the certification decision turns on a novel or unsettled question of law,” or when “the decision on certification is likely dispositive of the litigation,” as in a death-knell or reverse death-knell situation. Committee Note on Rule 23(f); see supra, at 4, and n. 2. Even so, the Rule allows courts of appeals to grant or deny review “on the basis of any consideration.” Committee Note on Rule 23(f) (emphasis added). B With this background in mind, we turn to the putative class action underlying our jurisdictional inquiry. The lawsuit is not the first of its kind. A few years after petitioner Microsoft Corporation released its popular videogame console, the Xbox 360, a group of Xbox owners brought a putative class action against Microsoft based on an alleged design defect in the device. See In re Microsoft Xbox 360 Scratched Disc Litigation, 2009 WL 10219350, *1 (WD Wash., Oct. 5, 2009). The named plaintiffs, advised by some of the same counsel representing respondents in this case, asserted that the Xbox scratched (and thus destroyed) game discs during normal game-playing conditions. See ibid. The District Court denied class certification, holding that individual issues of damages and causation predominated over common issues. See id., at *6–*7. The plaintiffs petitioned the Ninth Circuit under Rule 23(f) for leave to appeal the class-certification denial, but the Ninth Circuit denied the request. See 851 F. Supp. 2d 1274, 1276 (WD Wash. 2012). Thereafter, the Scratched Disc plaintiffs settled their claims individually. 851 F. Supp. 2d, at 1276. Two years later, in 2011, respondents filed this lawsuit in the same Federal District Court. They proposed a nationwide class of Xbox owners based on the same design defect alleged in Scratched Disc Litigation. See 851 F. Supp. 2d, at 1275–1276. The class-certification analysis in the earlier case did not control, respondents urged, because an intervening Ninth Circuit decision constituted a change in law sufficient to overcome the deference ordinarily due, as a matter of comity, the previous certification denial. Id., at 1277–1278. The District Court disagreed. Concluding that the relevant Circuit decision had not undermined Scratched Disc Litigation’s causation analysis, the court determined that comity required adherence to the earlier certification denial and therefore struck respondents’ class allegations. 851 F. Supp. 2d, at 1280–1281. Invoking Rule 23(f), respondents petitioned the Ninth Circuit for permission to appeal that ruling.[7] Interlocu-tory review was appropriate in this case, they argued, be-cause the District Court’s order striking the class allegations created a “death-knell situation”: The “small size of [their] claims ma[de] it economically irrational to bear the cost of litigating th[e] case to final judgment,” they asserted, so the order would “effectively kil[l] the case.” Pet. for Permission To Appeal Under Rule 23(f) in No. 12–80085 (CA9), App. 118. The Ninth Circuit denied the petition. Order in No. 12–80085 (CA9, June 12, 2012), App. 121. Respondents then had several options. They could have settled their individual claims like their Scratched Disc predecessors or petitioned the District Court, pursuant to §1292(b), to certify the interlocutory order for appeal, see supra, at 3–4. They could also have proceeded to litigate their case, mindful that the District Court could later reverse course and certify the proposed class. See Fed. Rule Civ. Proc. 23(c)(1)(C) (“An order that grants or denies class certification may be altered or amended before final judgment.”); Coopers & Lybrand, 437 U. S., at 469 (a certification order “is subject to revision in the District Court”). Or, in the event the District Court did not change course, respondents could have litigated the case to final judgment and then appealed. Id., at 469 (“an order denying class certification is subject to effective review after final judgment at the behest of the named plaintiff”). Instead of taking one of those routes, respondents moved to dismiss their case with prejudice. “After the [c]ourt has entered a final order and judgment,” respondents explained, they would “appeal the . . . order striking [their] class allegations.” Motion To Dismiss in No. 11–cv–00722 (WD Wash., Sept. 25, 2012), App. 122–123. In respondents’ view, the voluntary dismissal enabled them “to pursue their individual claims or to pursue relief solely on behalf of the class, should the certification decision be reversed.” Brief for Respondents 15. Microsoft stipulated to the dismissal, but maintained that respondents would have “no right to appeal” the order striking the class allegations after thus dismissing their claims. App. to Pet. for Cert. 35a–36a. The District Court granted the stipulated motion to dismiss, id., at 39a, and respondents appealed. They challenged only the District Court’s interlocutory order striking their class allegations, not the dismissal order which they invited. See Brief for Plaintiffs-Appellants in No. 12–35946 (CA9). The Ninth Circuit held it had jurisdiction to entertain the appeal under §1291. 797 F. 3d 607, 612 (2015). The Court of Appeals rejected Microsoft’s argument that respondents’ voluntary dismissal, explicitly engineered to appeal the District Court’s interlocutory order striking the class allegations, impermissibly circumvented Rule 23(f). Ibid., n. 3. Because the stipulated dismissal “did not involve a settlement,” the court reasoned, it was “ ‘a sufficiently adverse—and thus appealable—final decision’ ” under §1291. Id., at 612 (quoting Berger v. Home Depot USA, Inc., 741 F. 3d 1061, 1065 (CA9 2014)); see id., at 1065 (relying on 7B C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §1802, pp. 297–298 (3d ed. 2005), for the proposition “that finality for appeal purposes can be achieved in this manner”). Satisfied of its jurisdiction, the Ninth Circuit held that the District Court had abused its discretion in striking respondents’ class allegations. 797 F. 3d, at 615. The Court of Appeals “express[ed] no opinion on whether” respondents “should prevail on a motion for class certification,” ibid., concluding only that the District Court had misread recent Circuit precedent, see id., at 613–615, and therefore misapplied the comity doctrine, id., at 615. Whether a class should be certified, the court said, was a question for remand, “better addressed if and when [respondents] move[d] for class certification.” Ibid. We granted certiorari to resolve a Circuit conflict over this question: Do federal courts of appeals have jurisdiction under §1291 and Article III of the Constitution to review an order denying class certification (or, as here, an order striking class allegations) after the named plaintiffs have voluntarily dismissed their claims with prejudice?[8] 577 U. S. ___ (2016). Because we hold that §1291 does not countenance jurisdiction by these means, we do not reach the constitutional question, and therefore do not address the arguments and analysis discussed in the opinion concurring in the judgment. II “From the very foundation of our judicial system,” the general rule has been that “the whole case and every matter in controversy in it [must be] decided in a single appeal.” McLish v. Roff, 141 U. S. 661 –666 (1891). This final-judgment rule, now codified in §1291, preserves the proper balance between trial and appellate courts, minimizes the harassment and delay that would result from repeated interlocutory appeals, and promotes the efficient administration of justice. See Firestone Tire & Rubber Co. v. Risjord, 449 U. S. 368, 374 (1981) . Construing §1291 in line with these reasons for the rule, we have recognized that “finality is to be given a practical rather than a technical construction.” Eisen, 417 U. S., at 171 (internal quotation marks omitted). Repeatedly we have resisted efforts to stretch §1291 to permit appeals of right that would erode the finality principle and disserve its objectives. See, e.g., Mohawk Industries, Inc. v. Carpenter, 558 U. S. 100, 112 (2009) ; Digital Equipment Corp. v. Desktop Direct, Inc., 511 U. S. 863 –879, 884 (1994); Cobbledick v. United States, 309 U. S. 323 –325, 330 (1940) (construing §1291’s predecessor statute). Attempts to secure appeal as of right from adverse class-certification orders fit that bill. See supra, at 2–4. Because respondents’ dismissal device subverts the final-judgment rule and the process Congress has established for refining that rule and for determining when nonfinal orders may be immediately appealed, see §§2072(c) and 1292(e), the tactic does not give rise to a “final decisio[n]” under §1291. A Respondents’ voluntary-dismissal tactic, even more than the death-knell theory, invites protracted litigation and piecemeal appeals. Under the death-knell doctrine, a court of appeals could decline to hear an appeal if it determined that the plaintiff “ha[d] adequate incentive to continue” despite the denial of class certification. Coopers & Lybrand, 437 U. S., at 471. Appellate courts lack even that authority under respondents’ theory. Instead, the decision whether an immediate appeal will lie resides exclusively with the plaintiff; she need only dismiss her claims with prejudice, whereupon she may appeal the district court’s order denying class certification. And, as under the death-knell doctrine, she may exercise that option more than once, stopping and starting the district court proceedings with repeated interlocutory appeals. See id., at 474 (death-knell doctrine offered “no assurance that the trial process [would] not again be disrupted by interlocutory review”). Consider this case. The Ninth Circuit reviewed and rejected only the District Court’s application of comity as a basis for striking respondents’ class allegations. 797 F. 3d, at 615. The appeals court declined to reach Microsoft’s other arguments against class certification. See ibid. It remained open to the District Court, in the Court of Appeals’ view, to deny class certification on a differ-ent ground, and respondents would be free, under their theory, to force appellate review of any new order denying certification by again dismissing their claims. In designing Rule 23(f)’s provision for discretionary review, the Rules Committee sought to prevent such disruption and delay. See supra, at 6–8.[9] Respondents nevertheless maintain that their position promotes efficiency, observing that after dismissal with prejudice the case is over if the plaintiff loses on appeal. Brief for Respondents 38–39. Their way, they say, means prompt resolution of many lawsuits and infrequent use of the voluntary-dismissal tactic, for “most appeals lose” and few plaintiffs will “take th[e] risk” of losing their claims for good. Id., at 35–36. Respondents overlook the prospect that plaintiffs with weak merits claims may readily assume that risk, mindful that class certification often leads to a hefty settlement. See Coopers & Lybrand, 437 U. S., at 476 (defendant facing the specter of classwide liability may “abandon a meritorious defense”). Indeed, the same argument—that the case was over if the plaintiff lost on appeal—was evident in the death-knell context, yet this Court determined that the potential for piecemeal litigation was “apparent and serious.” Id., at 474.[10] And that potential is greater still under respondents’ theory, where plaintiffs alone determine whether and when to appeal an adverse certification ruling. B Another vice respondents’ theory shares with the death-knell doctrine, both allow indiscriminate appellate review of interlocutory orders. Ibid. Beyond disturbing the “appropriate relationship between the respective courts,” id., at 476 (internal quotation marks omitted), respondents’ dismissal tactic undercuts Rule 23(f)’s discretionary regime. This consideration is “[o]f prime significance to the jurisdictional issue before us.” Swint v. Chambers County Comm’n, 514 U. S. 35, 46 (1995) (pendent appellate jurisdiction in collateral-order context would undermine §1292(b)); see supra, at 3–4 (death-knell doctrine impermissibly circumvented §1292(b)). In the Rules Enabling Act, as earlier recounted, Congress authorized this Court to determine when a decision is final for purposes of §1291, and to provide for appellate review of interlocutory orders not covered by statute. See supra, at 5–6, and n. 4. These changes are to come from rulemaking, however, not judicial decisions in particular controversies or inventive litigation ploys. See Swint, 514 U. S., at 48. In this case, the rulemaking process has dealt with the matter, yielding a “measured, practical solutio[n]” to the questions whether and when adverse certification orders may be immediately appealed. Mohawk Industries, 558 U. S., at 114. Over years the Advisory Committee on the Federal Rules of Civil Procedure studied the data on class-certification rulings and appeals, weighed various proposals, received public comment, and refined the draft rule and Committee Note. See Solimine & Hines, 41 Wm. & Mary L. Rev., at 1564–1566, and nn. 178–189; Federal Judicial Center Study 80–87. Rule 23(f) reflects the rulemakers’ informed assessment, permitting, as explained supra, at 5–7, interlocutory appeals of adverse certification orders, whether sought by plaintiffs or defendants, solely in the discretion of the courts of appeals. That assessment “warrants the Judiciary’s full respect.” Swint, 514 U. S., at 48; see Mohawk Industries, 558 U. S., at 118–119 (Thomas, J., concurring in part and concurring in judgment). Here, however, the Ninth Circuit, after denying respondents permission to appeal under Rule 23(f), nevertheless assumed jurisdiction of their appeal challenging only the District Court’s order striking the class allegations. See supra, at 9–10. According to respondents, even plaintiffs who altogether bypass Rule 23(f) may force an appeal by dismissing their claims with prejudice. See Tr. of Oral Arg. 34. Rule 23(f), respondents say, is irrelevant, for it “address[es] interlocutory orders,” whereas this case involves “an actual final judgment.” Brief for Respondents 26, 28. We are not persuaded. If respondents’ voluntary-dismissal tactic could yield an appeal of right, Rule 23(f)’s careful calibration—as well as Congress’ designation of rulemaking “as the preferred means for determining whether and when prejudgment orders should be immediately appealable,” Mohawk Industries, 558 U. S., at 113 (majority opinion)—“would be severely undermined,” Swint, 514 U. S., at 47. Respondents, after all, “[sought] review of only the [inherently interlocutory] orde[r]” striking their class allegations; they “d[id] not complain of the ‘final’ orde[r] that dismissed their cas[e].” Camesi v. University of Pittsburgh Medical Center, 729 F. 3d 239, 244 (CA3 2013). Plaintiffs in putative class actions cannot transform a tentative interlocutory order, see supra, at 9, into a final judgment within the meaning of §1291 simply by dismissing their claims with prejudice—subject, no less, to the right to “revive” those claims if the denial of class certification is reversed on appeal, see Brief for Respondents 45; Tr. of Oral Arg. 31 (assertion by respondents’ counsel that, if the appeal succeeds, “everything would spring back to life” on remand). Were respondents’ reasoning embraced by this Court, “Congress[’] final decision rule would end up a pretty puny one.” Digital Equipment Corp., 511 U. S., at 872. Contrary to respondents’ argument, §1291’s firm final-judgment rule is not satisfied whenever a litigant persuades a district court to issue an order purporting to end the litigation. Finality, we have long cautioned, “is not a technical concept of temporal or physical termination.” Cobbledick, 309 U. S., at 326. It is one “means [geared to] achieving a healthy legal system,” ibid., and its contours are determined accordingly, see supra, at 12.[11] C The one-sidedness of respondents’ voluntary-dismissal device “reinforce[s] our conclusion that [it] does not support appellate jurisdiction of prejudgment orders denying class certification.” Coopers & Lybrand, 437 U. S., at 476; see supra, at 4. Respondents’ theory permits plaintiffs only, never defendants, to force an immediate appeal of an adverse certification ruling. Yet the “class issue” may be just as important to defendants, Coopers & Lybrand, 437 U. S., at 476, for “[a]n order granting certification . . . may force a defendant to settle rather than . . . run the risk of potentially ruinous liability,” Committee Note on Rule 23(f); see supra, at 4, and n. 2 (defendants may face a “reverse death knell”). Accordingly, we recognized in Coopers & Lybrand that “[w]hatever similarities or differences there are between plaintiffs and defendants in this context involve questions of policy for Congress.” 437 U. S., at 476. Congress chose the rulemaking process to settle the matter, and the rulemakers did so by adopting Rule 23(f)’s evenhanded prescription. It is not the prerogative of litigants or federal courts to disturb that settlement. See supra, at 14–15. * * * For the reasons stated, the judgment of the Court of Appeals for the Ninth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 Section 1292, which authorizes review of certain interlocutory decisions, does not include among those decisions class-action certifications. See 28 U. S. C. §1292. 2 This scenario has been called a “reverse death knell,” Sullivan & Trueblood, Rule 23(f): A Note on Law and Discretion in the Courts of Appeals, 246 F. R. D. 277, 280 (2008), or “inverse death knell,” 7B C. Wright, A. Miller, & M. Kane, Federal Practice and Procedure §1802, p. 299 (3d ed. 2005), for it too ends the litigation as a practical matter. 3 Coopers & Lybrand also rejected the collateral-order doctrine as a basis for invoking §1291 to appeal an order denying class certification. The collateral-order doctrine applies only to a “small class” of decisions that are conclusive, that resolve important issues “completely separate from the merits,” and that are “effectively unreviewable on appeal from a final judgment.” 437 U. S., at 468. An order concerning class certification, we explained, fails each of these criteria. See id., at 469. 4 Congress amended the Rules Enabling Act, 28 U. S. C. §2071 et seq., in 1990 to authorize this Court to prescribe rules “defin[ing] when a ruling of a district court is final for the purposes of appeal under section 1291.” §2072(c). Congress enacted §1292(e) two years later, and that same year the Advisory Committee on the Federal Rules of Civil Procedure began to review proposals for what would become Rule 23(f). See Solimine & Hines, Deciding To Decide: Class Action Certification and Interlocutory Review by the United States Courts of Appeals Under Rule 23(f), 41 Wm. & Mary L. Rev. 1531, 1563–1564, 1566, n. 189 (2000). 5 Rule 23(f) has changed little since its adoption in 1998. See Ad-visory Committee’s 2007 and 2009 Notes on subd. (f) of Fed. Rule Civ. Proc. 23, 28 U. S. C. App., p. 820 (deleting a redundancy and increas-ing the time to petition for permission to appeal from ten to 14 days, respectively). 6 Legislation striking this balance was also introduced in Congress. See H. R. 660, 105th Cong., 1st Sess. (1997). The bill, which would have amended §1292(b) to provide for interlocutory appeal of adverse class determinations, likewise committed the decision whether an immediate appeal would lie exclusively to the courts of appeals: “The court of appeals may, in its discretion, permit the appeal to be taken from such determination.” Ibid. Upon learning that “proposed Rule 23(f) [was] well advanced,” the bill’s sponsor, Representative Charles Canady, joined forces with the Rules Committee. See Judicial Conference of the United States, Advisory Committee on Civil Rules, Minutes of May 1–2, 1997. 7 An order striking class allegations is “functional[ly] equivalent” to an order denying class certification and therefore appealable under Rule 23(f). Scott v. Family Dollar Stores, Inc., 733 F. 3d 105, 110–111, n. 2 (CA4 2013) (quoting In re Bemis Co., 279 F. 3d 419, 421 (CA7 2002)). See also United Airlines, Inc. v. McDonald, 432 U. S. 385 , and n. 4 (1977) (equating order striking class allegations with “a denial of class certification”). 8 Compare Berger v. Home Depot USA, Inc., 741 F. 3d 1061, 1065 (CA9 2014) (assuming jurisdiction under these circumstances); Gary Plastic Packaging Corp. v. Merrill Lynch, 903 F. 2d 176, 179 (CA2 1990) (assuming jurisdiction after dismissal for failure to prosecute), with Camesi v. University of Pittsburgh Medical Center, 729 F. 3d 239, 245–247 (CA3 2013) (no jurisdiction under §1291 or Article III in this situation); Rhodes v. E. I. du Pont de Nemours & Co., 636 F. 3d 88, 100 (CA4 2011) (no jurisdiction under Article III). 9 Rule 23(f) avoids delay not only by limiting class-certification appeals to those permitted by the federal courts of appeals, but also by specifying that “[a]n appeal does not stay proceedings in the district court unless the district judge or the court of appeals so orders.” See Blair v. Equifax Check Services, Inc., 181 F. 3d 832, 835 (CA7 1999) (“Rule 23(f) is drafted to avoid delay.”). Respondents’ dismissal tactic, by contrast, halts district court proceedings whenever invoked. 10 The very premise of the death-knell doctrine was that plaintiffs “would not pursue their claims individually.” Coopers & Lybrand, 437 U. S., at 466. Having pressed such an argument for the benefit of immediate review, a death-knell plaintiff who lost on appeal would encounter the general proposition, long laid down, that “where a party assumes a certain position in a legal proceeding, and succeeds in maintaining that position, he may not thereafter, simply because his interests have changed, assume a contrary position.” Davis v. Wakelee, 156 U. S. 680, 689 (1895) . 11 Respondents also invoke our decision in United States v. Procter & Gamble Co., 356 U. S. 677 (1958) , but that case—a civil antitrust enforcement action—involved neither class-action certification nor the sort of dismissal tactic at issue here. See id., at 681 (the Government “did not consent to a judgment against [it]” (internal quotation marks omitted)).

581.US.2016_16-348

Petitioner Midland Funding filed a proof of claim in respondent Johnson’s Chapter 13 bankruptcy case, asserting that Johnson owed Midland credit-card debt and noting that the last time any charge appeared on Johnson’s account was more than 10 years ago. The relevant statute of limitations under Alabama law is six years. Johnson objected to the claim, and the Bankruptcy Court disallowed it. Johnson then sued Midland, claiming that its filing a proof of claim on an obviously time-barred debt was “false,” “deceptive,” “misleading,” “unconscionable,” and “unfair” within the meaning of the Fair Debt Collection Practices Act, 15 U. S. C. §§1692e, 1692f. The District Court held that the Act did not apply and dismissed the suit. The Eleventh Circuit reversed. Held: The filing of a proof of claim that is obviously time barred is not a false, deceptive, misleading, unfair, or unconscionable debt collection practice within the meaning of the Fair Debt Collection Practices Act. Pp. 2–10. (a) Midland’s proof of claim was not “false, deceptive, or misleading.” The Bankruptcy Code defines the term “claim” as a “right to payment,” 11 U. S. C. §101(5)(A), and state law usually determines whether a person has such a right, see Travelers Casualty & Surety Co. of America v. Pacific Gas & Elec. Co., 549 U. S. 443 –451. The relevant Alabama law provides that a creditor has the right to payment of a debt even after the limitations period has expired. Johnson argues that the word “claim” means “enforceable claim.” But the word “enforceable” does not appear in the Code’s definition, and Johnson’s interpretation is difficult to square with Congress’s intent “to adopt the broadest available definition of ‘claim,’ ” Johnson v. Home State Bank, 501 U. S. 78 . Other Code provisions are still more difficult to square with Johnson’s interpretation. For example, §502(b)(1) says that if a “claim” is “unenforceable” it will be disallowed, not that it is not a “claim.” Other provisions make clear that the running of a limitations period constitutes an affirmative defense that a debtor is to assert after the creditor makes a “claim.” §§502, 558. The law has long treated unenforceability of a claim (due to the expiration of the limitations period) as an affirmative defense, and there is nothing misleading or deceptive in the filing of a proof of claim that follows the Code’s similar system. Indeed, to determine whether a statement is misleading normally “requires consideration of the legal sophistication of its audience,” Bates v. State Bar of Ariz., 433 U. S. 350 , which in a Chapter 13 bankruptcy includes a trustee who is likely to understand that a proof of claim is a statement by the creditor that he or she has a right to payment that is subject to disallowance, including disallowance based on untimeliness. Pp. 2–5. (b) Several circumstances, taken together, lead to the conclusion that Midland’s proof of claim was not “unfair” or “unconscionable” within the terms of the Fair Debt Collection Practices Act. Johnson points out that several lower courts have found or indicated that, in the context of an ordinary civil action to collect a debt, a debt collector’s assertion of a claim known to be time barred is “unfair.” But those courts rested their conclusions upon their concern that a consumer might unwittingly repay a time-barred debt. Such considerations have significantly diminished force in a Chapter 13 bankruptcy, where the consumer initiates the proceeding, see §§301, 303(a); where a knowledgeable trustee is available, see §1302(a); where procedural rules more directly guide the evaluation of claims, see Fed. Rule Bkrtcy. Proc. 3001(c)(3)(A); and where the claims resolution process is “generally a more streamlined and less unnerving prospect for a debtor than facing a collection lawsuit,” In re Gatewood, 533 B. R. 905, 909. Also unpersuasive is Johnson’s argument that there is no legitimate reason for allowing a practice like this one that risks harm to the debtor. The bankruptcy system treats untimeliness as an affirmative defense and normally gives the trustee the burden of investigating claims to see if one is stale. And, at least on occasion, the assertion of even a stale claim can benefit the debtor. More importantly, a change in the simple affirmative-defense approach, carving out an exception, would require defining the exception’s boundaries. Does it apply only where a claim’s staleness appears on the face of the proof of claim? Does it apply to other affirmative defenses or only to the running of the limitations period? Neither the Fair Debt Collection Practices Act nor the Bankruptcy Code indicates that Congress intended an ordinary civil court applying the Act to determine answers to such bankruptcy-related questions. The Act and the Code have different purposes and structural features. The Act seeks to help consumers by preventing consumer bankruptcies in the first place, while the Code creates and maintains the “delicate balance of a debtor’s protections and obligations,” Kokoszka v. Belford, 417 U. S. 642 . Applying the Act in this context would upset that “delicate balance.” Contrary to the argument of the United States, the promulgation of Bankruptcy Rule 9011 did not resolve this issue. Pp. 5–10. 823 F. 3d 1334, reversed. Breyer, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, and Alito, JJ., joined. Sotomayor, J., filed a dissenting opinion, in which Ginsburg and Kagan, JJ., joined. Gorsuch, J., took no part in the consideration or decision of the case.

The Fair Debt Collection Practices Act, 91Stat. 874, 15 U. S. C. §1692 et seq., prohibits a debt collector from asserting any “false, deceptive, or misleading representation,” or using any “unfair or unconscionable means” to collect, or attempt to collect, a debt, §§1692e, 1692f. In this case, a debt collector filed a written statement in a Chapter 13 bankruptcy proceeding claiming that the debtor owed the debt collector money. The statement made clear, however, that the 6-year statute of limitations governing collection of the claimed debt had long since run. The question before us is whether the debt collector’s filing of that statement falls within the scope of the aforementioned provisions of the Fair Debt Collection Practices Act. We conclude that it does not. I In March 2014, Aleida Johnson, the respondent, filed for personal bankruptcy under Chapter 13 of the Bankruptcy Code (or Code), 11 U. S. C. §1301 et seq, in the Federal District Court for the Southern District of Alabama. Two months later, Midland Funding, LLC, the petitioner, filed a “proof of claim,” a written statement asserting that Johnson owed Midland a credit-card debt of $1,879.71. The statement added that the last time any charge appeared on Johnson’s account was in May 2003, more than 10 years before Johnson filed for bankruptcy. The relevant statute of limitations is six years. See Ala. Code §6–2–34 (2014). Johnson, represented by counsel, objected to the claim; Midland did not respond to the objection; and the Bankruptcy Court disallowed the claim. Subsequently, Johnson brought this lawsuit against Midland seeking actual damages, statutory damages, attorney’s fees, and costs for a violation of the Fair Debt Collection Practices Act. See 15 U. S. C. §1692k. The District Court decided that the Act did not apply and therefore dismissed the action. The Court of Appeals for the Eleventh Circuit disagreed and reversed the District Court. 823 F. 3d 1334 (2016). Midland filed a petition for certiorari, noting a division of opinion among the Courts of Appeals on the question whether the conduct at issue here is “false,” “deceptive,” “misleading,” “unconscionable,” or “unfair” within the meaning of the Act. Compare ibid. (finding the Fair Debt Collection Practices Act applicable) with In re Dubois, 834 F. 3d 522 (CA4 2016) (finding the Act inapplicable); Owens v. LVNV Funding, LLC, 832 F. 3d 726 (CA7 2016) (same); and Nelson v. MidlandCredit Management, Inc., 828 F. 3d 749 (CA8 2016) (same). We granted the petition. We now reverse the Court of Appeals. II Like the majority of Courts of Appeals that have considered the matter, we conclude that Midland’s filing of a proof of claim that on its face indicates that the limitations period has run does not fall within the scope of any of the five relevant words of the Fair Debt Collection Practices Act. We believe it reasonably clear that Midland’s proof of claim was not “false, deceptive, or misleading.” Midland’s proof of claim falls within the Bankruptcy Code’s definition of the term “claim.” A “claim” is a “right to payment.” 11 U. S. C. §101(5)(A). State law usually determines whether a person has such a right. See Travelers Casualty & Surety Co. of America v. Pacific Gas & Elec. Co., 549 U. S. 443 –451 (2007). The relevant state law is the law of Alabama. And Alabama’s law, like the law of many States, provides that a creditor has the right to payment of a debt even after the limitations period has expired. See Ex parte HealthSouth Corp., 974 S. 2d 288, 296 (Ala. 2007) (passage of time extinguishes remedy but the right remains); see also, e.g., Sallaz v. Rice, 161 Idaho 223, ___, 384 P. 3d 987, 992–993 (2016) (similar); Notte v. Merchants Mut. Ins. Co., 185 N. J. 490, 499–500, 888 A. 2d 464, 469 (2006) (similar); Potterton v. Ryland Group, Inc., 289 Md. 371, 375–376, 424 A. 2d 761, 764 (1981) (similar); Summers v. Connolly, 159 Ohio St. 396, 400–402, 112 N. E. 2d 391, 394 (1953) (similar); DeVries v. Secretary of State, 329 Mich. 68, 75, 44 N. W. 2d 872, 876 (1950) (similar); Fleming v. Yeazel, 379 Ill. 343, 344–346, 40 N. E. 2d 507, 508 (1942) (similar); Fidelity & Cas. Co. of N. Y. v. Lackland, 175 Va. 178, 185–187, 8 S. E. 2d 306, 309 (1940) (similar); Insurance Co. v. Dunscomb, 108 Tenn. 724, 728–731, 69 S. W. 345, 346 (1902) (similar); but see, e.g., Miss. Code Ann. §15–1–3(1) (2012) (expiration of the limitations period extinguishes the remedy and the right); Wis. Stat. §893.05 (2011–2012) (same). Johnson argues that the Code’s word “claim” means “enforceable claim.” She notes that this Court once referred to a bankruptcy “claim” as “an enforceable obligation.” Pennsylvania Dept. of Public Welfare v. Davenport, 495 U. S. 552, 559 (1990) . And, she concludes, Midland’s “proof of claim” was false (or deceptive or misleading) because its “claim” was not enforceable. Brief for Respondent 22; Brief for United States as Amicus Curiae 18–20 (making a similar argument). But we do not find this argument convincing. The word “enforceable” does not appear in the Code’s definition of “claim.” See 11 U. S. C. §101(5). The Court in Davenport likely used the word “enforceable” descriptively, for that case involved an enforceable debt. 495 U. S., at 559. And it is difficult to square Johnson’s interpretation with our later statement that “Congress intended . . . to adopt the broadest available definition of ‘claim.’ ” Johnson v. Home State Bank, 501 U. S. 78, 83 (1991) . It is still more difficult to square Johnson’s interpretation with other provisions of the Bankruptcy Code. Section 502(b)(1) of the Code, for example, says that, if a “claim” is “unenforceable,” it will be disallowed. It does not say that an “unenforceable” claim is not a “claim.” Similarly, §101(5)(A) says that a “claim” is a “right to payment,” “whether or not such right is . . . fixed, contingent, . . . [or] disputed.” If a contingency does not arise, or if a claimant loses a dispute, then the claim is unenforce-able. Yet this section makes clear that the unenforceable claim is nonetheless a “right to payment,” hence a “claim,” as the Code uses those terms. Johnson looks for support to other provisions that govern bankruptcy proceedings, including §502(a) of the Bankruptcy Code, which states that a claim will be allowed in the absence of an objection, and Rule 3001(f) of the Federal Rules of Bankruptcy Procedure, which states that a properly filed “proof of claim . . . shall constitute prima facie evidence of the validity and amount of the claim.” But these provisions do not discuss the scope of the term “claim.” Rather, they restate the Bankruptcy Code’s system for determining whether a claim will be allowed. Other provisions make clear that the running of a limitations period constitutes an affirmative defense, a defense that the debtor is to assert after a creditor makes a “claim.” §§502, 558. The law has long treated unenforceability of a claim (due to the expiration of the limitations period) as an affirmative defense. See, e.g., Fed. Rule Civ. Proc. 8(c)(1); 13 Encyclopaedia of Pleading and Practice 200 (W. McKinney ed. 1898). And we see nothing misleading or deceptive in the filing of a proof of claim that, in effect, follows the Code’s similar system. Indeed, to determine whether a statement is misleading normally “requires consideration of the legal sophistication of its audience.” Bates v. State Bar of Ariz., 433 U. S. 350 , n. 37 (1977). The audience in Chapter 13 bankruptcy cases includes a trustee, 11 U. S. C. §1302(a), who must examine proofs of claim and, where appropriate, pose an objection, §§704(a)(5), 1302(b)(1) (including any timeliness objection, §§502(b)(1), 558). And that trustee is likely to understand that, as the Code says, a proof of claim is a statement by the creditor that he or she has a right to payment subject to disallowance (includingdisallowance based upon, and following, the trustee’s objection for untimeliness). §§101(5)(A), 502(b), 704(a)(5), 1302(b)(1). (We do not address the appropriate standard in ordinary civil litigation.) III Whether Midland’s assertion of an obviously time-barred claim is “unfair” or “unconscionable” (within the terms of the Fair Debt Collection Practices Act) presents a closer question. First, Johnson points out that several lower courts have found or indicated that, in the context of an ordinary civil action to collect a debt, a debt collector’s assertion of a claim known to be time barred is “unfair.” See, e.g., Phillips v. Asset Acceptance, LLC, 736 F. 3d 1076, 1079 (CA7 2013) (holding as much); Kimber v. Federal Financial Corp., 668 F. Supp. 1480, 1487 (MD Ala. 1987) (same); Huertas v. Galaxy Asset Management, 641 F. 3d 28, 32–33 (CA3 2011) (indicating as much); Castro v. Collecto, Inc., 634 F. 3d 779, 783 (CA5 2011) (same); Freyermuth v. Credit Bureau Servs., Inc., 248 F. 3d 767, 771 (CA8 2001) (same). We are not convinced, however, by this precedent. It considers a debt collector’s assertion in a civil suit of a claim known to be stale. We assume, for argument’s sake, that the precedent is correct in that context (a matter this Court itself has not decided and does not now decide). But the context of a civil suit differs significantly from the present context, that of a Chapter 13 bankruptcy proceeding. The lower courts rested their conclusions upon their concern that a consumer might unwittingly repay a time-barred debt. Thus the Seventh Circuit pointed out that “ ‘few unsophisticated consumers would be aware that a statute of limitations could be used to defend against lawsuits based on stale debts.’ ” Phillips, supra, at 1079 (quoting Kimber, supra, at 1487). The “ ‘passage of time,’ ” the Circuit wrote, “ ‘dulls the consumer’s memory of the circumstances and validity of the debt’ ” and the consumer may no longer have “ ‘personal records.’ ” 736 F. 3d, at 1079 (quoting Kimber, supra, at 1487). Moreover, a consumer might pay a stale debt simply to avoid the cost and embarrassment of suit. 736 F. 3d, at 1079. These considerations have significantly diminished force in the context of a Chapter 13 bankruptcy. The consumer initiates such a proceeding, see 11 U. S. C. §§301, 303(a), and consequently the consumer is not likely to pay a stale claim just to avoid going to court. A knowledgeable trustee is available. See §1302(a). Procedural bankruptcyrules more directly guide the evaluation of claims. See Fed. Rule Bkrtcy. Proc. 3001(c)(3)(A); Advisory Committee’s Notes on Rule 3001–2011 Amdt., 11 U. S. C. App., p. 678. And, as the Eighth Circuit Bankruptcy Appellate Panel put it, the claims resolution process is “generally a more streamlined and less unnerving prospect for a debtor than facing a collection lawsuit.” In re Gatewood, 533 B. R. 905, 909 (2015); see also, e.g., 11 U. S. C. §502 (outlining generally the claims resolution process). These features of a Chapter 13 bankruptcy proceeding make it considerably more likely that an effort to collect upon a stale claim in bankruptcy will be met with resistance, objection, and disallowance. Second, Johnson argues that the practice at least risks harm to the debtor and that there is not “a single legitimate reason” for allowing this kind of behavior. Brief for Respondent 32. Would it not be obviously “unfair,” she asks, for a debt collector to adopt a practice of buying up stale claims cheaply and asserting them in bankruptcy knowing they are stale and hoping for careless trustees? The United States, supporting Johnson, adds its view that the Federal Rules of Bankruptcy Procedure make the practice open to sanction, and argues that sanctionable conduct is unfair conduct. Brief for United States as Amicus Curiae 20. See Fed. Rule Bkrtcy. Proc. 9011(b)(2) (sanction possible if party violates the Rule that by “presenting to the [bankruptcy] court” any “paper,” a “party is certifying that to the best of” his or her “knowledge, . . . the claims . . . therein are warranted by existing law”). We are ultimately not persuaded by these arguments. The bankruptcy system, as we have already noted, treats untimeliness as an affirmative defense. The trustee normally bears the burden of investigating claims and pointing out that a claim is stale. See supra, at 4–5. Moreover, protections available in a Chapter 13 bankruptcy proceeding minimize the risk to the debtor. See supra, at 6. And, at least on occasion, the assertion of even a stale claim can benefit a debtor. Its filing and disallowance “discharge[s]” the debt. 11 U. S. C. §1328(a). And that discharge means that the debt (even if unenforceable) will not remain on a credit report potentially affecting an individual’s ability to borrow money, buy a home, and perhaps secure employment. See 15 U. S. C. §1681c(a)(4) (debt may remain on a credit report for seven years); cf. Ala. Code §6–2–34 (6-year statute of limitations); Md. Cts. & Jud. Proc. Code Ann. §5–101 (2013) (3-year statute of limitations); cf. 16 CFR pt. 600, App. §607, ¶6 (1991) (a credit report may include discharged debt only if “the debt [is reported] as having a zero balance due to reflect the fact that the consumer is no longer liable for the discharged debt”); FTC, 40 Years of Experience with the Fair Credit Reporting Act: An FTC Staff Report with Summary of Interpretations 66 (2011) (similar). More importantly, a change in the simple affirmative-defense approach, carving out an exception, itself would require defining the boundaries of the exception. Does it apply only where (as Johnson alleged in the complaint) a claim’s staleness appears “on [the] face” of the proof of claim? Does it apply to other affirmative defenses or only to the running of a limitations period? At the same time, we do not find in either the Fair Debt Collection Practices Act or the Bankruptcy Code good reason to believe that Congress intended an ordinary civil court applying the Act to determine answers to these bankruptcy-related questions. The Act and the Code have different purposes and structural features. The Act seeks to help consumers, not necessarily by closing what Johnson and the United States characterize as a loophole inthe Bankruptcy Code, but by preventing consumer bankruptcies in the first place. See, e.g., 15 U. S. C. §1692(a) (recognizing the “abundant evidence of the use of abusive, deceptive, and unfair debt collection practices [which] contribute to the number of personal bankruptcies”); see also §1692(b) (“Existing laws and procedures . . . are inadequate to protect consumers”); §1692(e) (statute seeks to “eliminate abusive debt collection practices”). The Bankruptcy Code, by way of contrast, creates and maintains what we have called the “delicate balance of a debtor’s protections and obligations.” Kokoszka v. Belford, 417 U. S. 642, 651 (1974) . To find the Fair Debt Collection Practices Act applicable here would upset that “delicate balance.” From a sub-stantive perspective it would authorize a new significant bankruptcy-related remedy in the absence of language in the Code providing for it. Administratively, it would permit postbankruptcy litigation in an ordinary civil court concerning a creditor’s state of mind—a matter often hard to determine. See 15 U. S. C. §1692k(c) (safe harbor for any debt collector who “shows by a preponderance of evidence that the violation was not intentional and resulted from a bona fide error notwithstanding the maintenance of procedures reasonably adapted to avoid any such error”). Procedurally, it would require creditors (who assert a claim) to investigate the merits of an affirmative defense (typically the debtor’s job to assert and prove) lest the creditor later be found to have known the claim was untimely. The upshot could well be added complexity, changes in settlement incentives, and a shift from the debtor to the creditor the obligation to investigate the staleness of a claim. Unlike the United States, we do not believe that the Advisory Committee on Rules of Bankruptcy Procedure settled the issue when it promulgated Bankruptcy Rule 9011. The Committee, in considering amendments to the Federal Rules of Bankruptcy Procedure in 2009, specifically rejected a proposal that would have required a creditor to certify that there is no valid statute of limitations defense. See Agenda Book for Meeting 86–87 (Mar. 26–27, 2009). It did so in part because the working group did not want to impose an affirmative obligation on a creditor to make a prefiling investigation of a potential time-bar defense. Ibid. In rejecting that proposal, the Committee did note that Rule 9011 imposes a general “obligation on a claimant to undertake an inquiry reasonable under the circumstances to determine . . . that a claim is warranted by existing law and that factual contentions have evidentiary support,” and to certify as much on the proof of claim. Id., at 87. The Committee also acknowledged, however, that this requirement would “not addres[s] the statute of limitation issue,” but would only ensure “the accuracy of the information provided.” Ibid. We recognize that one Bankruptcy Court has held that filing a time-barred claim without a prefiling investigation of a potential time-bar defense merits sanctions under Rule 9011. In re Sekema, 523 B. R. 651, 654 (Bkrtcy. Ct. ND Ind. 2015). But others have held to the contrary. See, e.g., In re Freeman, 540 B. R. 129, 143–144 (Bkrtcy. Ct. ED Pa. 2015); In re Jenkins, 538 B. R. 129, 134–136 (Bkrcty. Ct. ND Ala. 2015); In re Keeler, 440 B. R. 354, 366–369 (Bkrtcy. Ct. ED Pa. 2009); see also In re Andrews, 394 B. R. 384, 387–388 (Bkrtcy. Ct. EDNC 2008) (recognizing that “[m]any courts have . . . found that sanctions [under Rule 9011] were not warranted for filing stale claims”). These circumstances, taken together, convince us that we cannot find the practice at issue here “unfair” or “unconscionable” within the terms of the Fair Debt Collection Practices Act. IV For these reasons, we conclude that filing (in a Chapter 13 bankruptcy proceeding) a proof of claim that is obviously time barred is not a false, deceptive, misleading, unfair,or unconscionable debt collection practice within the meaning of the Fair Debt Collection Practices Act. The judgment of the Eleventh Circuit is reversed. It is so ordered. JUSTICE GORSUCH took no part in the consideration or decision of this case.

581.US.2016_15-797

Petitioner Moore was convicted of capital murder and sentenced to death for fatally shooting a store clerk during a botched robbery that occurred when Moore was 20 years old. A state habeas court subsequently determined that, under Atkins v. Virginia, 536 U. S. 304 , and Hall v. Florida, 572 U. S. ___, Moore qualified as intellectually disabled and that his death sentence therefore violated the Eighth Amendment’s proscription of “cruel and unusual punishments.” The court consulted current medical diagnostic standards—the 11th edition of the American Association on Intellectual and Developmental Disabilities clinical manual (AAIDD–11) and the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders published by the American Psychiatric Association. The habeas court followed the generally accepted intellectual-disability definition, which identifies three core elements: (1) intellectual-functioning deficits, (2) adaptive deficits, and (3) the onset of these deficits while still a minor. Moore’s IQ scores, the court determined, established subaverage intellectual functioning. The court credited six scores, the average of which (70.66) indicated mild intellectual disability. And relying on testimony from mental-health professionals, the court found significant adaptive deficits in all three skill sets (conceptual, social, and practical). Based on its findings, the habeas court recommended to the Texas Court of Criminal Appeals (CCA) that Moore be granted relief. The CCA declined to adopt the judgment recommended by the habeas court. The CCA held instead that the habeas court erred by not following the CCA’s 2004 decision in Ex parte Briseno, 135 S. W. 3d 1, which adopted the definition of, and standards for assessing, intellectual disability contained in the 1992 (ninth) edition of the American Association on Mental Retardation manual (AAMR–9), predecessor to the current AAIDD–11 manual. Briseno also incorporated the AAMR–9’s requirement that adaptive deficits must be “related” to intellectual-functioning deficits, and it recited, without citation to any medical or judicial authority, seven evidentiary factors relevant to the intellectual-disability inquiry. Based on only two of Moore’s IQ scores (of 74 and 78), the CCA concluded that Moore had not shown significantly subaverage intellectual functioning. And even if he had, the CCA continued, his adaptive strengths undercut any adaptive weaknesses. The habeas court also failed, the CCA determined, to inquire into relatedness. Among alternative causes for Moore’s adaptive deficits, the CCA suggested, were an abuse-filled childhood, undiagnosed learning disorders, multiple elementary-school transfers, racially motivated harassment and violence at school, and a history of academic failure, drug abuse, and absenteeism. Briseno’s seven evidentiary factors, the CCA further determined, weighed against finding that Moore had satisfied the relatedness requirement. Held: By rejecting the habeas court’s application of medical guidance and by following the Briseno standard, including the nonclinical Briseno factors, the CCA’s decision does not comport with the Eighth Amendment and this Court’s precedents. Pp. 9–18. (a) The Eighth Amendment, which “ ‘reaffirms the duty of the government to respect the dignity of all persons,’ ” Hall, 572 U. S., at ___, prohibits the execution of any intellectually disabled individual, Atkins, 536 U. S., at 321. While Atkins and Hall left to the States “the task of developing appropriate ways to enforce” the restriction on executing the intellectually disabled, Hall, 572 U. S., at ___ (internal quotation marks omitted), States’ discretion is not “unfettered,” id., at ___, and must be “informed by the medical community’s diagnostic framework,” id., at ___–___. Relying on the most recent (and still current) versions of the leading diagnostic manuals, the Court concluded in Hall that Florida had “disregard[ed] established medical practice,” id., at ___, and had parted ways with practices and trends in other States, id., at ___–___. Hall indicated that being informed by the medical community does not demand adherence to everything stated in the latest medical guide. But neither does precedent license disregard of current medical standards. Pp. 9–10. (b) The CCA’s conclusion that Moore’s IQ scores established that he is not intellectually disabled is irreconcilable with Hall, which instructs that, where an IQ score is close to, but above, 70, courts must account for the test’s “standard error of measurement.” See 572 U. S., at ___–___, ___–___. Because the lower range of Moore’s adjusted IQ score of 74 falls at or below 70, the CCA had to move on to consider Moore’s adaptive functioning. Pp. 10–12. (c) The CCA’s consideration of Moore’s adaptive functioning also deviated from prevailing clinical standards and from the older clinical standards the CCA deemed applicable. Pp. 12–16. (1) The CCA overemphasized Moore’s perceived adaptive strengths—living on the streets, mowing lawns, and playing pool for money—when the medical community focuses the adaptive-functioning inquiry on adaptive deficits. The CCA also stressed Moore’s improved behavior in prison, but clinicians caution against reliance on adaptive strengths developed in controlled settings. Pp. 12–13. (2) The CCA further concluded that Moore’s record of academic failure, along with a history of childhood abuse and suffering, detracted from a determination that his intellectual and adaptive deficits were related. The medical community, however, counts traumatic experiences as risk factors for intellectual disability. The CCA also departed from clinical practice by requiring Moore to show that his adaptive deficits were not related to “a personality disorder.” Mental-health professionals recognize that intellectually disabled people may have other co-existing mental or physical impairments, including, e.g., attention-deficit/hyperactivity disorder, depressive and bipolar disorders, and autism. Pp. 13–14. (3) The CCA’s attachment to the seven Briseno evidentiary factors further impeded its assessment of Moore’s adaptive functioning. By design and in operation, the lay perceptions advanced by Briseno “creat[e] an unacceptable risk that persons with intellectual disability will be executed.” Hall, 572 U. S., at ___. The medical profession has endeavored to counter lay stereotypes, and the Briseno factors are an outlier, in comparison both to other States’ handling of intellectual-disability pleas and to Texas’ own practices in contexts other than the death penalty. Pp. 14–16. (d) States have some flexibility, but not “unfettered discretion,” in enforcing Atkins’ holding, Hall, 572 U. S., at ___, and the medical community’s current standards, reflecting improved understanding over time, constrain States’ leeway in this area. Here, the habeas court applied current medical standards in reaching its conclusion, but the CCA adhered to the standard it laid out in Briseno, including the nonclinical Briseno factors. The CCA therefore failed adequately to inform itself of the “medical community’s diagnostic framework,” Hall, 572 U. S., at ___–___. Because Briseno pervasively infected the CCA’s analysis, the decision of that court cannot stand. Pp. 17–18. 470 S. W. 3d 481, vacated and remanded. Ginsburg, J., delivered the opinion of the Court, in which Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Thomas and Alito, JJ., joined.

Bobby James Moore fatally shot a store clerk during a botched robbery. He was convicted of capital murder and sentenced to death. Moore challenged his death sentence on the ground that he was intellectually disabled and therefore exempt from execution. A state habeas court made detailed factfindings and determined that, under this Court’s decisions in Atkins v. Virginia, 536 U. S. 304 (2002) , and Hall v. Florida, 572 U. S. ___ (2014), Moore qualified as intellectually disabled. For that reason, the court concluded, Moore’s death sentence violated the Eighth Amendment’s proscription of “cruel and unusual punishments.” The habeas court therefore recommended that Moore be granted relief. The Texas Court of Criminal Appeals (CCA)[1] declined to adopt the judgment recommended by the state habeas court.[2] In the CCA’s view, the habeas court erroneously employed intellectual-disability guides currently used in the medical community rather than the 1992 guides adopted by the CCA in Ex parte Briseno, 135 S. W. 3d 1 (2004). See Ex parte Moore, 470 S. W. 3d 481, 486–487 (2015). The appeals court further determined that the evidentiary factors announced in Briseno “weigh[ed] heavily” against upsetting Moore’s death sentence. 470 S. W. 3d, at 526. We vacate the CCA’s judgment. As we instructed in Hall, adjudications of intellectual disability should be “informed by the views of medical experts.” 572 U. S., at ___ (slip op., at 19); see id., at ___ (slip op., at 7). That instruction cannot sensibly be read to give courts leave to diminish the force of the medical community’s consensus. Moreover, the several factors Briseno set out as indicators of intellectual disability are an invention of the CCA untied to any acknowledged source. Not aligned with the medical community’s information, and drawing no strength from our precedent, the Briseno factors “creat[e] an unacceptable risk that persons with intellectual dis-ability will be executed,” 572 U. S., at ___ (slip op., at 1). Accordingly, they may not be used, as the CCA used them, to restrict qualification of an individual as intellectually disabled. I In April 1980, then-20-year-old Bobby James Moore and two others were engaged in robbing a grocery store. Ex parte Moore, 470 S. W. 3d 481, 490–491 (Tex. Crim. App. 2015); App. 58. During the episode, Moore fatally shot a store clerk. 470 S. W. 3d, at 490. Some two months later, Moore was convicted and sentenced to death. See id., at 492. A federal habeas court later vacated that sentence based on ineffective assistance of trial counsel, see Moore v. Collins, 1995 U. S. Dist. LEXIS 22859, *35 (SD Tex., Sept. 29, 1995), and the Fifth Circuit affirmed, see Moore v. Johnson, 194 F. 3d 586, 622 (1999). Moore was resentenced to death in 2001, and the CCA affirmed on direct appeal. See Moore v. State, 2004 WL 231323, *1 (Jan. 14, 2004), cert. denied, 543 U. S. 931 (2004) . Moore subsequently sought state habeas relief. In 2014, the state habeas court conducted a two-day hearing on whether Moore was intellectually disabled. See Ex parte Moore, No. 314483–C (185th Jud. Dist., Harris Cty., Tex., Feb. 6, 2015), App. to Pet. for Cert. 129a. The court received affidavits and heard testimony from Moore’s family members, former counsel, and a number of court-appointed mental-health experts. The evidence revealed that Moore had significant mental and social difficulties beginning at an early age. At 13, Moore lacked basic understanding of the days of the week, the months of the year, and the seasons; he could scarcely tell time or comprehend the standards of measure or the basic principle that subtraction is the reverse of addition. Id., at 187a. At school, because of his limited ability to read and write, Moore could not keep up with lessons. Id., at 146a, 182a–183a. Often, he was separated from the rest of the class and told to draw pictures. Ibid. Moore’s father, teachers, and peers called him “stupid” for his slow reading and speech. Id., at 146a, 183a. After failing every subject in the ninth grade, Moore dropped out of high school. Id., at 188a. Cast out of his home, he survived on the streets, eating from trash cans, even after two bouts of food poisoning. Id., at 192a–193a. In evaluating Moore’s assertion of intellectual disability, the state habeas court consulted current medical diagnostic standards, relying on the 11th edition of the American Association on Intellectual and Developmental Disabilities (AAIDD) clinical manual, see AAIDD, Intellectual Disability: Definition, Classification, and Systems of Supports (2010) (hereinafter AAIDD–11), and on the 5th edition of the Diagnostic and Statistical Manual of Mental Disorders published by the American Psychiatric Association (APA), see APA, Diagnostic and Statistical Manual of Mental Disorders (2013) (hereinafter DSM–5). App. to Pet. for Cert. 150a–151a, 202a. The court followed the generally accepted, uncontroversial intellectual-disability diagnos-tic definition, which identifies three core elements: (1) intellectual-functioning deficits (indicated by an IQ score “approximately two standard deviations below the mean”—i.e., a score of roughly 70—adjusted for “the standard error of measurement,” AAIDD–11, at 27); (2) adaptive deficits (“the inability to learn basic skills and adjust behavior to changing circumstances,” Hall v. Flor-ida, 572 U. S. ___, ___ (2014) (slip op., at 8)); and (3) the onset of these deficits while still a minor. See App. to Pet. for Cert. 150a (citing AAIDD–11, at 1). See also Hall, 572 U. S., at ___ (slip op., at 8).[3] Moore’s IQ scores, the habeas court determined, established subaverage intellectual functioning. The court credited six of Moore’s IQ scores, the average of which (70.66) indicated mild intellectual disability. App. to Pet. for Cert. 167a–170a.[4] And relying on testimony from several mental-health experts, the habeas court found significant adaptive deficits. In determining the significance of adaptive deficits, clinicians look to whether an individual’s adaptive performance falls two or more standard deviations below the mean in any of the three adaptive skill sets (conceptual, social, and practical). See AAIDD–11, at 43. Moore’s performance fell roughly two standard deviations below the mean in all three skill categories. App. to Pet. for Cert. 200a–201a. Based on this evidence, the state habeas court recommended that the CCA reduce Moore’s sentence to life in prison or grant him a new trial on intellectual disability. See id., at 203a. The CCA rejected the habeas court’s recommendations and denied Moore habeas relief. See 470 S. W. 3d 481. At the outset of its opinion, the CCA reaffirmed Ex parte Briseno, 135 S. W. 3d 1 (Tex. Crim. App. 2004), as paramount precedent on intellectual disability in Texas capital cases. See 470 S. W. 3d, at 486–487. Briseno adopted the definition of, and standards for assessing, intellectual disability contained in the 1992 (ninth) edition of the American Association on Mental Retardation (AAMR) manual, predecessor to the current AAIDD–11 manual. See 135 S. W. 3d, at 7 (citing AAMR, Mental Retardation: Definition, Classification, and Systems of Supports (9th ed. 1992) (hereinafter AAMR–9)). Briseno incorporated the AAMR–9’s requirement that adaptive deficits be “related” to intellectual-functioning deficits. 135 S. W. 3d, at 7 (quoting AAMR–9, at 25).[5] To determine whether a defendant has satisfied the relatedness requirement, the CCA instructed in this case, Texas courts should attend to the “seven evidentiary factors” first set out in Briseno. 470 S. W. 3d, at 489.[6] No citation to any authority, medical or judicial, accompanied the Briseno court’s recitation of the seven factors. See 135 S. W. 3d, at 8–9. The habeas judge erred, the CCA held, by “us[ing] the most current position, as espoused by AAIDD, regarding the diagnosis of intellectual disability rather than the test . . . in Briseno.” 470 S. W. 3d, at 486. This Court’s decision in Atkins v. Virginia, 536 U. S. 304 (2002) , the CCA emphasized, “left it to the States to develop appropriate ways to enforce the constitutional restriction” on the execution of the intellectually disabled. 470 S. W. 3d, at 486. Thus, even though “[i]t may be true that the AAIDD’s and APA’s positions regarding the diagnosis of intellectual disability have changed since Atkins and Briseno,” the CCA retained Briseno’s instructions, both because of “the subjectivity surrounding the medical diagnosis of intellectual disability” and because the Texas Legislature had not displaced Briseno with any other guideposts. 470 S. W. 3d, at 486–487. The Briseno inquiries, the court said, “remai[n] adequately ‘informed by the medical community’s diagnostic framework.’ ” 470 S. W. 3d, at 487 (quoting Hall, 572 U. S., at ___ (slip op., at 19–20)). Employing Briseno, the CCA first determined that Moore had failed to prove significantly subaverage intellectual functioning. 470 S. W. 3d, at 514–519. Rejecting as unreliable five of the seven IQ tests the habeas court had considered, the CCA limited its appraisal to Moore’s scores of 78 in 1973 and 74 in 1989. Id., at 518–519. The court then discounted the lower end of the standard-error range associated with those scores. Id., at 519; see infra, at 10–11 (describing standard error of measurement). Regarding the score of 74, the court observed that Moore’s history of academic failure, and the fact that he took the test while “exhibit[ing] withdrawn and depressive behavior” on death row, might have hindered his performance. 470 S. W. 3d, at 519. Based on the two scores, but not on the lower portion of their ranges, the court concluded that Moore’s scores ranked “above the intellectually disabled range” (i.e., above 70). Ibid.; see id., at 513. “Even if [Moore] had proven that he suffers from significantly sub-average general intellectual functioning,” the court continued, he failed to prove “significant and related limitations in adaptive functioning.” Id., at 520. True, the court acknowledged, Moore’s and the State’s experts agreed that Moore’s adaptive-functioning test scores fell more than two standard deviations below the mean. Id., at 521; see supra, at 4. But the State’s expert ultimately discounted those test results because Moore had “no exposure” to certain tasks the testing included, “such as writing a check and using a microwave oven.” 470 S. W. 3d, at 521–522. Instead, the expert emphasized Moore’s adaptive strengths in school, at trial, and in prison. Id., at 522–524. The CCA credited the state expert’s appraisal. Id., at 524. The habeas court, the CCA concluded, had erred by concentrating on Moore’s adaptive weaknesses. Id., at 489. Moore had demonstrated adaptive strengths, the CCA spelled out, by living on the streets, playing pool and mowing lawns for money, committing the crime in a sophisticated way and then fleeing, testifying and representing himself at trial, and developing skills in prison. Id., at 522–523. Those strengths, the court reasoned, undercut the significance of Moore’s adaptive limitations. Id., at 524–525. The habeas court had further erred, the CCA determined, by failing to consider whether any of Moore’s adaptive deficits were related to causes other than his intellectual-functioning deficits. Id., at 488, 526. Among alterna-tive causes for Moore’s adaptive deficits, the CCA suggested, were an abuse-filled childhood, undiagnosed learning disorders, multiple elementary-school transfers, racially motivated harassment and violence at school, and a his-tory of academic failure, drug abuse, and absenteeism. Ibid. Moore’s significant improvement in prison, in the CCA’s view, confirmed that his academic and social difficulties were not related to intellectual-functioning deficits. Ibid. The court then examined each of the seven Briseno evidentiary factors, see supra, at 5–6, and n. 6, concluding that those factors “weigh[ed] heavily” against finding that Moore had satisfied the relatedness requirement. 470 S. W. 3d, at 526–527. Judge Alcala dissented. Atkins and Hall, she would have held, require courts to consult current medical standards to determine intellectual disability. 470 S. W. 3d, at 530. She criticized the majority for relying on manuals superseded in the medical community, id., at 530–534, 536–539, and for disregarding the habeas court’s credibility determinations, id., at 535–536, 538–539. Judge Alcala questioned the legitimacy of the seven Briseno factors, recounting wide criticism of the factors and explaining how they deviate from the current medical consensus. See 470 S. W. 3d, at 529–530, and n. 5. Most emphatically, she urged, the CCA “must consult the medical community’s current views and standards in determining whether a defendant is intellectually disabled”; “reliance on . . . standard[s] no longer employed by the medical community,” she objected, “is constitutionally unaccept-able.” Id., at 533. We granted certiorari to determine whether the CCA’s adherence to superseded medical standards and its reliance on Briseno comply with the Eighth Amendment and this Court’s precedents. 578 U. S. ___ (2016). II The Eighth Amendment prohibits “cruel and unusual punishments,” and “reaffirms the duty of the government to respect the dignity of all persons,” Hall, 572 U. S., at ___ (slip op., at 5) (quoting Roper v. Simmons, 543 U. S. 551, 560 (2005) ). “To enforce the Constitution’s protection of human dignity,” we “loo[k] to the evolving standards of decency that mark the progress of a maturing society,” recognizing that “[t]he Eighth Amendment is not fastened to the obsolete.” Hall, 572 U. S., at ___ (slip op., at 5) (internal quotation marks omitted). In Atkins v. Virginia, we held that the Constitution “restrict[s] . . . the State’s power to take the life of” any intellectually disabled individual. 536 U. S., at 321. See also Hall, 572 U. S., at ___ (slip op., at 6); Roper, 543 U. S., at 563–564. Executing intellectually disabled individuals, we concluded in Atkins, serves no penological purpose, see 536 U. S., at 318–320; runs up against a national consensus against the practice, see id., at 313–317; and creates a “risk that the death penalty will be imposed in spite of factors which may call for a less severe penalty,” id., at 320 (internal quotation marks omitted); see id., at 320–321. In Hall v. Florida, we held that a State cannot refuse to entertain other evidence of intellectual disability when a defendant has an IQ score above 70. 572 U. S., at ___–___ (slip op., at 21–22). Although Atkins and Hall left to the States “the task of developing appropriate ways to enforce” the restriction on executing the intellectually disabled, 572 U. S., at ___ (slip op., at 17) (quoting Atkins, 536 U. S., at 317), States’ discretion, we cautioned, is not “unfettered,” 572 U. S., at ___ (slip op., at 17). Even if “the views of medical experts” do not “dictate” a court’s intellectual-disability determination, id., at ___ (slip op., at 19), we clarified, the determination must be “informed by the medical community’s diagnostic framework,” id., at ___–___ (slip op., at 19–20). We relied on the most recent (and still current) versions of the leading diagnostic manuals—the DSM–5 and AAIDD–11. Id., at ___, ___, ___–___, ___–___ (slip op., at 3, 8, 10–11, 20–21). Florida, we concluded, had violated the Eighth Amendment by “disregard[ing] established medical practice.” Id., at ___ (slip op., at 10). We further noted that Florida had parted ways with practices and trends in other States. Id., at ___–___ (slip op., at 12–16). Hall indicated that being informed by the medical community does not demand adherence to everything stated in the latest medical guide. But neither does our precedent license disregard of current medical standards. III The CCA’s conclusion that Moore’s IQ scores established that he is not intellectually disabled is irreconcilable with Hall. Hall instructs that, where an IQ score is close to, but above, 70, courts must account for the test’s “standard error of measurement.” See id., at ___–___, ___–___ (slip op., at 10–11, 21–22). See also Brumfield v. Cain, 576 U. S. ___, ___ (2015) (slip op., at 10) (relying on Hall to find unreasonable a state court’s conclusion that a score of 75 precluded an intellectual-disability finding). As we explained in Hall, the standard error of measurement is “a statistical fact, a reflection of the inherent imprecision of the test itself.” 572 U. S., at ___ (slip op., at 10). “For purposes of most IQ tests,” this imprecision in the testing instrument “means that an individual’s score is best understood as a range of scores on either side of the recorded score . . . within which one may say an individual’s true IQ score lies.” Id., at ___ (slip op., at 11). A test’s standard error of measurement “reflects the reality that an individual’s intellectual functioning cannot be reduced to a single numerical score.” Ibid. See also id., at ___–___ (slip op., at 10–12); DSM–5, at 37; AAIDD, User’s Guide: Intellectual Disability: Definition, Classification, and Systems of Supports 22–23 (11th ed. 2012) (hereinafter AAIDD–11 User’s Guide). Moore’s score of 74, adjusted for the standard error of measurement, yields a range of 69 to 79, see 470 S. W. 3d, at 519, as the State’s retained expert acknowledged, see Brief for Petitioner 39, n. 18; App. 185, 189–190. Because the lower end of Moore’s score range falls at or below 70, the CCA had to move on to consider Moore’s adaptive functioning. See Hall, 572 U. S., at ___–___ (slip op., at 21–22); 470 S. W. 3d, at 536 (Alcala, J., dissenting) (even if the majority correctly limited the scores it would consider, “current medical standards . . . would still require [the CCA] to examine whether [Moore] has adaptive deficits”). Both Texas and the dissent maintain that the CCA properly considered factors unique to Moore in disregarding the lower end of the standard-error range. Post, at 14–15; Brief for Respondent 41–42; see supra, at 6–7; 470 S. W. 3d, at 519. But the presence of other sources of imprecision in administering the test to a particular individual, see post, at 14–16, and n. 3, cannot narrow the test-specific standard-error range.[7] In requiring the CCA to move on to consider Moore’s adaptive functioning in light of his IQ evidence, we do not suggest that “the Eighth Amendment turns on the slightest numerical difference in IQ score,” post, at 15–16. Hall invalidated Florida’s strict IQ cutoff because the cutoff took “an IQ score as final and conclusive evidence of a defendant’s intellectual capacity, when experts in the field would consider other evidence.” 572 U. S., at ___ (slip op., at 10). Here, by contrast, we do not end the intellectual-disability inquiry, one way or the other, based on Moore’s IQ score. Rather, in line with Hall, we require that courts continue the inquiry and consider other evidence of intellectual disability where an individual’s IQ score, adjusted for the test’s standard error, falls within the clinically established range for intellectual-functioning deficits. IV The CCA’s consideration of Moore’s adaptive functioning also deviated from prevailing clinical standards and from the older clinical standards the court claimed to apply. A In concluding that Moore did not suffer significant adaptive deficits, the CCA overemphasized Moore’s perceived adaptive strengths. The CCA recited the strengths it perceived, among them, Moore lived on the streets, mowed lawns, and played pool for money. See 470 S. W. 3d, at 522–523, 526–527. Moore’s adaptive strengths, in the CCA’s view, constituted evidence adequate to overcome the considerable objective evidence of Moore’s adaptive deficits, see supra, at 4; App. to Pet. for Cert. 180a–202a. See 470 S. W. 3d, at 522–524, 526–527. But the medical community focuses the adaptive-functioning inquiry on adaptive deficits. E.g., AAIDD–11, at 47 (“significant limitations in conceptual, social, or practical adaptive skills [are] not outweighed by the potential strengths in some adaptive skills”); DSM–5, at 33, 38 (inquiry should focus on “[d]eficits in adaptive functioning”; deficits in only one of the three adaptive-skills domains suffice to show adaptive deficits); see Brumfield, 576 U. S., at ___ (slip op., at 15) (“[I]ntellectually disabled persons may have ‘strengths in social or physical capabilities, strengths in some adaptive skill areas, or strengths in one aspect of an adaptive skill in which they otherwise show an overall limitation.’ ” (quoting AAMR, Mental Retardation: Definition, Classification, and Systems of Supports 8 (10th ed. 2002)).[8] In addition, the CCA stressed Moore’s improved behavior in prison. 470 S. W. 3d, at 522–524, 526–527. Clinicians, however, caution against reliance on adaptive strengths developed “in a controlled setting,” as a prison surely is. DSM–5, at 38 (“Adaptive functioning may be difficult to assess in a controlled setting (e.g., prisons, detention centers); if possible, corroborative information reflecting functioning outside those settings should be obtained.”); see AAIDD–11 User’s Guide 20 (counseling against reliance on “behavior in jail or prison”). B The CCA furthermore concluded that Moore’s record of academic failure, along with the childhood abuse and suffering he endured, detracted from a determination that his intellectual and adaptive deficits were related. See 470 S. W. 3d, at 488, 526; supra, at 5, 7–8. Those traumatic experiences, however, count in the medical community as “risk factors” for intellectual disability. AAIDD–11, at 59–60 (emphasis added). Clinicians rely on such factors as cause to explore the prospect of intellectual disability further, not to counter the case for a disability determination. See id., at 60 (“[A]t least one or more of the risk factors [described in the manual] will be found in every case of” intellectual disability.). The CCA also departed from clinical practice by requiring Moore to show that his adaptive deficits were not related to “a personality disorder.” 470 S. W. 3d, at 488; see id., at 526 (Moore’s problems in kindergarten were “more likely cause[d]” by “emotional problems” than by intellectual disability). As mental-health professionals recognize, how-ever, many intellectually disabled people also have other mental or physical impairments, for example, attention-deficit/hyperactivity disorder, depressive and bipolar dis-orders, and autism. DSM–5, at 40 (“[c]o-occurring men-tal, neurodevelopmental, medical, and physical conditions are frequent in intellectual disability, with rates of some conditions (e.g., mental disorders, cerebral palsy, and epilepsy) three to four times higher than in the general population”); see AAIDD–11, at 58–63. Coexisting conditions frequently encountered in intellectually disabled individ-uals have been described in clinical literature as “[c]omorbidit[ies].” DSM–5, at 40. See also Brief for AAIDD et al. as Amici Curiae 20, and n. 25. The existence of a personality disorder or mental-health issue, in short, is “not evidence that a person does not also have intellectual disability.” Brief for American Psychological Association, APA, et al. as Amici Curiae 19. C The CCA’s attachment to the seven Briseno evidentiary factors further impeded its assessment of Moore’s adaptive functioning. 1 By design and in operation, the Briseno factors “creat[e] an unacceptable risk that persons with intellectual dis-ability will be executed,” Hall, 572 U. S., at ___ (slip op., at 1). After observing that persons with “mild” intellectual disability might be treated differently under clinical standards than under Texas’ capital system, the CCA defined its objective as identifying the “consensus of Texas citizens” on who “should be exempted from the death penalty.” Briseno, 135 S. W. 3d, at 6 (emphasis added). Mild levels of intellectual disability, although they may fall outside Texas citizens’ consensus, nevertheless remain intellectual disabilities, see Hall, 572 U. S., at ___–___ (slip op., at 17–18); Atkins, 536 U. S., at 308, and n. 3; AAIDD–11, at 153, and States may not execute anyone in “the entire category of [intellectually disabled] offenders,” Roper, 543 U. S., at 563–564 (emphasis added); see supra, at 9. Skeptical of what it viewed as “exceedingly subjective” medical and clinical standards, the CCA in Briseno advanced lay perceptions of intellectual disability. 135 S. W. 3d, at 8; see supra, at 5–6, and n. 6. Briseno asks, for example, “Did those who knew the person best during the developmental stage—his family, friends, teachers, employers, authorities—think he was mentally retarded at that time, and, if so, act in accordance with that determination?” 135 S. W. 3d, at 8. Addressing that question here, the CCA referred to Moore’s education in “normal classrooms during his school career,” his father’s reactions to his academic challenges, and his sister’s perceptions of Moore’s intellectual abilities. 470 S. W. 3d, at 526–527. But the medical profession has endeavored to counter lay stereotypes of the intellectually disabled. See AAIDD–11 User’s Guide 25–27; Brief for AAIDD et al. as Amici Cu-riae 9–14, and nn. 11–15. Those stereotypes, much more than medical and clinical appraisals, should spark skepticism.[9] 2 The Briseno factors are an outlier, in comparison both to other States’ handling of intellectual-disability pleas and to Texas’ own practices in other contexts. See Hall, 572 U. S., at ___ (slip op., at 12) (consensus in the States provides “objective indicia of society’s standards in the context of the Eighth Amendment” (internal quotation marks omitted)). No state legislature has approved the use of the Briseno factors or anything similar. In the 12 years since Texas adopted the factors, only one other state high court and one state intermediate appellate court have authorized their use. See, e.g., Commonwealth v. Bracey, 632 Pa. 75, ___–___, 117 A. 3d 270, 286–287 (2015); Howell v. State, 2011 WL 2420378, *18 (Tenn. Crim. App., June 14, 2011). Indeed, Texas itself does not follow Briseno in contexts other than the death penalty. See Brief for Constitution Project as Amicus Curiae 14–17. For example, the related-ness requirement Texas defends here, see supra, at 5–6, is conspicuously absent from the standards the State uses to assess students for intellectual disabilities. See 19 Tex. Admin. Code §89.1040(c)(5) (2015). And even within Texas’ criminal-justice system, the State requires the intellectual-disability diagnoses of juveniles to be based on “the latest edition of the DSM.” 37 Tex. Admin. Code §380.8751(e)(3) (2016). Texas cannot satisfactorily explain why it applies current medical standards for diagnosing intellectual disability in other contexts, yet clings to superseded standards when an individual’s life is at stake.[10] V As noted supra, at 9, States have some flexibility, but not “unfettered discretion,” in enforcing Atkins’ holding. Hall, 572 U. S., at ___ (slip op., at 17). “If the States were to have complete autonomy to define intellectual disability as they wished,” we have observed, “Atkins could become a nullity, and the Eighth Amendment’s protection of human dignity would not become a reality.” Id., at ___–___ (slip op., at 18–19). The medical community’s current standards supply one constraint on States’ leeway in this area. Reflecting improved understanding over time, see DSM–5, at 7; AAIDD–11, at xiv–xv, current manuals offer “the best available description of how mental disorders are expressed and can be recognized by trained clinicians,” DSM–5, at xli. See also Hall, 572 U. S., at ___, ___, ___, ___–___, ___–___ (slip op., at 2, 3, 8, 10–11, 20–21) (employing current clinical standards); Atkins, 536 U. S., at 308, n. 3, 317, n. 22 (relying on then-current standards). In Moore’s case, the habeas court applied current medical standards in concluding that Moore is intellectually disabled and therefore ineligible for the death penalty. See, e.g., App. to Pet. for Cert. 150a–151a, 200a–203a. The CCA, however, faulted the habeas court for “disregarding [the CCA’s] case law and employing the definition of intellectual disability presently used by the AAIDD.” 470 S. W. 3d, at 486. The CCA instead fastened its intellectual-disability determination to “the AAMR’s 1992 definition of intellectual disability that [it] adopted in Briseno for Atkins claims presented in Texas death-penalty cases.” Ibid. By rejecting the habeas court’s application of medical guidance and clinging to the standard it laid out in Briseno, including the wholly nonclinical Briseno factors, the CCA failed adequately to inform itself of the “medical community’s diagnostic framework,” Hall, 572 U. S., at ___–___ (slip op., at 19–20). Because Briseno pervasively infected the CCA’s analysis, the decision of that court cannot stand. * * * For the reasons stated, the judgment of the Texas Court of Criminal Appeals is vacated, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered.Notes 1 The CCA is Texas’ court of last resort in criminal cases. See Tex. Const., Art. 5, §5. 2 Under Texas law, the CCA, not the court of first instance, is “the ultimate factfinder” in habeas corpus proceedings. Ex parte Reed, 271 S. W. 3d 698, 727 (Tex. Crim. App. 2008); see Ex parte Moore, 470 S. W. 3d 481, 489 (Tex. Crim. App. 2015). 3 The third element is not at issue here. 4 The habeas court considered a seventh score (of 59 on a WAIS–IV test administered in 2013) elsewhere in its opinion, see App. to Pet. for Cert. 170a–172a, but did not include that score in the calculation of Moore’s average IQ score, see id., at 170a. 5 This relatedness requirement, the CCA noted, is retained in the DSM–5. See 470 S. W. 3d, at 487, n. 5 (citing DSM–5, at 38). 6 The seven “Briseno factors” are: 7 The dissent suggests that Hall tacitly approved Idaho’s approach to capital sentencing, which the dissent characterizes as “grant[ing] trial courts discretion to draw ‘reasonable inferences’ about IQ scores and, where appropriate, decline to consider the full range of the [standard error of measurement].” Post, at 14–15 (quoting Hall, 572 U. S., at ___ (slip op., at 15) (quoting Pizzuto v. State, 146 Idaho 720, 729, 202 P. 3d 642, 651 (2008))). We referred in Hall to Idaho’s capital-sentencing scheme, however, only to note that the State had “passed legislation allowing a defendant to present additional evidence of intellectual disability even when an IQ test score is above 70.” 572 U. S., at ___ (slip op., at 15). 8 The dissent suggests that disagreement exists about the precise role of adaptive strengths in the adaptive-functioning inquiry. See post, at 11–12. But even if clinicians would consider adaptive strengths alongside adaptive weaknesses within the same adaptive-skill domain, neither Texas nor the dissent identifies any clinical authority permitting the arbitrary offsetting of deficits against unconnected strengths in which the CCA engaged, see 470 S. W. 3d, at 520–526. 9 As elsewhere in its opinion, the CCA, in its deployment of the Briseno factors, placed undue emphasis on adaptive strengths, see supra, at 12–13; 470 S. W. 3d, at 527, and regarded risk factors for intellectual disability as evidence of the absence of intellectual disability, see supra, at 13–14; 470 S. W. 3d, at 526–527. 10 Given the Briseno factors’ flaws, it is unsurprising that scholars and experts have long criticized the factors. See, e.g., American Bar Assn., Evaluating Fairness and Accuracy in State Death Penalty Systems: The Texas Capital Punishment Assessment Report 395 (2013) (“The Briseno factors create an especially high risk that [an intellectually disabled defendant] will be executed because, in many ways, they contradict established methods for diagnosing [intellectual disability].”); Blume, Johnson, & Seeds, Of Atkins and Men: Deviations from Clinical Definitions of Mental Retardation in Death Penalty Cases (footnote omitted), 18 Cornell J. L. & Pub. Pol’y 689, 710–712 (2009) (“The Briseno factors present an array of divergences from the clinical definitions.”); Macvaugh & Cunningham, Atkins v. Virginia: Implications and Recommendations for Forensic Practice, 37 J. Psychiatry & L. 131, 136 (2009) (“The seven criteria of the Briseno opinion operationalize an Atkins interpretation that [exempts only] a subcategory of persons with [intellectual disabilities] from execution.”). See also 470 S. W. 3d, at 529–530, and n. 5 (Alcala, J., dissenting) (summarizing, in this case, scholarly criticism of Briseno).

582.US.2016_15-214

The St. Croix River, which forms part of the boundary between Wisconsin and Minnesota, is protected under federal, state, and local law. Petitioners own two adjacent lots—Lot E and Lot F—along the lower portion of the river in the town of Troy, Wisconsin. For the area where petitioners’ property is located, state and local regulations prevent the use or sale of adjacent lots under common ownership as separate building sites unless they have at least one acre of land suitable for development. A grandfather clause relaxes this restriction for substandard lots which were in separate ownership from adjacent lands on January 1, 1976, the regulation’s effective date. Petitioners’ parents purchased Lots E and F separately in the 1960’s, and maintained them under separate ownership until transferring Lot F to petitioners in 1994 and Lot E to petitioners in 1995. Both lots are over one acre in size, but because of their topography they each have less than one acre suitable for development. The unification of the lots under common ownership therefore implicated the rules barring their separate sale or development. Petitioners became interested in selling Lot E as part of an improvement plan for the lots, and sought variances from the St. Croix County Board of Adjustment. The Board denied the request, and the state courts affirmed in relevant part. In particular, the State Court of Appeals found that the local ordinance effectively merged the lots, so petitioners could only sell or build on the single combined lot. Petitioners filed suit, alleging that the regulations worked a regulatory taking that deprived them of all, or practically all, of the use of Lot E. The County Circuit Court granted summary judgment to the State, explaining that petitioners had other options to enjoy and use their property, including eliminating the cabin and building a new residence on either lot or across both. The court also found that petitioners had not been deprived of all economic value of their property, because the decrease in market value of the unified lots was less than 10 percent. The State Court of Appeals affirmed, holding that the takings analysis properly focused on Lots E and F together and that, using that framework, the merger regulations did not effect a taking. Held: The State Court of Appeals was correct to analyze petitioners’ property as a single unit in assessing the effect of the challenged governmental action. Pp. 6–20. (a) The Court’s Takings Clause jurisprudence informs the analysis of this issue. Pp. 6–11. (1) Regulatory takings jurisprudence recognizes that if a “regulation goes too far it will be recognized as a taking.” Pennsylvania Coal Co. v. Mahon, 260 U. S. 393 , 415. This area of the law is characterized by “ad hoc, factual inquiries, designed to allow careful examination and weighing of all the relevant circumstances.” Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency, 535 U. S. 302 (citation and internal quotation marks omitted). The Court has, however, identified two guidelines relevant for determining when a government regulation constitutes a taking. First, “with certain qualifications . . . a regulation which ‘denies all economically beneficial or productive use of land’ will require compensation under the Takings Clause.” Palazzolo v. Rhode Island, 533 U. S. 606 (quoting Lucas v. South Carolina Coastal Council, 505 U. S. 1003 ). Second, a taking may be found based on “a complex of factors,” including (1) the economic impact of the regulation on the claimant; (2) the extent to which the regulation has interfered with distinct investment-backed expectations; and (3) the character of the governmental action. Palazzolo, supra, at 617 (citing Penn Central Transp. Co. v. New York City, 438 U. S. 104 ). Yet even the complete deprivation of use under Lucas will not require compensation if the challenged limitations “inhere . . . in the restrictions that background principles of the State’s law of property and nuisance already placed upon land ownership.” Lucas, 505 U. S., at 1029. A central dynamic of the Court’s regulatory takings jurisprudence thus is its flexibility. This is a means to reconcile two competing objectives central to regulatory takings doctrine: the individual’s right to retain the interests and exercise the freedoms at the core of private property ownership, cf. id., at 1027, and the government’s power to “adjus[t] rights for the public good,” Andrus v. Allard, 444 U. S. 51 . Pp. 6–9. (2) This case presents a critical question in determining whether a regulatory taking has occurred: What is the proper unit of property against which to assess the effect of the challenged governmental action? The Court has not set forth specific guidance on how to identify the relevant parcel. However, it has declined to artificially limit the parcel to the portion of property targeted by the challenged regulation, and has cautioned against viewing property rights under the Takings Clause as coextensive with those under state law. Pp. 9–11. (b) Courts must consider a number of factors in determining the proper denominator of the takings inquiry. Pp. 11–17. (1) The inquiry is objective and should determine whether reasonable expectations about property ownership would lead a landowner to anticipate that his holdings would be treated as one parcel or as separate tracts. First, courts should give substantial weight to the property’s treatment, in particular how it is bounded or divided, under state and local law. Second, courts must look to the property’s physical characteristics, including the physical relationship of any distinguishable tracts, topography, and the surrounding human and ecological environment. Third, courts should assess the property’s value under the challenged regulation, with special attention to the effect of burdened land on the value of other holdings. Pp. 11–14. (2) The formalistic rules for which the State of Wisconsin and petitioners advocate do not capture the central legal and factual principles informing reasonable expectations about property interests. Wisconsin would tie the definition of the parcel to state law, but it is also necessary to weigh whether the state enactments at issue accord with other indicia of reasonable expectations about property. Petitioners urge the Court to adopt a presumption that lot lines control, but lot lines are creatures of state law, which can be overridden by the State in the reasonable exercise of its power to regulate land. The merger provision here is such a legitimate exercise of state power, as reflected by its consistency with a long history of merger regulations and with the many merger provisions that exist nationwide today. Pp. 14–17. (c) Under the appropriate multifactor standard, it follows that petitioners’ property should be evaluated as a single parcel consisting of Lots E and F together. First, as to the property’s treatment under state and local law, the valid merger of the lots under state law informs the reasonable expectation that the lots will be treated as a single property. Second, turning to the property’s physical characteristics, the lots are contiguous. Their terrain and shape make it reasonable to expect their range of potential uses might be limited; and petitioners could have anticipated regulation of the property due to its location along the river, which was regulated by federal, state, and local law long before they acquired the land. Third, Lot E brings prospective value to Lot F. The restriction on using the individual lots is mitigated by the benefits of using the property as an integrated whole, allowing increased privacy and recreational space, plus an optimal location for any improvements. This relationship is evident in the lots’ combined valuation. The Court of Appeals was thus correct to treat the contiguous properties as one parcel. Considering petitioners’ property as a whole, the state court was correct to conclude that petitioners cannot establish a compensable taking. They have not suffered a taking under Lucas, as they have not been deprived of all economically beneficial use of their property. See 505 U. S., at 1019. Nor have they suffered a taking under the more general test of Penn Central, supra, at 124. Pp. 17–20. 2015 WI App 13, 359 Wis. 2d 675, 859 N. W. 2d 628, affirmed. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Roberts, C. J., filed a dissenting opinion, in which Thomas and Alito, JJ., joined. Thomas, J., filed a dissenting opinion. Gorsuch, J., took no part in the consideration or decision of the case.

The classic example of a property taking by the government is when the property has been occupied or otherwise seized. In the case now before the Court, petition-ers contend that governmental entities took their real property—an undeveloped residential lot—not by some physical occupation but instead by enacting burdensome regulations that forbid its improvement or separate sale because it is classified as substandard in size. The relevant governmental entities are the respondents. Against the background justifications for the challenged restrictions, respondents contend there is no regulatory taking because petitioners own an adjacent lot. The regulations, in effecting a merger of the property, permit the continued residential use of the property including for a single improvement to extend over both lots. This retained right of the landowner, respondents urge, is of sufficient offsetting value that the regulation is not severe enough to be a regulatory taking. To resolve the issue whether the landowners can insist on confining the analysis just to the lot in question, without regard to their ownership of the adjacent lot, it is necessary to discuss the background principles that define regulatory takings. I A The St. Croix River originates in northwest Wisconsin and flows approximately 170 miles until it joins the Mississippi River, forming the boundary between Minnesota and Wisconsin for much of its length. The lower portion of the river slows and widens to create a natural water area known as Lake St. Croix. Tourists and residents of the region have long extolled the picturesque grandeur of the river and surrounding area. E.g., E. Ellett, Summer Rambles in the West 136–137 (1853). Under the Wild and Scenic Rivers Act, the river was designated, by 1972, for federal protection. §3(a)(6), 82Stat. 908, 16 U. S. C. §1274(a)(6) (designating Upper St. Croix River); Lower Saint Croix River Act of 1972, §2, 86Stat. 1174, 16 U. S. C. §1274(a)(9) (adding Lower St. Croix River). The law required the States of Wisconsin and Minnesota to develop “a management and development program” for the river area. 41 Fed. Reg. 26237 (1976). In compliance, Wisconsin authorized the State Department of Natural Resources to promulgate rules limiting development in order to “guarantee the protection of the wild, scenic and recreational qualities of the river for present and future generations.” Wis. Stat. §30.27(l) (1973). Petitioners are two sisters and two brothers in the Murr family. Petitioners’ parents arranged for them to receive ownership of two lots the family used for recreation along the Lower St. Croix River in the town of Troy, Wisconsin. The lots are adjacent, but the parents purchased them separately, put the title of one in the name of the family business, and later arranged for transfer of the two lots, on different dates, to petitioners. The lots, which are referred to in this litigation as Lots E and F, are described in more detail below. For the area where petitioners’ property is located, the Wisconsin rules prevent the use of lots as separate building sites unless they have at least one acre of land suitable for development. Wis. Admin. Code §§ NR 118.04(4), 118.03(27), 118.06(1)(a)(2)(a), 118.06(1)(b) (2017). A grand-father clause relaxes this restriction for substandardlots which were “in separate ownership from abutting lands” on January 1, 1976, the effective date of the regulation. § NR 118.08(4)(a)(1). The clause permits the use of qualifying lots as separate building sites. The rules also include a merger provision, however, which provides that adjacent lots under common ownership may not be “sold or developed as separate lots” if they do not meet the size requirement. § NR 118.08(4)(a)(2). The Wisconsin rules require localities to adopt parallel provisions, see § NR 118.02(3), so the St. Croix County zoning ordinance contains identical restrictions, see St. Croix County, Wis., Ordinance §17.36I.4.a (2005). The Wisconsin rules also authorize the local zoning authority to grant variances from the regulations where enforcement would create “unnecessary hardship.” § NR 118.09(4)(b); St. Croix County Ordinance §17.09.232. B Petitioners’ parents purchased Lot F in 1960 and built a small recreational cabin on it. In 1961, they transferred title to Lot F to the family plumbing company. In 1963, they purchased neighboring Lot E, which they held in their own names. The lots have the same topography. A steep bluff cuts through the middle of each, with level land suitable for development above the bluff and next to the water below it. The line dividing Lot E from Lot F runs from the riverfront to the far end of the property, crossing the blufftop along the way. Lot E has approximately 60 feet of river frontage, and Lot F has approximately 100 feet. Though each lot is approximately 1.25 acres in size, because of the waterline and the steep bank they each have less than one acre of land suitable for development. Even when combined, the lots’ buildable land area is only 0.98 acres due to the steep terrain. The lots remained under separate ownership, with Lot F owned by the plumbing company and Lot E owned by petitioners’ parents, until transfers to petitioners. Lot F was conveyed to them in 1994, and Lot E was conveyed to them in 1995. Murr v. St. Croix County Bd. of Adjustment, 2011 WI App 29, 332 Wis. 2d 172, 177–178, 184–185, 796 N. W. 2d 837, 841, 844 (2011); 2015 WI App 13, 359 Wis. 2d 675, 859 N. W. 2d 628 (unpublished opinion), App. to Pet. for Cert. A–3, ¶¶4–5. (There are certain ambiguities in the record concerning whether the lots had merged earlier, but the parties and the courts below appear to have assumed the merger occurred upon transfer to petitioners.) A decade later, petitioners became interested in moving the cabin on Lot F to a different portion of the lot and selling Lot E to fund the project. The unification of the lots under common ownership, however, had implicated the state and local rules barring their separate sale or development. Petitioners then sought variances from the St. Croix County Board of Adjustment to enable their building and improvement plan, including a variance to allow the separate sale or use of the lots. The Board denied the requests, and the state courts affirmed in relevant part. In particular, the Wisconsin Court of Appeals agreed with the Board’s interpretation that the local ordinance “effectively merged” Lots E and F, so petitioners “could only sell or build on the single larger lot.” Murr, supra, at 184, 796 N. W. 2d, at 844. Petitioners filed the present action in state court, alleging that the state and county regulations worked a regulatory taking by depriving them of “all, or practically all, of the use of Lot E because the lot cannot be sold or developed as a separate lot.” App. 9. The parties each submitted appraisal numbers to the trial court. Respondents’ appraisal included values of $698,300 for the lots together as regulated; $771,000 for the lots as two distinct build-able properties; and $373,000 for Lot F as a single lot with improvements. Record 17–55, 17–56. Petitioners’ appraisal included an unrebutted, estimated value of $40,000 for Lot E as an undevelopable lot, based on the counterfactual assumption that it could be sold as a separate property. Id., at 22–188. The Circuit Court of St. Croix County granted summary judgment to the State, explaining that petitioners retained “several available options for the use and enjoyment of their property.” Case No. 12–CV–258 (Oct. 31, 2013), App. to Pet. for Cert. B–9. For example, they could preserve the existing cabin, relocate the cabin, or eliminate the cabin and build a new residence on Lot E, on Lot F, or across both lots. The court also found petitioners had not been deprived of all economic value of their property. Considering the valuation of the property as a single lot versus two separate lots, the court found the market value of the property was not significantly affected by the regulations because the decrease in value was less than 10 percent. Ibid. The Wisconsin Court of Appeals affirmed. The court explained that the regulatory takings inquiry required it to “ ‘first determine what, precisely, is the property at issue.’ ” Id., at A–9, ¶17. Relying on Wisconsin Supreme Court precedent in Zealy v. Waukesha, 201 Wis. 2d 365, 548 N. W. 2d 528 (1996), the Court of Appeals rejected petitioners’ request to analyze the effect of the regulations on Lot E only. Instead, the court held the takings analysis “properly focused” on the regulations’ effect “on the Murrs’ property as a whole”—that is, Lots E and F together. App. to Pet. for Cert. A–12, ¶22. Using this framework, the Court of Appeals concluded the merger regulations did not effect a taking. In particular, the court explained that petitioners could not reasonably have expected to use the lots separately because they were “ ‘charged with knowledge of the existing zoning laws’ ” when they acquired the property. Ibid. (quoting Murr, supra, at 184, 796 N. W. 2d, at 844). Thus, “even if [petitioners] did intend to develop or sell Lot E separately, that expectation of separate treatment became unreasonable when they chose to acquire Lot E in 1995, after their having acquired Lot F in 1994.” App. to Pet. for Cert. A–17, ¶30. The court also discounted the severity of the economic impact on petitioners’ property, recognizing the Circuit Court’s conclusion that the regulations diminished the property’s combined value by less than 10 percent. The Supreme Court of Wisconsin denied discretionary review. This Court granted certiorari, 577 U. S. ___ (2016). II A The Takings Clause of the Fifth Amendment provides that private property shall not “be taken for public use, without just compensation.” The Clause is made applicable to the States through the Fourteenth Amendment. Chicago, B. & Q. R. Co. v. Chicago, 166 U. S. 226 (1897) . As this Court has recognized, the plain language of the Takings Clause “requires the payment of compensation whenever the government acquires private property for a public purpose,” see Tahoe-Sierra Preservation Council, Inc. v. Tahoe Regional Planning Agency, 535 U. S. 302, 321 (2002) , but it does not address in specific terms the imposition of regulatory burdens on private property. Indeed, “[p]rior to Justice Holmes’s exposition in Pennsylvania Coal Co. v. Mahon, 260 U. S. 393 (1922) , it was generally thought that the Takings Clause reached only a direct appropriation of property, or the functional equivalent of a practical ouster of the owner’s possession,” like the permanent flooding of property. Lucas v. South Carolina Coastal Council, 505 U. S. 1003, 1014 (1992) (citation, brackets, and internal quotation marks omitted); accord, Horne v. Department of Agriculture, 576 U. S. ___, ___ (2015) (slip op., at 7); see also Loretto v. Teleprompter Manhattan CATV Corp., 458 U. S. 419, 427 (1982) . Mahon, however, initiated this Court’s regulatory takings jurisprudence, declaring that “while property may be regulated to a certain extent, if regulation goes too far it will be recognized as a taking.” 260 U. S., at 415. A regulation, then, can be so burdensome as to become a taking, yet the Mahon Court did not formulate more detailed guidance for determining when this limit is reached. In the near century since Mahon, the Court for the most part has refrained from elaborating this principle through definitive rules. This area of the law has been characterized by “ad hoc, factual inquiries, designed to allow careful examination and weighing of all the relevant circumstances.” Tahoe-Sierra, supra, at 322 (citation and internal quotation marks omitted). The Court has, however, stated two guidelines relevant here for determining when government regulation is so onerous that it constitutes a taking. First, “with certain qualifications . . . a regulation which ‘denies all economically beneficial or productive use of land’ will require compensation under the Takings Clause.” Palazzolo v. Rhode Island, 533 U. S. 606, 617 (2001) (quoting Lucas, supra, at 1015). Second, when a regulation impedes the use of property without depriving the owner of all economically beneficial use, a taking still may be found based on “a complex of factors,” including (1) the economic impact of the regulation on the claimant; (2) the extent to which the regulation has interfered with distinct investment-backed expectations; and (3) the character of the governmental action. Palazzolo, supra, at 617 (citing Penn Central Transp. Co. v. New York City, 438 U. S. 104, 124 (1978) ). By declaring that the denial of all economically beneficial use of land constitutes a regulatory taking, Lucas stated what it called a “categorical” rule. See 505 U. S., at 1015. Even in Lucas, however, the Court included a ca-veat recognizing the relevance of state law and land-use customs: The complete deprivation of use will not re-quire compensation if the challenged limitations “inhere . . . in the restrictions that background principles of the State’s law of property and nuisance already placed upon land ownership.” Id., at 1029; see also id., at 1030–1031 (listing factors for courts to consider in making thisdetermination). A central dynamic of the Court’s regulatory takings jurisprudence, then, is its flexibility. This has been and remains a means to reconcile two competing objectives central to regulatory takings doctrine. One is the individual’s right to retain the interests and exercise the freedoms at the core of private property ownership. Cf. id., at 1028 (“[T]he notion . . . that title is somehow held subject to the ‘implied limitation’ that the State may subsequently eliminate all economically valuable use is inconsistent with the historical compact recorded in the Takings Clause that has become part of our constitutional culture”). Property rights are necessary to preserve freedom, for property ownership empowers persons to shape and to plan their own destiny in a world where governments are always eager to do so for them. The other persisting interest is the government’s well-established power to “adjus[t] rights for the public good.” Andrus v. Allard, 444 U. S. 51, 65 (1979) . As Justice Holmes declared, “Government hardly could go on if to some extent values incident to property could not be diminished without paying for every such change in the general law.” Mahon, supra, at 413. In adjudicating regulatory takings cases a proper balancing of these principles requires a careful inquiry informed by the specifics of the case. In all instances, the analysis must be driven “by the purpose of the Takings Clause, which is to prevent the government from ‘forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.’ ” Palazzolo, supra, at 617–618 (quoting Armstrong v. United States, 364 U. S. 40, 49 (1960) ). B This case presents a question that is linked to the ultimate determination whether a regulatory taking has occurred: What is the proper unit of property against which to assess the effect of the challenged governmental action? Put another way, “[b]ecause our test for regulatory taking requires us to compare the value that has been taken from the property with the value that remains in the property, one of the critical questions is determining how to define the unit of property ‘whose value is to furnish the denominator of the fraction.’ ” Keystone Bituminous Coal Assn. v. DeBenedictis, 480 U. S. 470, 497 (1987) (quoting Michelman, Property, Utility, and Fairness, 80 Harv. L. Rev. 1165, 1992 (1967)). As commentators have noted, the answer to this question may be outcome determinative. See Eagle, The Four-Factor Penn Central Regulatory Takings Test, 118 Pa. St. L. Rev. 601, 631 (2014); see also Wright, A New Time for Denominators, 34 Env. L. 175, 180 (2004). This Court, too, has explained that the question is important to the regulatory takings inquiry. “To the extent that any portion of property is taken, that portion is always taken in its entirety; the relevant question, however, is whether the property taken is all, or only a portion of, the parcel in question.” Concrete Pipe & Products of Cal., Inc. v. Construction Laborers Pension Trust for Southern Cal., 508 U. S. 602, 644 (1993) . Defining the property at the outset, however, should not necessarily preordain the outcome in every case. In some, though not all, cases the effect of the challenged regulation must be assessed and understood by the effect on the entire property held by the owner, rather than just some part of the property that, considered just on its own, has been diminished in value. This demonstrates the contrast between regulatory takings, where the goal is usually to determine how the challenged regulation affects the property’s value to the owner, and physical takings, where the impact of physical appropriation or occupation of the property will be evident. While the Court has not set forth specific guidance on how to identify the relevant parcel for the regulatory taking inquiry, there are two concepts which the Court has indicated can be unduly narrow. First, the Court has declined to limit the parcel in an artificial manner to the portion of property targeted by the challenged regulation. In Penn Central, for example, the Court rejected a challenge to the denial of a permit to build an office tower above Grand Central Terminal. The Court refused to measure the effect of the denial only against the “air rights” above the terminal, cautioning that “ ‘[t]aking’ jurisprudence does not divide a single parcel into discrete segments and attempt to determine whether rights in a particular segment have been entirely abrogated.” 438 U. S., at 130. In a similar way, in Tahoe-Sierra, the Court refused to “effectively sever” the 32 months during which petitioners’ property was restricted by temporary moratoria on development “and then ask whether that segment ha[d] been taken in its entirety.” 535 U. S., at 331. That was because “defining the property interest taken in terms of the very regulation being challenged is circular.” Ibid. That approach would overstate the effect of regulation on property, turning “every delay” into a “total ban.” Ibid. The second concept about which the Court has expressed caution is the view that property rights under the Takings Clause should be coextensive with those under state law. Although property interests have their foundations in state law, the Palazzolo Court reversed a state- court decision that rejected a takings challenge to regulations that predated the landowner’s acquisition of title. 533 U. S., at 626–627. The Court explained that States do not have the unfettered authority to “shape and define property rights and reasonable investment-backed expectations,” leaving landowners without recourse against unreasonable regulations. Id., at 626. By the same measure, defining the parcel by reference to state law could defeat a challenge even to a state enactment that alters permitted uses of property in ways inconsistent with reasonable investment-backed expectations. For example, a State might enact a law that consolidates nonadjacent property owned by a single person or entity in different parts of the State and then imposes development limits on the aggregate set. If a court defined the parcel according to the state law requiring consolidation, this improperly would fortify the state law against a takings claim, because the court would look to the retained value in the property as a whole rather than considering whether individual holdings had lost all value. III A As the foregoing discussion makes clear, no single consideration can supply the exclusive test for determining the denominator. Instead, courts must consider a number of factors. These include the treatment of the land under state and local law; the physical characteristics of the land; and the prospective value of the regulated land. The endeavor should determine whether reasonable expectations about property ownership would lead a landowner to anticipate that his holdings would be treated as one parcel, or, instead, as separate tracts. The inquiry is objective, and the reasonable expectations at issue derive from background customs and the whole of our legal tradition. Cf. Lucas, 505 U. S., at 1035 (Kennedy, J., concurring) (“The expectations protected by the Constitution are based on objective rules and customs that can be understood as reasonable by all parties involved”). First, courts should give substantial weight to the treatment of the land, in particular how it is bounded or divided, under state and local law. The reasonable expectations of an acquirer of land must acknowledge legitimate restrictions affecting his or her subsequent use and dispensation of the property. See Ballard v. Hunter, 204 U. S. 241, 262 (1907) (“Of what concerns or may concern their real estate men usually keep informed, and on that probability the law may frame its proceedings”). A valid takings claim will not evaporate just because a purchaser took title after the law was enacted. See Palazzolo, 533 U. S., at 627 (some “enactments are unreasonable and do not become less so through passage of time or title”). A reasonable restriction that predates a landowner’s acquisition, however, can be one of the objective factors that most landowners would reasonably consider in forming fair expectations about their property. See ibid. (“[A] prospective enactment, such as a new zoning ordinance, can limit the value of land without effecting a taking because it can be understood as reasonable by all concerned”). Ina similar manner, a use restriction which is triggeredonly after, or because of, a change in ownership should also guide a court’s assessment of reasonable private expectations. Second, courts must look to the physical characteristics of the landowner’s property. These include the physical relationship of any distinguishable tracts, the parcel’s topography, and the surrounding human and ecological environment. In particular, it may be relevant that the property is located in an area that is subject to, or likely to become subject to, environmental or other regulation. Cf. Lucas, supra, at 1035 (Kennedy, J., concurring) (“Coastal property may present such unique concerns for a fragile land system that the State can go further in regulating its development and use than the common law of nuisance might otherwise permit”). Third, courts should assess the value of the property under the challenged regulation, with special attention to the effect of burdened land on the value of other holdings. Though a use restriction may decrease the market value of the property, the effect may be tempered if the regulated land adds value to the remaining property, such as by increasing privacy, expanding recreational space, or preserving surrounding natural beauty. A law that limits use of a landowner’s small lot in one part of the city by reason of the landowner’s nonadjacent holdings elsewhere may decrease the market value of the small lot in an unmitigated fashion. The absence of a special relationship between the holdings may counsel against consideration of all the holdings as a single parcel, making the restrictive law susceptible to a takings challenge. On the other hand, if the landowner’s other property is adjacent to the small lot, the market value of the properties may well increase if their combination enables the expansion of a structure, or if development restraints for one part of the parcel protect the unobstructed skyline views of another part. That, in turn, may counsel in favor of treatment as a single parcel and may reveal the weakness of a regulatory takings challenge to the law. State and federal courts have considerable experience in adjudicating regulatory takings claims that depart from these examples in various ways. The Court anticipates that in applying the test above they will continue to exercise care in this complex area. B The State of Wisconsin and petitioners each ask this Court to adopt a formalistic rule to guide the parcel inquiry. Neither proposal suffices to capture the central legal and factual principles that inform reasonable expectations about property interests. Wisconsin would tie the definition of the parcel to state law, considering the two lots here as a single whole due to their merger under the challenged regulations. That approach, as already noted, simply assumes the answer to the question: May the State define the relevant parcel in a way that permits it to escape its responsibility to justify regulation in light of legitimate property expectations? It is, of course, unquestionable that the law must recognize those legitimate expectations in order to give proper weight to the rights of owners and the right of the State to pass reasonable laws and regulations. See Palazzolo, supra, at 627. Wisconsin bases its position on a footnote in Lucas, which suggests the answer to the denominator question “may lie in how the owner’s reasonable expectations have been shaped by the State’s law of property—i.e., whether and to what degree the State’s law has accorded legal recognition and protection to the particular interest in land with respect to which the takings claimant alleges a diminution in (or elimination of) value.” 505 U. S., at 1017, n. 7. As an initial matter, Lucas referenced the parcel problem only in dicta, unnecessary to the announcement or application of the rule it established. See ibid. (“[W]e avoid th[e] difficulty” of determining the relevant parcel “in the present case”). In any event, the test the Court adopts today is consistent with the respect for state law described in Lucas. The test considers state law but in addition weighs whether the state enactments at issue accord with other indicia of reasonable expectations about property. Petitioners propose a different test that is also flawed. They urge the Court to adopt a presumption that lot lines define the relevant parcel in every instance, making Lot E the necessary denominator. Petitioners’ argument, however, ignores the fact that lot lines are themselves creatures of state law, which can be overridden by the State in the reasonable exercise of its power. In effect, petitioners ask this Court to credit the aspect of state law that favors their preferred result (lot lines) and ignore that which does not (merger provision). This approach contravenes the Court’s case law, which recognizes that reasonable land-use regulations do not work a taking. See Palazzolo, 533 U. S., at 627; Mahon, 260 U. S., at 413. Among other cases, Agins v. City of Tiburon, 447 U. S. 255 (1980) , demonstrates the validity of this proposition because it upheld zoning regulations as a legitimate exercise of the government’s police power. Of course, the Court’s later opinion in Lingle v. Chevron U. S. A. Inc. recognized that the test articulated in Agins—that regulation effects a taking if it “ ‘does not substantially advance legitimate state interests’ ”—was improper because it invited courts to engage in heightened review of the effectiveness of government regulation. 544 U. S. 528, 540 (2005) (quoting Agins, supra, at 260). Lingle made clear, however, that the holding of Agins survived, even if its test was “imprecis[e].” See 544 U. S., at 545–546, 548. The merger provision here is likewise a legitimate exercise of government power, as reflected by its consistency with a long history of state and local merger regulations that originated nearly a century ago. See Brief for National Association of Counties et al. as Amici Curiae 5–10. Merger provisions often form part of a regulatory scheme that establishes a minimum lot size in order to preserve open space while still allowing orderly development. See E. McQuillin, Law of Municipal Corporations §25:24 (3d ed. 2010); see also Agins, supra, at 262 (challenged “zoning ordinances benefit[ed] the appellants as well as the public by serving the city’s interest in assuring careful and orderly development of residential property with provision for open-space areas”). When States or localities first set a minimum lot size, there often are existing lots that do not meet the new requirements, and so local governments will strive to reduce substandard lots in a gradual manner. The regulations here represent a classic way of doing this: by implementing a merger provision, which combines contiguous substandard lots under common ownership, alongside a grandfather clause, which preserves adjacent substandard lots that are in separate ownership. Also, as here, the harshness of a merger provision may be ameliorated by the availability of a variance from the local zoning authority for landowners in special circumstances. See 3 E. Ziegler, Rathkopf’s Law of Zoning and Planning §49:13 (39th ed. 2017). Petitioners’ insistence that lot lines define the relevant parcel ignores the well-settled reliance on the merger provision as a common means of balancing the legitimate goals of regulation with the reasonable expectations of landowners. Petitioners’ rule would frustrate municipalities’ ability to implement minimum lot size regulations by casting doubt on the many merger provisions that exist nationwide today. See Brief for National Association of Counties et al. as Amici Curiae 12–31 (listing over 100 examples of merger provisions). Petitioners’ reliance on lot lines also is problematic for another reason. Lot lines have varying degrees of formality across the States, so it is difficult to make them a standard measure of the reasonable expectations of property owners. Indeed, in some jurisdictions, lot lines may be subject to informal adjustment by property owners, with minimal government oversight. See Brief for California et al. as Amici Curiae 17; 1 J. Kushner, Subdivision Law and Growth Management §5:8 (2d ed. 2017) (lot line adjustments that create no new parcels are often exempt from subdivision review); see, e.g., Cal. Govt. Code Ann. §66412(d) (West 2016) (permitting adjustment of lot lines subject to limited conditions for government approval). The ease of modifying lot lines also creates the risk of gamesmanship by landowners, who might seek to alter the lines in anticipation of regulation that seems likely to affect only part of their property. IV Under the appropriate multifactor standard, it follows that for purposes of determining whether a regulatory taking has occurred here, petitioners’ property should be evaluated as a single parcel consisting of Lots E and F together. First, the treatment of the property under state and local law indicates petitioners’ property should be treated as one when considering the effects of the restrictions. As the Wisconsin courts held, the state and local regulations merged Lots E and F. E.g., App. to Pet. for Cert. A–3, ¶6 (“The 1995 transfer of Lot E brought the lots under common ownership and resulted in a merger of the two lots under [the local ordinance]”). The decision to adopt the merger provision at issue here was for a specific and legitimate purpose, consistent with the widespread understanding that lot lines are not dominant or controlling in every case. See supra, at ___. Petitioners’ land was subject to this regulatory burden, moreover, only because of voluntary conduct in bringing the lots under common ownership after the regulations were enacted. As a result, the valid merger of the lots under state law informs the reasonable expectation they will be treated as a single property. Second, the physical characteristics of the property support its treatment as a unified parcel. The lots are contiguous along their longest edge. Their rough terrain and narrow shape make it reasonable to expect their range of potential uses might be limited. Cf. App. to Pet. for Cert. A–5, ¶8 (“[Petitioners] asserted Lot E could not be put to alternative uses like agriculture or commerce due to its size, location and steep terrain”). The land’s location along the river is also significant. Petitioners could have anticipated public regulation might affect their enjoyment of their property, as the Lower St. Croix was a regulated area under federal, state, and local law long before petitioners possessed the land. Third, the prospective value that Lot E brings to Lot F supports considering the two as one parcel for purposes of determining if there is a regulatory taking. Petitioners are prohibited from selling Lots E and F separately or from building separate residential structures on each. Yet this restriction is mitigated by the benefits of using the property as an integrated whole, allowing increased privacy and recreational space, plus the optimal location of any improvements. See Case No. 12–CV–258, App. to Pet. for Cert. B–9 (“They have an elevated level of privacy because they do not have close neighbors and are able to swim and play volleyball at the property”). The special relationship of the lots is further shown by their combined valuation. Were Lot E separately saleable but still subject to the development restriction, petitioners’ appraiser would value the property at only $40,000. We express no opinion on the validity of this figure. We also note the number is not particularly helpful for understanding petitioners’ retained value in the properties because Lot E, under the regulations, cannot be sold without Lot F. The point that is useful for these purposes is that the combined lots are valued at $698,300, which is far greater than the summed value of the separate regulated lots (Lot F with its cabin at $373,000, according to respondents’ appraiser, and Lot E as an undevelopable plot at $40,000, according to petitioners’ appraiser). The value added by the lots’ combination shows their complementarity and supports their treatment as one parcel. The State Court of Appeals was correct in analyzing petitioners’ property as a single unit. Petitioners allege that in doing so, the state court applied a categorical rule that all contiguous, commonly owned holdings must be combined for Takings Clause analysis. See Brief for Petitioners i (“[D]oes the ‘parcel as a whole’ concept . . . establish a rule that two legally distinct, but commonly owned contiguous parcels, must be combined for takings analysis purposes”). This does not appear to be the case, however, for the precedent relied on by the Court of Appeals addressed multiple factors before treating contiguous properties as one parcel. See App. to Pet. for Cert. A–9–A–11, ¶¶17–19 (citing Zealy v. Waukesha, 201 Wis. 2d 365, 548 N. W. 2d 528); see id., at 378, 548 N. W. 2d, at 533 (considering the property as a whole because it was “part of a single purchase” and all 10.4 acres were undeveloped). The judgment below, furthermore, may be affirmed on any ground permitted by the law and record. See Thigpen v. Roberts, 468 U. S. 27, 30 (1984) . To the extent the state court treated the two lots as one parcel based on a bright-line rule, nothing in this opinion approves that methodology, as distinct from the result. Considering petitioners’ property as a whole, the state court was correct to conclude that petitioners cannot establish a compensable taking in these circumstances. Petitioners have not suffered a taking under Lucas, as they have not been deprived of all economically beneficial use of their property. See 505 U. S., at 1019. They can use the property for residential purposes, including an enhanced, larger residential improvement. See Palazzolo, 533 U. S., at 631 (“A regulation permitting a landowner to build a substantial residence . . . does not leave the property ‘economically idle’ ”). The property has not lost all economic value, as its value has decreased by less than 10 percent. See Lucas, supra, at 1019, n. 8 (suggesting that even a landowner with 95 percent loss may not recover). Petitioners furthermore have not suffered a taking under the more general test of Penn Central. See 438 U. S., at 124. The expert appraisal relied upon by the state courts refutes any claim that the economic impact of the regulation is severe. Petitioners cannot claim that they reasonably expected to sell or develop their lots separately given the regulations which predated their acquisition of both lots. Finally, the governmental action was a reasonable land-use regulation, enacted as part of a coordinated federal, state, and local effort to preserve the river and surrounding land. * * * Like the ultimate question whether a regulation has gone too far, the question of the proper parcel in regulatory takings cases cannot be solved by any simple test. See Arkansas Game and Fish Comm’n v. United States, 568 U. S. 23, 31 (2012) . Courts must instead define the parcel in a manner that reflects reasonable expectations about the property. Courts must strive for consistency with the central purpose of the Takings Clause: to “bar Government from forcing some people alone to bear public burdens which, in all fairness and justice, should be borne by the public as a whole.” Armstrong, 364 U. S., at 49. Treating the lot in question as a single parcel is legitimate for purposes of this takings inquiry, and this supports the conclusion that no regulatory taking occurred here. The judgment of the Wisconsin Court of Appeals is affirmed. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.

580.US.2016_15-1251

Article II of the Constitution requires that the President obtain “the Advice and Consent of the Senate” before appointing “Officers of the United States.” §2, cl. 2. Given this provision, the responsibilities of an office requiring Presidential appointment and Senate confirmation (PAS office) may go unperformed if a vacancy arises and the President and Senate cannot promptly agree on a replacement. Congress has accounted for this reality by giving the President limited authority to appoint acting officials to temporarily perform the functions of a vacant PAS office without first obtaining Senate approval. The current version of that authorization is the Federal Vacancies Reform Act of 1998 (FVRA). Section 3345(a) of the FVRA permits three categories of Government officials to perform acting service in a vacant PAS office. Subsection (a)(1) prescribes the general rule that, if a vacancy arises in a PAS office, the first assistant to that office “shall perform” the office’s “functions and duties temporarily in an acting capacity.” Subsections (a)(2) and (a)(3) provide that, “notwithstanding paragraph (1),” the President “may direct” a person already serving in another PAS office, or a senior employee in the relevant agency, to serve in an acting capacity instead. Section 3345 also makes certain individuals ineligible for acting service. Subsection (b)(1) states: “Notwithstanding subsection (a)(1), a person may not serve as an acting officer for an office under this section” if the President nominates him for the vacant PAS office and, during the 365-day period preceding the vacancy, the person “did not serve in the position of first assistant” to that office or “served in [that] position . . . for less than 90 days.” The general counsel of the National Labor Relations Board (NLRB or the Board) is a PAS office. In June 2010, a vacancy arose in that office, and the President directed Lafe Solomon to serve as acting general counsel. Solomon qualified for acting service under subsection (a)(3) of the FVRA, because he was a senior employee at the NLRB. In January 2011, the President nominated Solomon to serve as the NLRB’s general counsel on a permanent basis. The Senate never took action on the nomination, and the President ultimately withdrew Solomon’s name in favor of a new candidate, whom the Senate confirmed in October 2013. Throughout this entire period Solomon served as the acting general counsel to the NLRB. In January 2013, an NLRB Regional Director, exercising authority on Solomon’s behalf, issued an unfair labor practices complaint against respondent SW General, Inc. An Administrative Law Judge concluded that SW General had committed unfair labor practices, and the NLRB agreed. SW General sought review in the United States Court of Appeals for the District of Columbia Circuit, arguing that the complaint was invalid because, under subsection (b)(1) of the FVRA, Solomon could not perform the duties of general counsel to the NLRB after having been nominated to fill that position. The NLRB countered that subsection (b)(1) applies only to first assistants who automatically assume acting duties under subsection (a)(1), not to acting officers who, like Solomon, serve under (a)(2) or (a)(3). The Court of Appeals vacated the Board’s order. It concluded that the prohibition on acting service by nominees contained in subsection (b)(1) applies to all acting officers, regardless of whether they serve pursuant to subsection (a)(1), (a)(2), or (a)(3). As a result, Solomon became ineligible to perform the duties of general counsel in an acting capacity once the President nominated him to fill that post. Held: 1. Subsection (b)(1) of the FVRA prevents a person who has been nominated to fill a vacant PAS office from performing the duties of that office in an acting capacity. The prohibition applies to anyone performing acting service under the FVRA. It is not limited to first assistants performing acting service under subsection (a)(1). Pp. 8–18. (a) The text of the FVRA requires this conclusion. Pp. 8–14. (1) Subsection (b)(1) applies to any “person” and prohibits service “as an acting officer for an office under this section.” “Person” has an expansive meaning that can encompass anyone who performs acting duties under the FVRA. See Pfizer Inc. v. Government of India, 434 U. S. 308 . And “under this section” clarifies that subsection (b)(1) applies to all of §3345: The FVRA contains cross-references to specific subsections and paragraphs. But subsection (b)(1) refers to §3345, which contains all of the ways a person may become an acting officer. The rest of the FVRA also uses the pairing of “person” and “section” to encompass anyone serving as an acting officer under the FVRA, and Congress could readily have used more specific language if it intended subsection (b)(1) to apply only to first assistants acting under (a)(1). The dependent clause at the beginning of subsection (b)(1)—“[n]otwithstanding subsection (a)(1)”—confirms the breadth of the prohibition on acting service by nominees. In statutes, “notwithstanding” clauses show that one provision prevails over another in the event of a conflict. Here, that means that subsection (b)(1) applies even when it conflicts with the default rule in (a)(1) that first assistants “shall perform” acting duties. Pp. 8–10. (2) The Board argues that, because the phrase “notwithstanding subsection (a)(1)” does not mention (a)(2) or (a)(3), Congress did not intend the prohibition in subsection (b)(1) to apply to people serving as acting officers under those provisions. The Board relies on the “interpretive canon, expressio unius est exclusio alterius, expressing one item of [an] associated group or series excludes another left unmentioned.” Chevron U. S. A. Inc. v. Echazabal, 536 U. S. 73 (internal quotation marks omitted). This interpretive canon applies, however, only when “circumstances support[ ] a sensible inference that the term left out must have been meant to be excluded.” Id., at 81. A “notwithstanding” clause does not naturally give rise to such an inference; it just shows which of two or more provisions prevails in the event of a conflict. Singling out one conflict generally does not suggest that other, unaddressed conflicts should be resolved in the opposite manner. Here, the conflict between (a)(1) and (b)(1) is unique: The former uses mandatory language—the first assistant “shall perform” acting duties—while the latter identifies who “may not” serve as an acting officer. The “notwithstanding” clause clarifies that the mandatory language in subsection (a)(1) does not prevail over subsection (b)(1) in the event of a conflict. Subsections (a)(2) and (a)(3) lack that mandatory language, so the natural inference is that Congress left these provisions out of the “notwithstanding” clause because they differ from subsection (a)(1), not to implicitly exempt them from the prohibition in subsection (b)(1). Moreover, subsection (b)(2) specifies that (b)(1) “shall not apply” to certain people who are “serving as the first assistant.” If (b)(1) applied only to first assistants, stating that limitation would be superfluous. Pp. 10–14. (b) Because the text is clear, the Board’s arguments about legislative history, purpose, and post-enactment practice need not be considered. In any event, its arguments are not compelling. The original draft of the FVRA contained a prohibition on nominees serving as acting officers, but explicitly limited that prohibition to first assistants. The Board argues that, when Congress revised this original draft, it made changes to give the President more flexibility to appoint acting officers and did not intend to broaden the prohibition on nominees performing acting service. The glitch in this argument is that Congress did change the prohibition on nominees performing acting service, revising it to clearly apply to all acting officers. The fact that certain Senators stated that they wanted to give the President more flexibility to appoint acting officials does not mean that they got exactly what they wanted. Nor does a statement by one of the sponsors of the FVRA—who said that subsection (b)(1) applies only to first assistants—overcome the clear text, particularly given that the very next Senator to speak offered a contradictory account of the provision. The Board also argues that, since the FVRA was enacted, Congress has not objected when Presidents have nominated individuals who were serving as acting officers under subsection (a)(2) or (a)(3), and that the Office of Legal Counsel and Government Accountability Office have issued guidance construing subsection (b)(1) to apply only to first assistants. Relying on NLRB v. Noel Canning, the Board contends that this “historical practice” is entitled to “significant weight.” 573 U. S. ___. “[H]istorical practice” is too grand a title for the Board’s evidence. The FVRA was not enacted until 1998, and the evidence the Board cites is not significant enough to warrant the conclusion that Congress’s failure to speak up implies that it has acquiesced in the view that subsection (b)(1) applies only to first assistants. By contrast, the Court’s decision in Noel Canning dealt with the President’s constitutional authority under the Recess Appointments Clause; an issue that had attracted intense attention from Presidents, Attorneys General, and the Senate dating back to the beginning of the Republic. Pp. 14–18. 2. Applying the FVRA to this case is straightforward. Subsection (b)(1) prohibited Solomon from continuing his service as acting general counsel once the President nominated him to fill the position permanently. The President could have appointed another person to serve as acting officer in Solomon’s place, but did not do so. P. 18. 796 F. 3d 67, affirmed. Roberts, C. J., delivered the opinion of the Court, in which Kennedy, Thomas, Breyer, Alito, and Kagan, JJ., joined. Thomas, J., filed a concurring opinion. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, J., joined.

Article II of the Constitution requires that the President obtain “the Advice and Consent of the Senate” before appointing “Officers of the United States.” §2, cl. 2. Given this provision, the responsibilities of an office requiring Presidential appointment and Senate confirmation—known as a “PAS” office—may go unperformed if a va-cancy arises and the President and Senate cannot promptly agree on a replacement. Congress has long accountedfor this reality by authorizing the President to direct certain officials to temporarily carry out the duties of a vacant PAS office in an acting capacity, without Senate confirmation. The Federal Vacancies Reform Act of 1998 (FVRA), 5 U. S. C. §3345 et seq., is the latest version of that authorization. Section 3345(a) of the FVRA authorizes three classes of Government officials to become acting officers. The general rule is that the first assistant to a vacant office shall become the acting officer. The President may override that default rule by directing either a person serving in a different PAS office or a senior employee within the relevant agency to become the acting officer instead. The FVRA, however, prohibits certain persons from serving as acting officers if the President has nominated them to fill the vacant office permanently. The question presented is whether that limitation applies only to first assistants who have automatically assumed acting duties, or whether it also applies to PAS officers and senior employees serving as acting officers at the President’s behest. We hold that it applies to all three categories of acting officers. I A The Senate’s advice and consent power is a critical “structural safeguard[ ] of the constitutional scheme.” Edmond v. United States, 520 U. S. 651, 659 (1997) . The Framers envisioned it as “an excellent check upon a spirit of favoritism in the President” and a guard against “the appointment of unfit characters . . . from family connection, from personal attachment, or from a view to popularity.” The Federalist No. 76, p. 457 (C. Rossiter ed. 1961) (A. Hamilton). The constitutional process of Presidential appointment and Senate confirmation, however, can take time: The President may not promptly settle on a nominee to fill an office; the Senate may be unable, or unwilling, to speedily confirm the nominee once submitted. Yet neither may desire to see the duties of the vacant office go unperformed in the interim. Since President Washington’s first term, Congress has given the President limited authority to appoint acting officials to temporarily perform the functions of a vacant PAS office without first obtaining Senate approval. The earliest statutes authorized the appointment of “any person or persons” to fill specific vacancies in the Departments of State, Treasury, and War. Act of May 8, 1792, ch. 37, §8, 1Stat. 281. Congress at first allowed acting officers to serve until the permanent officeholder could resume his duties or a successor was appointed, ibid., but soon imposed a six-month limit on acting service, Act of Feb. 13, 1795, ch. 21, 1Stat. 415. Congress revisited the issue in the 1860s, ultimately passing the Vacancies Act of 1868. The Vacancies Act expanded the number of PAS offices that the President could fill with acting officers. Act of July 23, 1868, ch. 227, 15Stat. 168; see also Act of Feb. 20, 1863, ch. 45, 12Stat. 656. With that expansion came new constraints. The authority to appoint “any person or persons” as an acting officer gave way to a default rule that the “first or sole assistant . . . shall” perform that function, with an exception allowing the President to instead fill the post with a person already serving in a PAS office. 15Stat. 168. And rather than six months of acting service, the Vacancies Act generally authorized only ten days. Ibid. That narrow window of acting service was later lengthened to 30 days. Act of Feb. 6, 1891, ch. 113, 26Stat. 733. During the 1970s and 1980s, interbranch conflict arose over the Vacancies Act. The Department of Justice took the position that, in many instances, the head of an executive agency had independent authority apart from the Vacancies Act to temporarily fill vacant offices. The Comptroller General disagreed, arguing that the Act was the exclusive authority for temporarily filling vacancies in executive agencies. See M. Rosenberg, Congressional Research Service Report for Congress, The New Vacancies Act: Congress Acts to Protect the Senate’s Confirmation Prerogative 2–4 (1998) (Rosenberg). Congress then amended the Vacancies Act to clarify that it applies to such agencies, while at the same time lengthening the term of permissible acting service to 120 days, with a tolling period while a nomination is pending. Id., at 3; see Presidential Transitions Effectiveness Act, §7, 102Stat. 988. But tensions did not ease. By 1998, approximately 20 percent of PAS offices in executive agencies were occupied by “temporary designees, most of whom had served beyond the 120-day limitation period . . . without presidential submissions of nominations.” Rosenberg 1. These acting officers filled high-level positions, sometimes in obvious contravention of the Senate’s wishes. One, for instance, was brought in from outside Government to serve as Acting Assistant Attorney General for the Civil Rights Division of the Justice Department, immediately after the Senate refused to confirm him for that very office. Ibid.; see M. Rosenberg, Congressional Research Service, Valid-ity of Designation of Bill Lann Lee as Acting Assistant Attorney General for Civil Rights 1–3 (1998). Perceiving a threat to the Senate’s advice and consent power, see Rosen-berg 6, Congress acted again. In 1998, it replaced the Vacancies Act with the FVRA. Section 3345(a) of the FVRA permits three categories of Government officials to perform acting service in a vacant PAS office. Subsection (a)(1) prescribes a general rule: If a person serving in a PAS office dies, resigns, or is otherwise unable to perform his duties, the first assistant to that office “shall perform” the office’s “functions and duties . . . temporarily in an acting capacity.” The next two paragraphs of §3345(a) identify alternatives. Subsection (a)(2) provides that “notwithstanding paragraph (1),” the President “may direct a person” who already serves in a PAS office to “perform the functions and duties of the vacant office temporarily in an acting capacity.” Subsection (a)(3) adds that “notwithstanding paragraph (1),” the President “may direct” a person to perform acting duties if the person served in a senior position in the relevant agency for at least 90 days in the 365-day period preceding the vacancy.[1] Section 3345 also makes certain individuals ineligible for acting service. Subsection (b)(1) states: “Notwithstanding subsection (a)(1), a person may not serve as an acting officer for an office under this section” if the President nominates him for the vacant PAS office and, during the 365-day period preceding the vacancy, the individual “did not serve in the position of first assistant” to that office or “served in [that] position . . . for less than 90 days.” Subsection (b)(2) creates an exception to this prohibition, providing that “[p]aragraph (1) shall not apply to any person” serving in a first assistant position that itself requires the Senate’s advice and consent. Other sections of the FVRA establish time limits on acting service and penalties for noncompliance. In most cases, the statute permits acting service for “210 days beginning on the date the vacancy occurs”; tolls that time limit while a nomination is pending; and starts a new 210-day clock if the nomination is “rejected, withdrawn, or returned.” §§3346(a)–(b)(1). Upon a second nomination, the time limit tolls once more, and an acting officer can serve an additional 210 days if the second nomination proves unsuccessful. §3346(b)(2). The FVRA ensures compliance by providing that, in general, “any function or duty of a vacant office” performed by a person not properly serving under the statute “shall have no force or effect.” §3348(d). B The National Labor Relations Board (NLRB or Board) is charged with administering the National Labor Relations Act. By statute, its general counsel must be appointed by the President with the advice and consent of the Senate. 29 U. S. C. §153(d). In June 2010, the NLRB’s general counsel—who had been serving with Senate confirmation—resigned. The President directed Lafe Solomon to serve temporarily as the NLRB’s acting general counsel, citing the FVRA as the basis for the appointment. See Memorandum from President Barack Obama to L. Solomon (June 18, 2010). Solomon satisfied the requirements for acting service under subsection (a)(3) of the FVRA because he had spent the previous ten years in the senior position of Director of the NLRB’s Office of Representation Appeals. The President had bigger plans for Solomon than acting service. On January 5, 2011, he nominated Solomon to serve as the NLRB’s general counsel on a permanent basis. The Senate had other ideas. That body did not act upon the nomination during the 112th Congress, so it was returned to the President when the legislative session expired. 159 Cong. Rec. S17 (Jan. 3, 2013). The President resubmitted Solomon’s name for consideration in the spring of 2013, id., at S3884 (May 23, 2013), but to no avail. The President ultimately withdrew Solomon’s nomination and put forward a new candidate, whom the Senate confirmed on October 29, 2013. Id., at S7635. Throughout this entire period, Solomon served as the NLRB’s acting general counsel. Solomon’s responsibilities included exercising “final authority” to issue complaints alleging unfair labor practices. 29 U. S. C. §§153(d), 160(b). In January 2013, an NLRB Regional Director, exercising authority on Solomon’s behalf, issued a complaint alleging that respondent SW General, Inc.—a company that provides ambulance services—had improperly failed to pay certain bonuses to long-term employees. An Administrative Law Judge concluded that SW General had committed unfair labor practices, and the NLRB agreed. 360 N. L. R. B. 109 (2014). SW General filed a petition for review in the United States Court of Appeals for the District of Columbia Circuit. It argued that the unfair labor practices complaint was invalid because, under subsection (b)(1) of the FVRA, Solomon could not legally perform the duties of general counsel after having been nominated to fill that position. The NLRB defended Solomon’s actions. It contended that subsection (b)(1) applies only to first assistants who automatically assume acting duties under subsection (a)(1), not to acting officers who, like Solomon, serve under (a)(2) or (a)(3). The Court of Appeals granted SW General’s petition for review and vacated the Board’s order. It reasoned that “the text of subsection (b)(1) squarely supports” the conclusion that the provision’s restriction on nominees serving as acting officers “applies to all acting officers, no matter whether they serve pursuant to subsection (a)(1), (a)(2) or (a)(3).” 796 F. 3d 67, 78 (CADC 2015). As a result, Solomon became “ineligible to serve as Acting General Counsel once the President nominated him to be General Counsel.” Id., at 72.[2] We granted certiorari, 579 U. S. ___ (2016), and now affirm. II Subsection (b)(1) of the FVRA prevents a person who has been nominated for a vacant PAS office from performing the duties of that office in an acting capacity. In full, it states: “(1) Notwithstanding subsection (a)(1), a person may not serve as an acting officer for an office under this section, if— (A) during the 365-day period preceding the date of the death, resignation, or beginning of inability to serve, such person— (i) did not serve in the position of first assistant to the office of such officer; or (ii) served in the position of first assistant to the office of such officer for less than 90 days; and (B) the President submits a nomination of such person to the Senate for appointment to such office.” Subsection (b)(2) adds that “[p]aragraph (1) shall not apply” to a person serving in a first assistant position that itself requires the advice and consent of the Senate. We conclude that the prohibition in subsection (b)(1) applies to anyone performing acting service under the FVRA. It is not, as the Board contends, limited to first assistants performing acting service under subsection (a)(1). The text of the prohibition extends to any “person” who serves “as an acting officer . . . under this section,” not just to “first assistants” serving under subsection (a)(1). The phrase “[n]otwithstanding subsection (a)(1)” does not limit the reach of (b)(1), but instead clarifies that the prohibition applies even when it conflicts with the default rule that first assistants shall perform acting duties. A 1 Our analysis of subsection (b)(1) begins with its text. Subsection (b)(1) applies to any “person” and prohibits service “as an acting officer for an office under this section.” The key words are “person” and “section.” They clearly indicate that (b)(1) applies to all acting officers under §3345, regardless of the means of appointment. Start with “person.” The word has a naturally expansive meaning that can encompass anyone who performs acting duties under the FVRA. See Pfizer Inc. v. Government of India, 434 U. S. 308, 312 (1978) . Important as they may be, first assistants are not the only “person[s]” of the bunch. Now add “under this section.” The language clarifies that subsection (b)(1) applies to all persons serving under §3345. Congress often drafts statutes with hierarchical schemes—section, subsection, paragraph, and on down the line. See Koons Buick Pontiac GMC, Inc. v. Nigh, 543 U. S. 50 –61 (2004); L. Filson, The Legislative Drafter’s Desk Reference 222 (1992). Congress used that structure in the FVRA and relied on it to make precise cross-references. When Congress wanted to refer only to a particular subsection or paragraph, it said so. See, e.g., §3346(a)(2) (“subsection (b)”); §3346(b)(2) (“paragraph (1)”). But in (b)(1) Congress referred to the entire section—§3345—which subsumes all of the ways a person may become an acting officer. The rest of the FVRA uses the pairing of “person” and “section” the same way. Section 3346, for example, specifies how long “the person serving as an acting officer as described under section 3345 may serve in the office.” (Emphasis added.) And §3348(d)(1) describes the consequences of noncompliance with the FVRA by referring to the actions “taken by any person who is not acting under section 3345, 3346, or 3347.” (Emphasis added.) No one disputes that both provisions apply to anyone serving as an acting officer under the FVRA, not just first assistants serving under subsection (a)(1). Had Congress intended subsection (b)(1) to apply only to first assistants acting under (a)(1), it could easily have chosen clearer language. Replacing “person” with “first assistant” would have done the trick. So too would replacing “under this section” with “under subsection (a)(1).” “The fact that [Congress] did not adopt [either] readily available and apparent alternative strongly supports” the conclusion that subsection (b)(1) applies to any acting officer appointed under any provision within §3345. Knight v. Commissioner, 552 U. S. 181, 188 (2008) . The dependent clause at the beginning of subsection (b)(1)—“[n]otwithstanding subsection (a)(1)”—confirms that the prohibition on acting service applies even when it conflicts with the default rule that the first assistant shall perform acting duties. The ordinary meaning of “notwithstanding” is “in spite of,” or “without prevention or obstruction from or by.” Webster’s Third New International Dictionary 1545 (1986); Black’s Law Dictionary 1091 (7th ed. 1999) (“Despite; in spite of”). In statutes, the word “shows which provision prevails in the event of a clash.” A. Scalia & B. Garner, Reading Law: The Interpretation of Legal Texts 126–127 (2012). Subsection (a)(1) sets the rule that first assistants “shall perform” the vacant office’s “functions and duties . . . in an acting capacity.” But the “notwithstanding” clause in subsection (b)(1) means that, even if a first assistant is serving as an acting officer under this statutory mandate, he must cease that service if the President nominates him to fill the vacant PAS office. That subsection (b)(1) also applies to acting officers serving at the President’s behest is already clear from the broad text of the independent clause—they are all “person[s]” serving “under this section.” 2 The Board takes a different view of the phrase “[n]otwithstanding subsection (a)(1).” It begins by noting that §3345(a) uses three different subsections to “create three separate paths for becoming an acting official.” Reply Brief 2. The prohibition in subsection (b)(1), the Board continues, “applies ‘[n]otwithstanding’ only one of these subsections—‘subsection (a)(1).’ ” Ibid. In the Board’s view, singling out subsection (a)(1) carries a negative implication: that “Congress did not intend Subsection (b)(1) to override the alternative mechanisms for acting service in Subsections (a)(2) and (a)(3).” Id., at 3. We disagree. The Board relies on the “interpretive canon, expressio unius est exclusio alterius, ‘expressing one item of [an] associated group or series excludes another left unmentioned.’ ” Chevron U. S. A. Inc. v. Echazabal, 536 U. S. 73, 80 (2002) (quoting United States v. Vonn, 535 U. S. 55, 65 (2002) ). If a sign at the entrance to a zoo says “come see the elephant, lion, hippo, and giraffe,” and a temporary sign is added saying “the giraffe is sick,” you would reasonably assume that the others are in good health. “The force of any negative implication, however, depends on context.” Marx v. General Revenue Corp., 568 U. S. ___, ___ (2013) (slip op., at 9). The expressio unius canon applies only when “circumstances support[ ] a sensible inference that the term left out must have been meant to be excluded.” Echazabal, 536 U. S., at 81. A “notwithstanding” clause does not naturally give rise to such an inference; it just shows which of two or more provisions prevails in the event of a conflict. Such a clause confirms rather than constrains breadth. Singling out one potential conflict might suggest that Congress thought the conflict was particularly difficult to resolve, or was quite likely to arise. But doing so generally does not imply anything about other, unaddressed conflicts, much less that they should be resolved in the opposite manner. Suppose a radio station announces: “We play your favorite hits from the ’60s, ’70s, and ’80s. Notwithstanding the fact that we play hits from the ’60s, we do not play music by British bands.” You would not tune in expecting to hear the 1970s British band “The Clash” any more than the 1960s “Beatles.” The station, after all, has announced that “we do not play music by British bands.” The “notwithstanding” clause just establishes that this applies even to music from the ’60s, when British bands were prominently featured on the charts. No one, however, would think the station singled out the ’60s to convey implicitly that its categorical statement “we do not play music by British bands” actually did not apply to the ’70s and ’80s. Drawing a negative inference from the “notwithstanding” clause in subsection (b)(1) is similarly inapt. Without that clause, subsection (b)(1) plainly would apply to all persons serving as acting officers under §3345(a). Adding “notwithstanding subsection (a)(1)” makes sense because (a)(1) conflicts with (b)(1) in a unique manner. The former is mandatory and self-executing: The first assistant “shall perform” acting duties. The latter, by contrast, speaks to who “may not” be an acting officer. So if a vacancy arises and the President nominates the first assistant to fill the position, (a)(1) says the first assistant “shall perform” the duties of that office in an acting capacity while the nomination is pending, and (b)(1) says he “may not.” The “notwithstanding” clause clarifies that the language of (a)(1) does not prevail if that conflict occurs. Compare the mandatory language of subsection (a)(1) to (a)(2) and (a)(3). People appointed under those provisions are just as much acting officers as first assistants who assume the role. But there is no freestanding directive that they perform acting duties; subsections (a)(2) and (a)(3) just say that the President “may direct” them to do so. The natural inference, then, is that Congress left these provisions out of the “notwithstanding” clause because they are different from subsection (a)(1), not to exempt from the broad prohibition in subsection (b)(1) those officers serving under (a)(2) and (a)(3). Indeed, “notwithstanding” is used the same way in other parts of §3345. Subsections (a)(2) and (a)(3) are each preceded by the phrase “notwithstanding paragraph (1).” The phrase recognizes that subsection (a)(1) is unique, and resolves the potential conflict between the mandatory “shall perform” in that provision and the permissive “may direct” in (a)(2) and (a)(3). But it implies nothing about other potential conflicts that may arise in the statutory scheme. In subsection (b)(1), it works the same way: The “notwithstanding” clause simply shows that (b)(1) overrides (a)(1), and nothing more. Step back from the Board’s focus on “notwithstanding” and another problem appears: Its interpretation of subsection (b)(1) makes a mess of (b)(2). Subsection (b)(2) specifies that (b)(1) “shall not apply to any person” if (A) that person “is serving as the first assistant”; (B) the first assistant position is itself a PAS office; and (C) “the Senate has approved the appointment of such person” to that office. The Board’s interpretation makes the first requirement superfluous, a result we typically try to avoid. Williams v. Taylor, 529 U. S. 362, 404 (2000) (“It is . . . a cardinal principle of statutory construction that we must give effect, if possible, to every clause and word of a statute.” (internal quotation marks omitted)). If subsection (b)(1) applied only to first assistants, there would be no need to state the requirement in (b)(2)(A) that “such person is serving as the first assistant.” The Board proposes that Congress did so for clarity, but the same could be said of most superfluous language. The Board and the dissent counter that applying the prohibition in subsection (b)(1) to anyone performing acting service under §3345(a) has its own problem: Doing so would also require applying it to §3345(c)(1), which “would nullify” that provision. Reply Brief 9. The dissent deems this “no way to read a statute.” Post, at 6. We agree, and it is not the way we read it. Under our reading, subsection (b)(1) has no effect on (c)(1). Subsection (b)(1) addresses nominations generally, prohibiting any person who has been nominated to fill any vacant office from performing that office’s duties in an acting capacity. Subsection (c)(1) speaks to a specific nomination scenario: When a person is “nominated by the President for reappointment for an additional term to the same office . . . without a break in service.” In this particular situation, the FVRA authorizes the nominee “to continue to serve in that office.” §3345(c). “[I]t is a commonplace of statutory construction that the specific governs the general.” RadLAX Gateway Hotel, LLC v. Amalgamated Bank, 566 U. S. 639, 645 (2012) . The general prohibition on acting service by nominees yields to the more specific authorization allowing officers up for reappointment to remain at their posts. Applying subsection (b)(1) to §3345(a) hardly compels a different result. The text of subsection (b)(1) is clear: Subject to one narrow exception, it prohibits anyone who has been nominated to fill a vacant PAS office from performing the duties of that office in an acting capacity, regardless of whether the acting officer was appointed under subsection (a)(1), (a)(2), or (a)(3). It is not limited to first assistants who automatically assume acting duties under (a)(1). B The Board contends that legislative history, purpose, and post-enactment practice uniformly show that subsection (b)(1) applies only to first assistants. The text is clear, so we need not consider this extra-textual evidence. See State Farm Fire & Casualty Co. v. United States ex rel. Rigsby, 580 U. S. ___ (2016) (slip op., at 9). In any event, the Board’s evidence is not compelling. The Board argues that subsection (b)(1) was designed to serve a specific purpose: preventing the President from having his nominee serve as an acting officer by making him first assistant after (or right before) a vacancy arises. Brief for Petitioner 38. The original draft of the FVRA authorized first assistants and PAS officers to perform acting service. Subsection (b) of that draft provided that if a first assistant was nominated to fill the vacant office, he could not perform that office’s duties in an acting capacity unless he had been the first assistant for at least 180 days before the vacancy. Several Senators thought the FVRA too restrictive. They asked to add senior agency officials to the list of potential acting officers and to shorten the 180-day length-of-service requirement in subsection (b). Their requests, the Board says, were granted; the final version of the FVRA included subsection (a)(3) for senior employees and shortened the length-of-service requirement to 90 days. There was no intent to extend the pro-hibition in subsection (b) beyond first assistants. Id., at 45–46. The glitch in this argument is of course the text of subsection (b)(1). Congress did amend the statute to allow senior employees to become acting officers under subsection (a)(3). The only substantive change that was requested in (b) was to reduce the length-of-service requirement. Congress could have done that with a few tweaks to the original version of subsection (b). Instead, Congress went further: It also removed language that expressly limited subsection (b) to first assistants. And it added a provision—subsection (b)(2)—that makes sense only if (b)(1) applies to all acting officers. In short, Congress took a provision that explicitly applied only to first assistants and turned it into one that applies to all acting officers. The Board protests that Congress would not have expanded the prohibition on nominees serving as acting officers after Senators asked to give the President more flexibility. See Brief for Petitioner 45–46. That certain Senators made specific demands, however, does not mean that they got exactly what they wanted. Passing a law often requires compromise, where even the most firm public demands bend to competing interests. See Ragsdale v. Wolverine World Wide, Inc., 535 U. S. 81 –94 (2002). What Congress ultimately agrees on is the text that it enacts, not the preferences expressed by certain legislators. See Oncale v. Sundowner Offshore Services, Inc., 523 U. S. 75, 79 (1998) (“[I]t is ultimately the provisions of our laws rather than the principal concerns of our legislators by which we are governed.”). Compromise is precisely what happened here: “[A] period of intense negotiations” took place after Senators demanded changes to the original draft of the FVRA, and the final bill was “a compromise measure.” Rosenberg 9. The legislation as passed did expand the pool of individuals the President could appoint as acting officers, by adding senior employees in subsection (a)(3). But it also expanded the scope of the limitation on acting service in (b)(1), by dropping the language making (b)(1) applicable only to first assistants. The Board contends that this compromise must not have happened because Senator Thompson, one of the sponsors of the FVRA, said that subsection (b)(1) “applies only when the acting officer is the first assistant, and not when the acting officer is designated by the President pursuant to §§3345(a)(2) or 3345(a)(3).” 144 Cong. Rec. 27496 (1998). But Senator Byrd—the very next speaker—offered a contradictory account: A nominee may not “serve as an acting officer” if “he is not the first assistant” or “has been the first assistant for less than 90 . . . days, and has not been confirmed for the position.” Id., at 27498. This is a good example of why floor statements by individual legislators rank among the least illuminating forms of legislative history. See Milner v. Department of Navy, 562 U. S. 562, 572 (2011) (“Those of us who make use of legislative history believe that clear evidence of congressional intent may illuminate ambiguous text. We will not take the opposite tack of allowing ambiguous legislative history to muddy clear statutory language.”). Finally, the Board supports its interpretation with post-enactment practice. It notes that the Office of Legal Counsel and the Government Accountability Office have issued guidance construing subsection (b)(1) to apply only to first assistants. And three Presidents have, without congressional objection, submitted the nominations of 112 individuals who were serving as acting officers under subsections (a)(2) and (a)(3). The Board contends that this “historical practice” is entitled to “significant weight” because the FVRA “concern[s] the allocation of power between two elected branches of Government.” Brief for Petitioner 49 (quoting NLRB v. Noel Canning, 573 U. S. ___, ___–___ (2014) (slip op., at 6–7); internal quotation marks omitted). “[H]istorical practice” is too grand a title for the Board’s evidence. The FVRA was not enacted until 1998, and the 112 nominations that the Board cites make up less than two percent of the thousands of nominations to positions in executive agencies that the Senate has considered in the years since its passage. Even the guidance documents the Board cites paid the matter little attention; both made conclusory statements about subsection (b)(1), with no analysis. In this context, Congress’s failure to speak up does not fairly imply that it has acquiesced in the Board’s interpretation. See Zuber v. Allen, 396 U. S. 168 , n. 21 (1969); Alexander v. Sandoval, 532 U. S. 275, 292 (2001) . The Senate may not have noticed that certain nominees were serving as acting officers in violation of the FVRA, or it may have chosen not to reject a qualified candidate just to make a point about compliance with the statute. Either is at least as plausible as the theory that the Legislature’s inaction reflects considered acceptance of the Executive’s practice. Our decision in Noel Canning—the chief opinion on which the Board relies—is a sharp contrast. That case dealt with the President’s constitutional authority under the Recess Appointments Clause, an issue that has attracted intense attention and written analysis from Presidents, Attorneys General, and the Senate. 573 U. S., at ___–___ (slip op., at 22–32). The voluminous historical record dated back to “the beginning of the Republic,” and included “thousands of intra-session recess appointments.” Id., at ___, ___ (slip op., at 8, 12). That the chronicle of the Recess Appointments Clause weighed heavily in Noel Canning offers no support to the Board here. III Applying the FVRA to this case is straightforward. Solomon was appointed as acting general counsel under subsection (a)(3). Once the President submitted his nomination to fill that position in a permanent capacity, subsection (b)(1) prohibited him from continuing his acting service. This does not mean that the duties of general counsel to the NLRB needed to go unperformed; the President could have appointed another person to serve as the acting officer in Solomon’s place. And he had a wide array of individuals to choose from: any one of the approximately 250 senior NLRB employees or the hundreds of individuals in PAS positions throughout the Government. The President, however, did not do so, and Solomon’s continued service violated the FVRA. Accordingly, the judgment of the Court of Appeals is affirmed. It is so ordered. APPENDIX Section 3345 of the FVRA provides: “(a) If an officer of an Executive agency (including the Executive Office of the President, and other than the Government Accountability Office) whose appointment to office is required to be made by the President, by and with the advice and consent of the Senate, dies, resigns, or is otherwise unable to perform the functions and duties of the office— (1) the first assistant to the office of such officer shall perform the functions and duties of the office temporarily in an acting capacity subject to the time limitations of section 3346; (2) notwithstanding paragraph (1), the President (and only the President) may direct a person who serves in an office for which appointment is required to be made by the President, by and with the advice and consent of the Senate, to perform the functions and duties of the vacant office temporarily in an acting capacity subject to the time limitations of section 3346; or (3) notwithstanding paragraph (1), the President (and only the President) may direct an officer or employee of such Executive agency to perform the functions and duties of the vacant office temporarily in an acting capacity, subject to the time limitations of section 3346, if— (A) during the 365-day period preceding the date of death, resignation, or beginning of inability to serve of the applicable officer, the officer or employee served in a position in such agency for not less than 90 days; and (B) the rate of pay for the position described under subparagraph (A) is equal to or greater than the minimum rate of pay payable for a position at GS–15 of the General Schedule. (b)(1) Notwithstanding subsection (a)(1), a person may not serve as an acting officer for an office under this section, if— (A) during the 365-day period preceding the date of the death, resignation, or beginning of inability to serve, such person— (i) did not serve in the position of first assistant to the office of such officer; or (ii) served in the position of first assistant to the office of such officer for less than 90 days; and (B) the President submits a nomination of such person to the Senate for appointment to such office. (2) Paragraph (1) shall not apply to any person if— (A) such person is serving as the first assistant to the office of an officer described under subsection (a); (B) the office of such first assistant is an office for which appointment is required to be made by the President, by and with the advice and consent of the Senate; and (C) the Senate has approved the appointment of such person to such office. (c)(1) Notwithstanding subsection (a)(1), the President (and only the President) may direct an officer who is nominated by the President for reappointment for an office in an Executive department without a break in service, to continue to serve in that office subject to the time limitations in section 3346, until such time as the Senate has acted to confirm or reject the nomination, notwithstanding adjournment sine die. (2) For purposes of this section and sections 3346, 3347, 3348, 3349, 3349a, and 3349d, the expiration of a term of office is an inability to perform the functions and duties of such office.”Notes 1 A senior position is one that has a rate of pay equal to or greater than the minimum rate “for a position at GS–15 of the General Schedule.” 5 U. S. C. §3345(a)(3)(B). 2 The FVRA exempts “the General Counsel of the National Labor Relations Board” from the general rule that actions taken in violation of the FVRA are void ab initio. 5 U. S. C. §3348(e)(1). The Court of Appeals “assume[d] that section 3348(e)(1) renders the actions of an improperly serving Acting General Counsel voidable” and rejected the Board’s argument against voiding Solomon’s actions. 796 F. 3d, at 79–82. The Board did not seek certiorari on this issue, so we do not con-sider it.

581.US.2016_15-1256

Petitioner Shannon Nelson was convicted by a Colorado jury of two felonies and three misdemeanors arising from the alleged sexual and physical abuse of her four children. The trial court imposed a prison term of 20 years to life and ordered her to pay $8,192.50 in court costs, fees, and restitution. On appeal, Nelson’s conviction was reversed for trial error, and on retrial, she was acquitted of all charges. Petitioner Louis Alonzo Madden was convicted by a Colorado jury of attempting to patronize a prostituted child and attempted sexual assault. The trial court imposed an indeterminate prison sentence and ordered him to pay $4,413.00 in costs, fees, and restitution. After one of Madden’s convictions was reversed on direct review and the other vacated on postconviction review, the State elected not to appeal or retry the case. The Colorado Department of Corrections withheld $702.10 from Nelson’s inmate account between her conviction and acquittal, and Madden paid the State $1,977.75 after his conviction. In both cases, the funds were allocated to costs, fees, and restitution. Once their convictions were invalidated, both petitioners moved for return of the funds. Nelson’s trial court denied her motion outright, and Madden’s postconviction court allowed a refund of costs and fees, but not restitution. The Colorado Court of Appeals concluded that both petitioners were entitled to seek refunds of all they had paid, but the Colorado Supreme Court reversed. It reasoned that Colorado’s Compensation for Certain Exonerated Persons statute (Exoneration Act or Act), Colo. Rev. Stat. §§13–65–101, 13–65–102, 13–65–103, provided the exclusive authority for refunds and that, because neither Nelson nor Madden had filed a claim under that Act, the courts lacked authority to order refunds. The Colorado Supreme Court also held that there was no due process problem under the Act, which permits Colorado to retain conviction-related assessments unless and until the prevailing defendant institutes a discrete civil proceeding and proves her innocence by clear and convincing evidence. Held: The Exoneration Act’s scheme does not comport with the Fourteenth Amendment’s guarantee of due process. Pp. 5–11. (a) The procedural due process inspection required by Mathews v. Eldridge, 424 U. S. 319 , governs these cases. Medina v. California, 505 U. S. 437 , controls when state procedural rules that are part of the criminal process are at issue. These cases, in contrast, concern the continuing deprivation of property after a conviction has been reversed or vacated, with no prospect of reprosecution. Pp. 5–6. (b) The three considerations balanced under Mathews—the private interest affected; the risk of erroneous deprivation of that interest through the procedures used; and the governmental interest at stake—weigh decisively against Colorado’s scheme. Pp. 6–10. (1) Nelson and Madden have an obvious interest in regaining the money they paid to Colorado. The State may not retain these funds simply because Nelson’s and Madden’s convictions were in place when the funds were taken, for once those convictions were erased, the presumption of innocence was restored. See, e.g., Johnson v. Mississippi, 486 U. S. 578 . And Colorado may not presume a person, adjudged guilty of no crime, nonetheless guilty enough for monetary exactions. Pp. 6–8. (2) Colorado’s scheme creates an unacceptable risk of the erroneous deprivation of defendants’ property. The Exoneration Act conditions refund on defendants’ proof of innocence by clear and convincing evidence, but defendants in petitioners’ position are presumed innocent. Moreover, the Act provides no remedy for assessments tied to invalid misdemeanor convictions. And when, as here, the recoupment amount sought is not large, the cost of mounting a claim under the Act and retaining counsel to pursue it would be prohibitive. Colorado argues that an Act that provides sufficient process to compensate a defendant for the loss of her liberty must suffice to compensate a defendant for the lesser deprivation of money. But Nelson and Madden seek the return of their property, not compensation for its temporary deprivation. Just as restoration of liberty on reversal of a conviction is not compensation, neither is the return of money taken by the State on account of the conviction. Other procedures cited by Colorado—the need for probable cause to support criminal charges, the jury-trial right, and the State’s burden to prove guilt beyond a reasonable doubt—do not address the risk faced by a defendant whose conviction has been overturned that she will not recover funds taken from her based solely on a conviction no longer valid. Pp. 8–10. (3) Colorado has no interest in withholding from Nelson and Madden money to which the State currently has zero claim of right. The State has identified no equitable considerations favoring its position, nor indicated any way in which the Exoneration Act embodies such considerations. P. 10. 362 P. 3d 1070 (first judgment) and 364 P. 3d 866 (second judgment), reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in the judgment. Thomas, J., filed a dissenting opinion. Gorsuch, J., took no part in the consideration or decision of the cases.Notes 1 Together with Madden v. Colorado, also on certiorari to the same court (see this Court’s Rule 12.4).

When a criminal conviction is invalidated by a reviewing court and no retrial will occur, is the State obliged to refund fees, court costs, and restitution exacted from the defendant upon, and as a consequence of, the conviction? Our answer is yes. Absent conviction of a crime, one is presumed innocent. Under the Colorado law before us in these cases, however, the State retains conviction-related assessments unless and until the prevailing defendant institutes a discrete civil proceeding and proves her innocence by clear and convincing evidence. This scheme, we hold, offends the Fourteenth Amendment’s guarantee of due process. I A Two cases are before us for review. Petitioner Shannon Nelson, in 2006, was convicted by a Colorado jury of five counts—two felonies and three misdemeanors—arising from the alleged sexual and physical abuse of her four children. 362 P. 3d 1070, 1071 (Colo. 2015); App. 25–26. The trial court imposed a prison sentence of 20 years to life and ordered Nelson to pay court costs, fees, and restitution totaling $8,192.50. 362 P. 3d, at 1071. On appeal, Nelson’s conviction was reversed for trial error. Ibid. On retrial, a new jury acquitted Nelson of all charges. Ibid. Petitioner Louis Alonzo Madden, in 2005, was convicted by a Colorado jury of attempting to patronize a prostituted child and attempted third-degree sexual assault by force. See 364 P. 3d 866, 867 (Colo. 2015). The trial court imposed an indeterminate prison sentence and ordered Madden to pay costs, fees, and restitution totaling $4,413.00. Ibid. The Colorado Supreme Court reversed one of Madden’s convictions on direct review, and a postconviction court vacated the other. Ibid. The State elected not to appeal or retry the case. Ibid. Between Nelson’s conviction and acquittal, the Colorado Department of Corrections withheld $702.10 from her inmate account, $287.50 of which went to costs and fees[1] and $414.60 to restitution. See 362 P. 3d, at 1071, and n. 1. Following Madden’s conviction, Madden paid Colorado $1,977.75, $1,220 of which went to costs and fees[2] and $757.75 to restitution. See 364 P. 3d, at 867. The sole legal basis for these assessments was the fact of Nelson’s and Madden’s convictions.[3] Absent those convictions, Colorado would have no legal right to exact and retain petitioners’ funds. Their convictions invalidated, both petitioners moved for return of the amounts Colorado had taken from them. In Nelson’s case, the trial court denied the motion outright. 362 P. 3d, at 1071. In Madden’s case, the postconviction court allowed the refund of costs and fees, but not restitution. 364 P. 3d, at 867–868. The same Colorado Court of Appeals panel heard both cases and concluded that Nelson and Madden were entitled to seek refunds of all they had paid, including amounts allocated to restitution. See People v. Nelson, 369 P. 3d 625, 628–629 (2013); People v. Madden, 2013 WL 1760869, *1 (Apr. 25, 2013). Costs, fees, and restitution, the court held, must be “tied to a valid conviction,” 369 P. 3d, at 627–628, absent which a court must “retur[n] the defendant to the status quo ante,” 2013 WL 1760869, at *2. The Colorado Supreme Court reversed in both cases. A court must have statutory authority to issue a refund, that court stated. 362 P. 3d, at 1077; 364 P. 3d, at 868. Colorado’s Compensation for Certain Exonerated Persons statute (Exoneration Act or Act), Colo. Rev. Stat. §§13–65–101, 13–65–102, 13–65–103 (2016), passed in 2013, “provides the proper procedure for seeking a refund,” the court ruled. 362 P. 3d, at 1075, 1077. As no other statute addresses refunds, the court concluded that the Exoneration Act is the “exclusive process for exonerated defendants seeking a refund of costs, fees, and restitution.” Id., at 1078.[4] Because neither Nelson nor Madden had filed a claim under the Act, the court further determined, their trial courts lacked authority to order a refund. Id., at 1075, 1078; 364 P. 3d, at 867.[5] There was no due process problem, the court continued, because the Act “provides sufficient process for defendants to seek refunds of costs, fees, and restitution that they paid in connection with their conviction.” 362 P. 3d, at 1078. Justice Hood dissented in both cases. Because neither petitioner has been validly convicted, he explained, each must be presumed innocent. Id., at 1079 (Nelson); 364 P. 3d, at 870 (adopting his reasoning from Nelson in Madden). Due process therefore requires some mechanism “for the return of a defendant’s money,” Justice Hood maintained, 362 P. 3d, at 1080; as the Exoneration Act required petitioners to prove their innocence, the Act, he concluded, did not supply the remedy due process demands, id., at 1081. We granted certiorari. 579 U. S. ___ (2016). B The Exoneration Act provides a civil claim for relief “to compensate an innocent person who was wrongly con-victed.” 362 P. 3d, at 1075. Recovery under the Act is avail-able only to a defendant who has served all or part of a term of incarceration pursuant to a felony conviction, and whose conviction has been overturned for reasons other than insufficiency of evidence or legal error unrelated to actual innocence. See §13–65–102. To succeed on an Exoneration Act claim, a petitioner must show, by clear and convincing evidence, her actual innocence of the offense of conviction. §§13–65–101(1), 13–65–102(1). A successful petitioner may recoup, in addition to compensation for time served,[6] “any fine, penalty, court costs, or restitution . . . paid . . . as a result of his or her wrongful conviction.” Id., at 1075 (quoting §13–65–103(2)(e)(V)). Under Colorado’s legislation, as just recounted, a defendant must prove her innocence by clear and convincing evidence to obtain the refund of costs, fees, and restitution paid pursuant to an invalid conviction. That scheme, we hold, does not comport with due process. Accordingly, we reverse the judgment of the Supreme Court of Colorado. II The familiar procedural due process inspection instructed by Mathews v. Eldridge, 424 U. S. 319 (1976) , governs these cases. Colorado argues that we should instead apply the standard from Medina v. California, 505 U. S. 437, 445 (1992) , and inquire whether Nelson and Madden were exposed to a procedure offensive to a fundamental principle of justice. Medina “provide[s] the appropriate framework for assessing the validity of state procedural rules” that “are part of the criminal process.” Id., at 443. Such rules concern, for example, the allocation of burdens of proof and the type of evidence qualifying as admissible.[7] These cases, in contrast, concern the continuing deprivation of property after a conviction has been reversed or vacated, with no prospect of reprosecution. See Kaley v. United States, 571 U. S. ___, ___, n. 4 (2014) (Roberts, C. J., dissenting) (slip op., at 10–11, n. 4) (explaining the different offices of Mathews and Medina). Because no further criminal process is implicated, Mathews “provides the relevant inquiry.” 571 U. S., at ___ (slip op., at 11, n. 4). III Under the Mathews balancing test, a court evaluates (A) the private interest affected; (B) the risk of erroneous deprivation of that interest through the procedures used; and (C) the governmental interest at stake. 424 U. S., at 335. All three considerations weigh decisively against Colorado’s scheme. A Nelson and Madden have an obvious interest in regaining the money they paid to Colorado. Colorado urges, however, that the funds belong to the State because Nelson’s and Madden’s convictions were in place when the funds were taken. Tr. of Oral Arg. 29–31. But once those convictions were erased, the presumption of their innocence was restored. See, e.g., Johnson v. Mississippi, 486 U. S. 578, 585 (1988) (After a “conviction has been reversed, unless and until [the defendant] should be retried, he must be presumed innocent of that charge.”).[8] “[A]xiomatic and elementary,” the presumption of innocence “lies at the foundation of our criminal law.” Coffin v. United States, 156 U. S. 432, 453 (1895) .[9] Colorado may not retain funds taken from Nelson and Madden solely because of their now-invalidated convictions, see supra, at 2–3, and n. 3, for Colorado may not presume a person, adjudged guilty of no crime, nonetheless guilty enough for monetary exactions.[10] That petitioners prevailed on subsequent review rather than in the first instance, moreover, should be inconsequential. Suppose a trial judge grants a motion to set aside a guilty verdict for want of sufficient evidence. In that event, the defendant pays no costs, fees, or restitution. Now suppose the trial court enters judgment on a guilty verdict, ordering cost, fee, and restitution payments by reason of the conviction, but the appeals court upsets the conviction for evidentiary insufficiency. By what right does the State retain the amount paid out by the defendant? “[I]t should make no difference that the reviewing court, rather than the trial court, determined the evidence to be insufficient.” Burks v. United States, 437 U. S. 1, 11 (1978) . The vulnerability of the State’s argument that it can keep the amounts exacted so long as it prevailed in the court of first instance is more apparent still if we assume a case in which the sole penalty is a fine. On Colorado’s reasoning, an appeal would leave the defendant emptyhanded; regardless of the outcome of an appeal, the State would have no refund obligation. See Tr. of Oral Arg. 41, 44.[11] B Is there a risk of erroneous deprivation of defendants’ interest in return of their funds if, as Colorado urges, the Exoneration Act is the exclusive remedy? Indeed yes, for the Act conditions refund on defendants’ proof of innocence by clear and convincing evidence. §13–65–101(1)(a). But to get their money back, defendants should not be saddled with any proof burden. Instead, as explained supra, at 6–7, they are entitled to be presumed innocent. Furthermore, as Justice Hood noted in dissent, the Act provides no remedy at all for any assessments tied to invalid misdemeanor convictions (Nelson had three). 362 P. 3d, at 1081, n. 1; see §13–65–102(1)(a). And when amounts a defendant seeks to recoup are not large, as is true in Nelson’s and Madden’s cases, see supra, at 2, the cost of mounting a claim under the Exoneration Act and retaining a lawyer to pursue it would be prohibitive.[12] Colorado argued on brief that if the Exoneration Act provides sufficient process to compensate a defendant for the loss of her liberty, the Act should also suffice “when a defendant seeks compensation for the less significant deprivation of monetary assessments paid pursuant to a conviction that is later overturned.” Brief for Respondent 40. The comparison is inapt. Nelson and Madden seek restoration of funds they paid to the State, not compensation for temporary deprivation of those funds. Petitioners seek only their money back, not interest on those funds for the period the funds were in the State’s custody. Just as the restoration of liberty on reversal of a conviction is not compensation, neither is the return of money taken by the State on account of the conviction. Colorado also suggests that “numerous pre- and post-deprivation procedures”—including the need for probable cause to support criminal charges, the jury-trial right, and the State’s burden to prove guilt beyond a reasonable doubt—adequately minimize the risk of erroneous deprivation of property. Id., at 31; see id., at 31–35. But Colorado misperceives the risk at issue. The risk here involved is not the risk of wrongful or invalid conviction any criminal defendant may face. It is, instead, the risk faced by a defendant whose conviction has already been overturned that she will not recover funds taken from her solely on the basis of a conviction no longer valid. None of the above-stated procedures addresses that risk, and, as just explained, the Exoneration Act is not an adequate rem-edy for the property deprivation Nelson and Madden experienced.[13] C Colorado has no interest in withholding from Nelson and Madden money to which the State currently has zero claim of right. “Equitable [c]onsiderations,” Colorado suggests, may bear on whether a State may withhold funds from criminal defendants after their convictionsare overturned. Brief for Respondent 20–22. Colorado, however, has identified no such consideration relevantto petitioners’ cases, nor has the State indicated anyway in which the Exoneration Act embodies “equitable considerations.” IV Colorado’s scheme fails due process measurement because defendants’ interest in regaining their funds is high, the risk of erroneous deprivation of those funds under the Exoneration Act is unacceptable, and the State has shown no countervailing interests in retaining the amounts in question. To comport with due process, a State may not impose anything more than minimal procedures on the refund of exactions dependent upon a conviction subsequently invalidated. * * * The judgments of the Colorado Supreme Court are reversed, and the cases are remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of these cases.Notes 1 Of the $287.50 for costs and fees, $125 went to the victim compensation fund and $162.50 to the victims and witnesses assistance and law enforcement fund (VAST fund). See 362 P. 3d 1070, 1071, n. 1 (Colo. 2015). 2 Of the $1,220 for costs and fees, $125 went to the victim compensation fund and $1,095 to the VAST fund ($1,000 of which was for the special advocate surcharge). See App. 79; 364 P. 3d 866, 869 (Colo. 2015). 3 See Colo. Rev. Stat. §24–4.1–119(1)(a) (2005) (levying victim-compensation-fund fees for “each criminal action resulting in a conviction or in a deferred judgment and sentence”); §24–4.2–104(1)(a)(1)(I) (2005) (same, for VAST fund fees); §24–4.2–104(1)(a)(1)(II) (same, for special advocate surcharge); §18–1.3–603(1) (2005) (with one exception, “[e]very order of conviction . . . shall include consideration of restitution”). See also 362 P. 3d, at 1073 (“[T]he State pays the cost of criminal cases when a defendant is acquitted.” (citing Colo. Rev. Stat. §16–18–101(1) (2015))). Under Colorado law, a restitution order tied to a criminal conviction is rendered as a separate civil judgment. See §18–1.3–603(4)(a) (2005). If the conviction is reversed, any restitution order dependent on that conviction is simultaneously vacated. See People v. Scearce, 87 P. 3d 228, 234–235 (Colo. App. 2003). 4 While these cases were pending in this Court, Colorado passed new legislation to provide “[r]eimbursement of amounts paid following a vacated conviction.” See Colo. House Bill 17–1071 (quoting language for Colo. Rev. Stat. §18–1.3–703, the new provision). That legislation takes effect September 1, 2017, and has no effect on the cases before us. 5 Prior to the Exoneration Act, the Colorado Supreme Court recognized the competence of courts, upon reversal of a conviction, to order the refund of monetary exactions imposed on a defendant solely by reason of the conviction. Toland v. Strohl, 147 Colo. 577, 586, 364 P. 2d 588, 593 (1961). 6 Compensation under the Exoneration Act includes $70,000 per year of incarceration for the wrongful conviction; additional sums per year served while the defendant is under a sentence of death, or placed on parole or probation or on a sex offender registry; compensation for child support payments due during incarceration; tuition waivers at state institutions of higher education for the exonerated person and for any children conceived or legally adopted before the incarceration; and reasonable attorney’s fees for bringing an Exoneration Act claim. §13–65–103(2), (3) (2016). 7 See Cooper v. Oklahoma, 517 U. S. 348 –362 (1996) (standard of proof to establish incompetence to stand trial); Dowling v. United States, 493 U. S. 342 –344, 352 (1990) (admissibility of testimony about a prior crime of which the defendant was acquitted); Patterson v. New York, 432 U. S. 197 –202 (1977) (burden of proving affirmative defense); Medina v. California, 505 U. S. 437 –446, 457 (1992) (burden of proving incompetence to stand trial). 8 Citing Bell v. Wolfish, 441 U. S. 520 (1979) , Colorado asserts that “[t]he presumption of innocence applies only at criminal trials” and thus has no application here. Brief for Respondent 40, n. 19. Colorado misapprehends Wolfish. Our opinion in that case recognized that “under the Due Process Clause,” a detainee who “has not been adjudged guilty of any crime” may not be punished. 441 U. S., at 535–536; see id., at 535–540. Wolfish held only that the presumption does not prevent the government from “detain[ing a defendant] to ensure his presence at trial . . . so long as [the] conditions and restrictions [of his detention] do not amount to punishment, or otherwise violate the Constitution.” Id., at 536–537. 9 Were Medina applicable, Colorado’s Exoneration Act scheme would similarly fail due process measurement. Under Medina, a criminal procedure violates due process if “it offends some principle of justice so rooted in the traditions and conscience of our people as to be ranked as fundamental.” 505 U. S., at 445 (quoting Patterson, 432 U. S., at 202). The presumption of innocence unquestionably fits that bill. 10 Colorado invites a distinction between convictions merely “void-able,” rather than “void,” and urges that the invalidated convictions here fall in the voidable category. See Brief for Respondent 32–33, and n. 11. As Justice Hood noted in dissent, however, “reversal is reversal,” regardless of the reason, “[a]nd an invalid conviction is no conviction at all.” 362 P. 3d, at 1080. 11 The dissent echoes Colorado’s argument. If Nelson and Madden prevailed at trial, the dissent agrees, no costs, fees, or restitution could be exacted. See post, at 6. But if they prevailed on appellate inspection, the State gets to keep their money. See ibid. Under Colorado law, as the dissent reads the Colorado Supreme Court’s opinion, “moneys lawfully exacted pursuant to a valid conviction become public funds (or[, in the case of restitution,] the victims’ money).” Post, at 3–4. Shut from the dissent’s sights, however, the convictions pursuant to which the State took petitioners’ money were invalid, hence the State had no legal right to retain their money. Given the invalidity of the convictions, does the Exoneration Act afford sufficient process to enable the State to retain the money? Surely, it does not. 12 A successful petitioner under the Exoneration Act can recover reasonable attorney’s fees, §13–65–103(2)(e)(IV), but neither a defendant nor counsel is likely to assume the risk of loss when amounts to be gained are not worth the candle. 13 Colorado additionally argues that defendants can request a stay of sentence pending appeal, thereby reducing the risk of erroneous deprivation. See Brief for Respondent 32; §§16–12–103, 18–1.3–702(1)(a) (2016). But the State acknowledged at oral argument that few defendants can meet the requirements a stay pending appeal entails. Tr. of Oral Arg. 33–34. And even when a stay is available, a trial court “may require the defendant to deposit the whole or any part of the . . . costs.” Colo. App. Rule 8.1(a)(3) (2016).

582.US.2016_15-1194

North Carolina law makes it a felony for a registered sex offender “to access a commercial social networking Web site where the sex offender knows that the site permits minor children to become members or to create or maintain personal Web pages.” N. C. Gen. Stat. Ann. §§14–202.5(a), (e). According to sources cited to the Court, the State has prosecuted over 1,000 people for violating this law, including petitioner, who was indicted after posting a statement on his personal Facebook profile about a positive experience in traffic court. The trial court denied petitioner’s motion to dismiss the indictment on the ground that the law violated the First Amendment. He was convicted and given a suspended prison sentence. On appeal, the State Court of Appeals struck down §14–202.5 on First Amendment grounds, but the State Supreme Court reversed. Held: The North Carolina statute impermissibly restricts lawful speech in violation of the First Amendment. Pp. 4–10. (a) A fundamental First Amendment principle is that all persons have access to places where they can speak and listen, and then, after reflection, speak and listen once more. Today, one of the most important places to exchange views is cyberspace, particularly social media, which offers “relatively unlimited, low-cost capacity for communication of all kinds,” Reno v. American Civil Liberties Union, 521 U. S. 844 , to users engaged in a wide array of protected First Amendment activity on any number of diverse topics. The Internet’s forces and directions are so new, so protean, and so far reaching that courts must be conscious that what they say today may be obsolete tomorrow. Here, in one of the first cases the Court has taken to address the relationship between the First Amendment and the modern Internet, the Court must exercise extreme caution before suggesting that the First Amendment provides scant protection for access to vast networks in that medium. Pp. 4–6. (b) This background informs the analysis of the statute at issue. Even assuming that the statute is content neutral and thus subject to intermediate scrutiny, the provision is not “ ‘ “narrowly tailored to serve a significant governmental interest.” ’ ” McCullen v. Coakley, 573 U. S. ___, ___. Like other inventions heralded as advances in human progress, the Internet and social media will be exploited by the criminal mind. It is also clear that “sexual abuse of a child is a most serious crime and an act repugnant to the moral instincts of a decent people,” Ashcroft v. Free Speech Coalition, 535 U. S. 234 , and that a legislature “may pass valid laws to protect children” and other sexual assault victims, id., at 245. However, the assertion of a valid governmental interest “cannot, in every context, be insulated from all constitutional protections.” Stanley v. Georgia, 394 U. S. 557 . Two assumptions are made in resolving this case. First, while the Court need not decide the statute’s precise scope, it is enough to assume that the law applies to commonplace social networking sites like Facebook, LinkedIn, and Twitter. Second, the Court assumes that the First Amendment permits a State to enact specific, narrowly-tailored laws that prohibit a sex offender from engaging in conduct that often presages a sexual crime, like contacting a minor or using a website to gather information about a minor. Even with these assumptions, the statute here enacts a prohibition unprecedented in the scope of First Amendment speech it burdens. Social media allows users to gain access to information and communicate with one another on any subject that might come to mind. With one broad stroke, North Carolina bars access to what for many are the principal sources for knowing current events, checking ads for employment, speaking and listening in the modern public square, and otherwise exploring the vast realms of human thought and knowledge. Foreclosing access to social media altogether thus prevents users from engaging in the legitimate exercise of First Amendment rights. Even convicted criminals—and in some instances especially convicted criminals—might receive legitimate benefits from these means for access to the world of ideas, particularly if they seek to reform and to pursue lawful and rewarding lives. Pp. 6–8. (c) The State has not met its burden to show that this sweeping law is necessary or legitimate to serve its purpose of keeping convicted sex offenders away from vulnerable victims. No case or holding of this Court has approved of a statute as broad in its reach. The State relies on Burson v. Freeman, 504 U. S. 191 , but that case considered a more limited restriction—prohibiting campaigning within 100 feet of a polling place—in order to protect the fundamental right to vote. The Court noted, moreover, that a larger buffer zone could “become an impermissible burden” under the First Amendment. Id., at 210. The better analogy is Board of Airport Comm’rs of Los Angeles v. Jews for Jesus, Inc., 482 U. S. 569 . If an ordinance prohibiting any “ First Amendment activities” at a single Los Angeles airport could be struck down because it covered all manner of protected, nondisruptive behavior, including “talking and reading, or the wearing of campaign buttons or symbolic clothing,” id., at 571, 575, it follows with even greater force that the State may not enact this complete bar to the exercise of First Amendment rights on websites integral to the fabric of modern society and culture. Pp. 9–10. 368 N. C. 380, 777 S. E. 2d 738, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Alito, J., filed an opinion concurring in the judgment, in which Roberts, C. J., and Thomas, J., joined. Gorsuch, J., took no part in the consideration or decision of the case.

In 2008, North Carolina enacted a statute making it a felony for a registered sex offender to gain access to a number of websites, including commonplace social media websites like Facebook and Twitter. The question presented is whether that law is permissible under the First Amendment’s Free Speech Clause, applicable to theStates under the Due Process Clause of the Fourteenth Amendment. I A North Carolina law makes it a felony for a registered sex offender “to access a commercial social networking Web site where the sex offender knows that the site permits minor children to become members or to create or maintain personal Web pages.” N. C. Gen. Stat. Ann. §§14–202.5(a), (e) (2015). A “commercial social networking Web site” is defined as a website that meets four criteria. First, it “[i]s operated by a person who derives revenue from membership fees, advertising, or other sources related to the operation of the Web site.” §14–202.5(b). Second, it “[f]acilitates the social introduction between two or more persons for the purposes of friendship, meeting other persons, or information exchanges.” Ibid. Third, it “[a]llows users to create Web pages or personal profiles that contain information such as the name or nickname of the user, photographs placed on the personal Web page by the user, other personal information about the user, and links to other personal Web pages on the commercial social networking Web site of friends or associates of the user that may be accessed by other users or visitors to the Web site.” Ibid. And fourth, it “[p]rovides users or visitors . . . mechanisms to communicate with other users, such as a message board, chat room, electronic mail, or instant messenger.” Ibid. The statute includes two express exemptions. The statutory bar does not extend to websites that “[p]rovid[e] only one of the following discrete services: photo-sharing, electronic mail, instant messenger, or chat room or message board platform.” §14–202.5(c)(1). The law also does not encompass websites that have as their “primary purpose the facilitation of commercial transactions involving goods or services between [their] members or visitors.” §14–202.5(c)(2). According to sources cited to the Court, §14–202.5 applies to about 20,000 people in North Carolina and the State has prosecuted over 1,000 people for violating it. Brief for Petitioner 6–8. B In 2002, petitioner Lester Gerard Packingham—then a 21-year-old college student—had sex with a 13-year-old girl. He pleaded guilty to taking indecent liberties with a child. Because this crime qualifies as “an offense against a minor,” petitioner was required to register as a sex offender—a status that can endure for 30 years or more. See §14–208.6A; see §14–208.7(a). As a registered sex offender, petitioner was barred under §14–202.5 from gaining access to commercial social networking sites. In 2010, a state court dismissed a traffic ticket against petitioner. In response, he logged on to Facebook.com and posted the following statement on his personal profile: “Man God is Good! How about I got so much favor they dismissed the ticket before court even started? No fine, no court cost, no nothing spent. . . . . .Praise be to GOD, WOW! Thanks JESUS!” App. 136. At the time, a member of the Durham Police Department was investigating registered sex offenders who were thought to be violating §14–202.5. The officer noticed that a “ ‘J. R. Gerrard’ ” had posted the statement quoted above. 368 N. C. 380, 381, 777 S. E. 2d 738, 742 (2015). By checking court records, the officer discovered that a traffic citation for petitioner had been dismissed around the time of the post. Evidence obtained by search warrant confirmed the officer’s suspicions that petitioner was J. R. Gerrard. Petitioner was indicted by a grand jury for violating §14–202.5. The trial court denied his motion to dismiss the indictment on the grounds that the charge against him violated the First Amendment. Petitioner was ultimately convicted and given a suspended prison sentence. At no point during trial or sentencing did the State allege that petitioner contacted a minor—or committed any other illicit act—on the Internet. Petitioner appealed to the Court of Appeals of North Carolina. That court struck down §14–202.5 on First Amendment grounds, explaining that the law is not narrowly tailored to serve the State’s legitimate interest in protecting minors from sexual abuse. 229 N. C. App. 293, 304, 748 S. E. 2d 146, 154 (2013). Rather, the law “arbitrarily burdens all registered sex offenders by preventing a wide range of communication and expressive activity unrelated to achieving its purported goal.” Ibid. The North Carolina Supreme Court reversed, concluding that the law is “constitutional in all respects.” 368 N. C., at 381, 777 S. E. 2d, at 741. Among other things, the court explained that the law is “carefully tailored . . . to prohibit registered sex offenders from accessing only those Web sites that allow them the opportunity to gather information about minors.” Id., at 389, 777 S. E. 2d, at 747. The court also held that the law leaves open adequate alternative means of communication because it permits petitioner to gain access to websites that the court believed perform the “same or similar” functions as social media, such as the Paula Deen Network and the website for the local NBC affiliate. Id., at 390, 777 S. E. 2d, at 747. Two justices dissented. They stated that the law impermissibly “creates a criminal prohibition of alarming breadth and extends well beyond the evils the State seeks to combat.” Id., at 401, 777 S. E. 2d, at 754 (opinion of Hudson, J.) (alteration, citation, and internal quotation marks omitted). The Court granted certiorari, 580 U. S. ___ (2016), and now reverses. II A fundamental principle of the First Amendment is that all persons have access to places where they can speak and listen, and then, after reflection, speak and listen once more. The Court has sought to protect the right to speak in this spatial context. A basic rule, for example, is that a street or a park is a quintessential forum for the exercise of First Amendment rights. See Ward v. Rock Against Racism, 491 U. S. 781, 796 (1989) . Even in the modern era, these places are still essential venues for public gatherings to celebrate some views, to protest others, or simply to learn and inquire. While in the past there may have been difficulty in identifying the most important places (in a spatial sense) for the exchange of views, today the answer is clear. It is cyberspace—the “vast democratic forums of the Internet” in general, Reno v. American Civil Liberties Union, 521 U. S. 844, 868 (1997) , and social media in particular. Seven in ten American adults use at least one Internet social networking service. Brief for Electronic Frontier Foundation et al. as Amici Curiae 5–6. One of the most popular of these sites is Facebook, the site used by petitioner leading to his conviction in this case. According to sources cited to the Court in this case, Facebook has 1.79 billion active users. Id., at 6. This is about three times the population of North America. Social media offers “relatively unlimited, low-cost capacity for communication of all kinds.” Reno, supra, at 870. On Facebook, for example, users can debate religion and politics with their friends and neighbors or share vacation photos. On LinkedIn, users can look for work, advertise for employees, or review tips on entrepreneurship. And on Twitter, users can petition their elected representatives and otherwise engage with them in a direct manner. Indeed, Governors in all 50 States and almost every Member of Congress have set up accounts for this purpose. See Brief for Electronic Frontier Foundation 15–16. In short, social media users employ these websites to engage in a wide array of protected First Amendment activity on topics “as diverse as human thought.” Reno, supra, at 870 (internal quotation marks omitted). The nature of a revolution in thought can be that, in its early stages, even its participants may be unaware of it. And when awareness comes, they still may be unable to know or foresee where its changes lead. Cf. D. Hawke, Benjamin Rush: Revolutionary Gadfly 341 (1971) (quoting Rush as observing: “ ‘The American war is over; but this is far from being the case with the American revolution. On the contrary, nothing but the first act of the great drama is closed’ ”). So too here. While we now may be coming to the realization that the Cyber Age is a revolution of historic proportions, we cannot appreciate yet its full dimensions and vast potential to alter how we think, express ourselves, and define who we want to be. The forces and directions of the Internet are so new, so protean, and so far reaching that courts must be conscious that what they say today might be obsolete tomorrow. This case is one of the first this Court has taken to address the relationship between the First Amendment and the modern Internet. As a result, the Court must exercise extreme caution before suggesting that the First Amendment provides scant protection for access to vast networks in that medium. III This background informs the analysis of the North Carolina statute at issue. Even making the assumption that the statute is content neutral and thus subject to intermediate scrutiny, the provision cannot stand. In order to survive intermediate scrutiny, a law must be “narrowly tailored to serve a significant governmental interest.” McCullen v. Coakley, 573 U. S. ___, ___ (2014) (slip op., at 18) (internal quotation marks omitted). In other words, the law must not “burden substantially more speech than is necessary to further the government’s legitimate interests.” Id., at ___ (slip op., at 19) (internal quotation marks omitted). For centuries now, inventions heralded as advances in human progress have been exploited by the criminal mind. New technologies, all too soon, can become instruments used to commit serious crimes. The railroad is one example, see M. Crichton, The Great Train Robbery, p. xv (1975), and the telephone another, see 18 U. S. C. §1343. So it will be with the Internet and social media. There is also no doubt that, as this Court has recognized, “[t]he sexual abuse of a child is a most serious crime and an act repugnant to the moral instincts of a decent people.” Ashcroft v. Free Speech Coalition, 535 U. S. 234, 244 (2002) . And it is clear that a legislature “may pass valid laws to protect children” and other victims of sexual assault “from abuse.” See id., at 245; accord, New York v. Ferber, 458 U. S. 747, 757 (1982) . The government, of course, need not simply stand by and allow these evils to occur. But the assertion of a valid governmental interest “cannot, in every context, be insulated from all constitutional protections.” Stanley v. Georgia, 394 U. S. 557, 563 (1969) . It is necessary to make two assumptions to resolve this case. First, given the broad wording of the North Carolina statute at issue, it might well bar access not only to commonplace social media websites but also to websitesas varied as Amazon.com, Washingtonpost.com, and Webmd.com. See post, at 6–9; see also Brief for Electronic Frontier Foundation 24–27; Brief for Cato Institute et al. as Amici Curiae 10–12, and n. 6. The Court need not decide the precise scope of the statute. It is enough to assume that the law applies (as the State concedes it does) to social networking sites “as commonly understood”—that is, websites like Facebook, LinkedIn, and Twitter. See Brief for Respondent 54; Tr. of Oral Arg. 27. Second, this opinion should not be interpreted as barring a State from enacting more specific laws than the one at issue. Specific criminal acts are not protected speech even if speech is the means for their commission. See Brandenburg v. Ohio, 395 U. S. 444 –449 (1969) ( per curiam). Though the issue is not before the Court, it can be assumed that the First Amendment permits a State to enact specific, narrowly tailored laws that prohibit a sex offender from engaging in conduct that often presages a sexual crime, like contacting a minor or using a website to gather information about a minor. Cf. Brief for Respondent 42–43. Specific laws of that type must be the State’s first resort to ward off the serious harm that sexual crimes inflict. (Of importance, the troubling fact that the law imposes severe restrictions on persons who already have served their sentence and are no longer subject to the supervision of the criminal justice system is also not an issue before the Court.) Even with these assumptions about the scope of the law and the State’s interest, the statute here enacts a prohibition unprecedented in the scope of First Amendment speech it burdens. Social media allows users to gain access to information and communicate with one another about it on any subject that might come to mind. Supra, at 5. By prohibiting sex offenders from using those websites, North Carolina with one broad stroke bars access to what for many are the principal sources for knowing current events, checking ads for employment, speaking and listening in the modern public square, and otherwise exploring the vast realms of human thought and knowledge. These websites can provide perhaps the most powerful mechanisms available to a private citizen to make his or her voice heard. They allow a person with an Internet connection to “become a town crier with a voice that resonates farther than it could from any soapbox.” Reno, 521 U. S., at 870. In sum, to foreclose access to social media altogether is to prevent the user from engaging in the legitimate exercise of First Amendment rights. It is unsettling to suggest that only a limited set of websites can be used even by persons who have completed their sentences. Even convicted criminals—and in some instances especially convicted criminals—might receive legitimate benefits from these means for access to the world of ideas, in particular if they seek to reform and to pursue lawful and rewarding lives. IV The primary response from the State is that the law must be this broad to serve its preventative purpose of keeping convicted sex offenders away from vulnerable victims. The State has not, however, met its burden to show that this sweeping law is necessary or legitimate to serve that purpose. See McCullen, 573 U. S., at ___ (slip op., at 28). It is instructive that no case or holding of this Court has approved of a statute as broad in its reach. The closest analogy that the State has cited is Burson v. Freeman, 504 U. S. 191 (1992) . There, the Court upheld a prohibition on campaigning within 100 feet of a polling place. That case gives little or no support to the State. The law in Burson was a limited restriction that, in a context consistent with constitutional tradition, was enacted to protect another fundamental right—the right to vote. The restrictions there were far less onerous than those the State seeks to impose here. The law in Burson meant only that the last few seconds before voters entered a polling place were “their own, as free from interference as possible.” Id., at 210. And the Court noted that, were the buffer zone larger than 100 feet, it “could effectively become an impermissible burden” under the First Amendment. Ibid. The better analogy to this case is Board of Airport Comm’rs of Los Angeles v. Jews for Jesus, Inc., 482 U. S. 569 (1987) , where the Court struck down an ordinance prohibiting any “ First Amendment activities” at Los Angeles International Airport because the ordinance covered all manner of protected, nondisruptive behavior including “talking and reading, or the wearing of campaign buttons or symbolic clothing,” id., at 571, 575. If a law prohibiting “all protected expression” at a single airport is not constitutional, id., at 574 (emphasis deleted), it follows with even greater force that the State may not enact this complete bar to the exercise of First Amendment rights on websites integral to the fabric of our modern society and culture. * * * It is well established that, as a general rule, the Government “may not suppress lawful speech as the means to suppress unlawful speech.” Ashcroft v. Free Speech Coalition, 535 U. S., at 255. That is what North Carolina has done here. Its law must be held invalid. The judgment of the North Carolina Supreme Court is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.

580.US.2016_15-606

A Colorado jury convicted petitioner Peña-Rodriguez of harassment and unlawful sexual contact. Following the discharge of the jury, two jurors told defense counsel that, during deliberations, Juror H. C. had expressed anti-Hispanic bias toward petitioner and petitioner’s alibi witness. Counsel, with the trial court’s supervision, obtained affidavits from the two jurors describing a number of biased statements by H. C. The court acknowledged H. C.’s apparent bias but denied petitioner’s motion for a new trial on the ground that Colorado Rule of Evidence 606(b) generally prohibits a juror from testifying as to statements made during deliberations in a proceeding inquiring into the validity of the verdict. The Colorado Court of Appeals affirmed, agreeing that H. C.’s alleged statements did not fall within an exception to Rule 606(b). The Colorado Supreme Court also affirmed, relying on Tanner v. United States, 483 U. S. 107 , and Warger v. Shauers, 574 U. S. ___, both of which rejected constitutional challenges to the federal no-impeachment rule as applied to evidence of juror misconduct or bias. Held: Where a juror makes a clear statement indicating that he or she relied on racial stereotypes or animus to convict a criminal defendant, the Sixth Amendment requires that the no-impeachment rule give way in order to permit the trial court to consider the evidence of the juror’s statement and any resulting denial of the jury trial guarantee. Pp. 6–21. (a) At common law jurors were forbidden to impeach their verdict, either by affidavit or live testimony. Some American jurisdictions adopted a more flexible version of the no-impeachment bar, known as the “Iowa rule,” which prevented jurors from testifying only about their own subjective beliefs, thoughts, or motives during deliberations. An alternative approach, later referred to as the federal approach, permitted an exception only for events extraneous to the deliberative process. This Court’s early decisions did not establish a clear preference for a particular version of the no-impeachment rule, appearing open to the Iowa rule in United States v. Reid, 12 How. 361, and Mattox v. United States, 146 U. S. 140 , but rejecting that approach in McDonald v. Pless, 238 U. S. 264 . The common-law development of the rule reached a milestone in 1975 when Congress adopted Federal Rule of Evidence 606(b), which sets out a broad no-impeachment rule, with only limited exceptions. This version of the no-impeachment rule has substantial merit, promoting full and vigorous discussion by jurors and providing considerable assurance that after being discharged they will not be summoned to recount their deliberations or otherwise harassed. The rule gives stability and finality to verdicts. Pp. 6–9. (b) Some version of the no-impeachment rule is followed in every State and the District of Columbia, most of which follow the Federal Rule. At least 16 jurisdictions have recognized an exception for juror testimony about racial bias in deliberations. Three Federal Courts of Appeals have also held or suggested there is a constitutional exception for evidence of racial bias. In addressing the common-law no-impeachment rule, this Court noted the possibility of an exception in the “gravest and most important cases.” United States v. Reid, supra, at 366; McDonald v. Pless, supra, at 269. The Court has addressed the question whether the Constitution mandates an exception to Rule 606(b) just twice, rejecting an exception each time. In Tanner, where the evidence showed that some jurors were under the influence of drugs and alcohol during the trial, the Court identified “long-recognized and very substantial concerns” supporting the no-impeachment rule. 483 U. S., at 127. The Court also outlined existing, significant safeguards for the defendant’s right to an impartial and competent jury beyond post-trial juror testimony: members of the venire can be examined for impartiality during voir dire; juror misconduct may be observed the court, counsel, and court personnel during the trial; and jurors themselves can report misconduct to the court before a verdict is rendered. In Warger, a civil case where the evidence indicated that the jury forewoman failed to disclose a prodefendant bias during voir dire, the Court again put substantial reliance on existing safeguards for a fair trial. But the Court also warned, as in Reid and McDonald, that the no-impeachment rule may admit of exceptions for “juror bias so extreme that, almost by definition, the jury trial right has been abridged.” 574 U. S., at ___–___, n. 3. Reid, McDonald, and Warger left open the question here: whether the Constitution requires an exception to the no-impeachment rule when a juror’s statements indicate that racial animus was a significant motivating factor in his or her finding of guilt. Pp. 9–13. (c) The imperative to purge racial prejudice from the administration of justice was given new force and direction by the ratification of the Civil War Amendments. “[T]he central purpose of the Fourteenth Amendment was to eliminate racial discrimination emanating from official sources in the States.” McLaughlin v. Florida, 379 U. S. 184 . Time and again, this Court has enforced the Constitution’s guarantee against state-sponsored racial discrimination in the jury system. The Court has interpreted the Fourteenth Amendment to prohibit the exclusion of jurors based on race, Strauder v. West Virginia, 100 U. S. 303 –309; struck down laws and practices that systematically exclude racial minorities from juries, see, e.g., Neal v. Delaware, 103 U. S. 370 ; ruled that no litigant may exclude a prospective juror based on race, see, e.g., Batson v. Kentucky, 476 U. S. 79 ; and held that defendants may at times be entitled to ask about racial bias during voir dire, see, e.g., Ham v. South Carolina, 409 U. S. 524 . The unmistakable principle of these precedents is that discrimination on the basis of race, “odious in all aspects, is especially pernicious in the administration of justice,” Rose v. Mitchell, 443 U. S. 545 , damaging “both the fact and the perception” of the jury’s role as “a vital check against the wrongful exercise of power by the State,” Powers v. Ohio, 499 U. S. 400 . Pp. 13–15. (d) This case lies at the intersection of the Court’s decisions endorsing the no-impeachment rule and those seeking to eliminate racial bias in the jury system. Those lines of precedent need not conflict. Racial bias, unlike the behavior in McDonald, Tanner, or Warger, implicates unique historical, constitutional, and institutional concerns and, if left unaddressed, would risk systemic injury to the administration of justice. It is also distinct in a pragmatic sense, for the Tanner safeguards may be less effective in rooting out racial bias. But while all forms of improper bias pose challenges to the trial process, there is a sound basis to treat racial bias with added precaution. A constitutional rule that racial bias in the justice system must be addressed—including, in some instances, after a verdict has been entered—is necessary to prevent a systemic loss of confidence in jury verdicts, a confidence that is a central premise of the Sixth Amendment trial right. Pp. 15–17. (e) Before the no-impeachment bar can be set aside to allow further judicial inquiry, there must be a threshold showing that one or more jurors made statements exhibiting overt racial bias that cast serious doubt on the fairness and impartiality of the jury’s deliberations and resulting verdict. To qualify, the statement must tend to show that racial animus was a significant motivating factor in the juror’s vote to convict. Whether the threshold showing has been satisfied is committed to the substantial discretion of the trial court in light of all the circumstances, including the content and timing of the alleged statements and the reliability of the proffered evidence. The practical mechanics of acquiring and presenting such evidence will no doubt be shaped and guided by state rules of professional ethics and local court rules, both of which often limit counsel’s post-trial contact with jurors. The experience of those jurisdictions that have already recognized a racial-bias exception to the no-impeachment rule, and the experience of courts going forward, will inform the proper exercise of trial judge discretion. The Court need not address what procedures a trial court must follow when confronted with a motion for a new trial based on juror testimony of racial bias or the appropriate standard for determining when such evidence is sufficient to require that the verdict be set aside and a new trial be granted. Standard and existing safeguards may also help prevent racial bias in jury deliberations, including careful voir dire and a trial court’s instructions to jurors about their duty to review the evidence, deliberate together, and reach a verdict in a fair and impartial way, free from bias of any kind. Pp. 17–21. 350 P. 3d 287, reversed and remanded. Kennedy, J., delivered the opinion of the Court, in which Ginsburg, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed a dissenting opinion. Alito, J., filed a dissenting opinion, in which Roberts, C. J., and Thomas, J., joined.

The jury is a central foundation of our justice system and our democracy. Whatever its imperfections in a particular case, the jury is a necessary check on governmental power. The jury, over the centuries, has been an inspired, trusted, and effective instrument for resolving factual disputes and determining ultimate questions of guilt or innocence in criminal cases. Over the long course its judgments find acceptance in the community, an acceptance essential to respect for the rule of law. The jury is a tangible implementation of the principle that the law comes from the people. In the era of our Nation’s founding, the right to a jury trial already had existed and evolved for centuries, through and alongside the common law. The jury was considered a fundamental safeguard of individual liberty. See The Federalist No. 83, p. 451 (B. Warner ed. 1818) (A. Hamilton). The right to a jury trial in criminal cases was part of the Constitution as first drawn, and it was restated in the Sixth Amendment. Art. III, §2, cl. 3; Amdt. 6. By operation of the Fourteenth Amendment, it is applicable to the States. Duncan v. Louisiana, 391 U. S. 145 –150 (1968). Like all human institutions, the jury system has its flaws, yet experience shows that fair and impartial verdicts can be reached if the jury follows the court’s instructions and undertakes deliberations that are honest, candid, robust, and based on common sense. A general rule has evolved to give substantial protection to verdict final-ity and to assure jurors that, once their verdict has been entered, it will not later be called into question based on the comments or conclusions they expressed during deliberations. This principle, itself centuries old, is often referred to as the no-impeachment rule. The instant case presents the question whether there is an exception to the no-impeachment rule when, after the jury is discharged, a juror comes forward with compelling evidence that an-other juror made clear and explicit statements indicating that racial animus was a significant motivating factor in his or her vote to convict. I State prosecutors in Colorado brought criminal charges against petitioner, Miguel Angel Peña-Rodriguez, based on the following allegations. In 2007, in the bathroom of a Colorado horse-racing facility, a man sexually assaulted two teenage sisters. The girls told their father and identified the man as an employee of the racetrack. The police located and arrested petitioner. Each girl separately identified petitioner as the man who had assaulted her. The State charged petitioner with harassment, unlawful sexual contact, and attempted sexual assault on a child. Before the jury was empaneled, members of the venire were repeatedly asked whether they believed that they could be fair and impartial in the case. A written questionnaire asked if there was “anything about you that you feel would make it difficult for you to be a fair juror.” App. 14. The court repeated the question to the panel of prospective jurors and encouraged jurors to speak in private with the court if they had any concerns about their impartiality. Defense counsel likewise asked whether anyone felt that “this is simply not a good case” for them to be a fair juror. Id., at 34. None of the empaneled jurors expressed any reservations based on racial or any other bias. And none asked to speak with the trial judge. After a 3-day trial, the jury found petitioner guilty of unlawful sexual contact and harassment, but it failed to reach a verdict on the attempted sexual assault charge. When the jury was discharged, the court gave them this instruction, as mandated by Colorado law: “The question may arise whether you may now discuss this case with the lawyers, defendant, or other persons. For your guidance the court instructs you that whether you talk to anyone is entirely your own decision. . . . If any person persists in discussing the case over your objection, or becomes critical of your service either before or after any discussion has begun, please report it to me.” Id., at 85–86. Following the discharge of the jury, petitioner’s counsel entered the jury room to discuss the trial with the jurors. As the room was emptying, two jurors remained to speak with counsel in private. They stated that, during deliberations, another juror had expressed anti-Hispanic bias toward petitioner and petitioner’s alibi witness. Petitioner’s counsel reported this to the court and, with the court’s supervision, obtained sworn affidavits from the two jurors. The affidavits by the two jurors described a number of biased statements made by another juror, identified as Juror H. C. According to the two jurors, H. C. told the other jurors that he “believed the defendant was guilty because, in [H. C.’s] experience as an ex-law enforcement officer, Mexican men had a bravado that caused them to believe they could do whatever they wanted with women.” Id., at 110. The jurors reported that H. C. stated his belief that Mexican men are physically controlling of women because of their sense of entitlement, and further stated, “ ‘I think he did it because he’s Mexican and Mexican men take whatever they want.’ ” Id., at 109. According to the jurors, H. C. further explained that, in his experience, “nine times out of ten Mexican men were guilty of being aggressive toward women and young girls.” Id., at 110. Finally, the jurors recounted that Juror H. C. said that he did not find petitioner’s alibi witness credible because, among other things, the witness was “ ‘an illegal.’ ” Ibid. (In fact, the witness testified during trial that he was a legal resident of the United States.) After reviewing the affidavits, the trial court acknowledged H. C.’s apparent bias. But the court denied petitioner’s motion for a new trial, noting that “[t]he actual deliberations that occur among the jurors are protected from inquiry under [Colorado Rule of Evidence] 606(b).” Id., at 90. Like its federal counterpart, Colorado’s Rule 606(b) generally prohibits a juror from testifying as to any statement made during deliberations in a proceeding inquiring into the validity of the verdict. See Fed. Rule Evid. 606(b). The Colorado Rule reads as follows: “(b) Inquiry into validity of verdict or indictment. Upon an inquiry into the validity of a verdict or indictment, a juror may not testify as to any matter or statement occurring during the course of the jury’s deliberations or to the effect of anything upon his or any other juror’s mind or emotions as influencing him to assent to or dissent from the verdict or indictment or concerning his mental processes in connection therewith. But a juror may testify about (1) whether extraneous prejudicial information was improperly brought to the jurors’ attention, (2) whether any outside influence was improperly brought to bear upon any juror, or (3) whether there was a mistake in entering the verdict onto the verdict form. A juror’s affidavit or evidence of any statement by the juror may not be received on a matter about which the juror would be precluded from testifying.” Colo. Rule Evid. 606(b) (2016). The verdict deemed final, petitioner was sentenced to two years’ probation and was required to register as a sex offender. A divided panel of the Colorado Court of Appeals affirmed petitioner’s conviction, agreeing that H. C.’s alleged statements did not fall within an exception to Rule 606(b) and so were inadmissible to undermine the validity of the verdict. ___ P. 3d ___, 2012 WL 5457362. The Colorado Supreme Court affirmed by a vote of 4 to 3. 350 P. 3d 287 (2015). The prevailing opinion relied on two decisions of this Court rejecting constitutional challenges to the federal no-impeachment rule as applied to evidence of juror misconduct or bias. See Tanner v. United States, 483 U. S. 107 (1987) ; Warger v. Shauers, 574 U. S. ___ (2014). After reviewing those precedents, the court could find no “dividing line between different types of juror bias or misconduct,” and thus no basis for permitting impeachment of the verdicts in petitioner’s trial, notwithstanding H. C.’s apparent racial bias. 350 P. 3d, at 293. This Court granted certiorari to decide whether there is a constitutional exception to the no-impeachment rule for instances of racial bias. 578 U. S. ___ (2016). Juror H. C.’s bias was based on petitioner’s Hispanic identity, which the Court in prior cases has referred to as ethnicity, and that may be an instructive term here. See, e.g., Hernandez v. New York, 500 U. S. 352, 355 (1991) (plurality opinion). Yet we have also used the language of race when discussing the relevant constitutional principles in cases involving Hispanic persons. See, e.g., ibid.; Fisher v. University of Tex. at Austin, 570 U. S. ___ (2013); Rosales-Lopez v. United States, 451 U. S. 182 –190 (1981) (plurality opinion). Petitioner and respondent both refer to race, or to race and ethnicity, in this more expansive sense in their briefs to the Court. This opinion refers to the nature of the bias as racial in keeping with the primary terminology employed by the parties and used in our precedents. II A At common law jurors were forbidden to impeach their verdict, either by affidavit or live testimony. This rule originated in Vaise v. Delaval, 1 T. R. 11, 99 Eng. Rep. 944 (K. B. 1785). There, Lord Mansfield excluded juror testimony that the jury had decided the case through a game of chance. The Mansfield rule, as it came to be known, prohibited jurors, after the verdict was entered, from testifying either about their subjective mental processes or about objective events that occurred during deliberations. American courts adopted the Mansfield rule as a matter of common law, though not in every detail. Some jurisdictions adopted a different, more flexible version of the no-impeachment bar known as the “Iowa rule.” Under that rule, jurors were prevented only from testifying about their own subjective beliefs, thoughts, or motives during deliberations. See Wright v. Illinois & Miss. Tel. Co., 20 Iowa 195 (1866). Jurors could, however, testify about objective facts and events occurring during deliberations, in part because other jurors could corroborate that testimony. An alternative approach, later referred to as the federal approach, stayed closer to the original Mansfield rule. See Warger, supra, at ___ (slip op., at 5). Under this version of the rule, the no-impeachment bar permitted an exception only for testimony about events extraneous to the deliberative process, such as reliance on outside evidence—newspapers, dictionaries, and the like—or personal investigation of the facts. This Court’s early decisions did not establish a clear preference for a particular version of the no-impeachment rule. In United States v. Reid, 12 How. 361 (1852), the Court appeared open to the admission of juror testimony that the jurors had consulted newspapers during deliberations, but in the end it barred the evidence because the newspapers “had not the slightest influence” on the verdict. Id., at 366. The Reid Court warned that juror testimony “ought always to be received with great caution.” Ibid. Yet it added an important admonition: “cases might arise in which it would be impossible to refuse” juror testimony “without violating the plainest principles of justice.” Ibid. In a following case the Court required the admission of juror affidavits stating that the jury consulted information that was not in evidence, including a prejudicial news-paper article. Mattox v. United States, 146 U. S. 140, 151 (1892) . The Court suggested, furthermore, that the admission of juror testimony might be governed by a more flexible rule, one permitting jury testimony even where it did not involve consultation of prejudicial extraneous information. Id., at 148–149; see also Hyde v. United States, 225 U. S. 347 –384 (1912) (stating that the more flexible Iowa rule “should apply,” but excluding evidence that the jury reached the verdict by trading certain defendants’ acquittals for others’ convictions). Later, however, the Court rejected the more lenient Iowa rule. In McDonald v. Pless, 238 U. S. 264 (1915) , the Court affirmed the exclusion of juror testimony about objective events in the jury room. There, the jury allegedly had calculated a damages award by averaging thenumerical submissions of each member. Id., at 265–266. As the Court explained, admitting that evidence would have “dangerous consequences”: “no verdict would be safe” and the practice would “open the door to the most pernicious arts and tampering with jurors.” Id., at 268 (internal quotation marks omitted). Yet the Court reiterated its admonition from Reid, again cautioning that the no-impeachment rule might recognize exceptions “in the gravest and most important cases” where exclusion of juror affidavits might well violate “the plainest principles of justice.” 238 U. S., at 269 (quoting Reid, supra, at 366; internal quotation marks omitted). The common-law development of the no-impeachment rule reached a milestone in 1975, when Congress adopted the Federal Rules of Evidence, including Rule 606(b). Congress, like the McDonald Court, rejected the Iowa rule. Instead it endorsed a broad no-impeachment rule, with only limited exceptions. The version of the rule that Congress adopted was “no accident.” Warger, 574 U. S., at ___ (slip op., at 7). The Advisory Committee at first drafted a rule reflecting the Iowa approach, prohibiting admission of juror testimony only as it related to jurors’ mental processes in reaching a verdict. The Department of Justice, however, expressed concern over the preliminary rule. The Advisory Committee then drafted the more stringent version now in effect, prohibiting all juror testimony, with exceptions only where the jury had considered prejudicial extraneous evidence or was subject to other outside influence. Rules of Evidence for United States Courts and Magistrates, 56 F. R. D. 183, 265 (1972). The Court adopted this second version and transmitted it to Congress. The House favored the Iowa approach, but the Senate expressed concern that it did not sufficiently address the public policy interest in the finality of verdicts. S. Rep. No. 93–1277, pp. 13–14 (1974). Siding with the Senate, the Conference Committee adopted, Congress enacted, and the President signed the Court’s proposed rule. The substance of the Rule has not changed since 1975, except for a 2006 modification permitting evidence of a clerical mistake on the verdict form. See 574 U. S., at ___. The current version of Rule 606(b) states as follows: “(1) Prohibited Testimony or Other Evidence. During an inquiry into the validity of a verdict or indictment, a juror may not testify about any statement made or incident that occurred during the jury’s deliberations; the effect of anything on that juror’s or another juror’s vote; or any juror’s mental processes concerning the verdict or indictment. The court may not receive a juror’s affidavit or evidence of a juror’s statement on these matters. “(2) Exceptions. A juror may testify about whether: “(A) extraneous prejudicial information was improperly brought to the jury’s attention; “(B) an outside influence was improperly brought to bear on any juror; or “(C) a mistake was made in entering the verdict on the verdict form.” This version of the no-impeachment rule has substantial merit. It promotes full and vigorous discussion by providing jurors with considerable assurance that after being discharged they will not be summoned to recount their deliberations, and they will not otherwise be harassed or annoyed by litigants seeking to challenge the verdict. The rule gives stability and finality to verdicts. B Some version of the no-impeachment rule is followed in every State and the District of Columbia. Variations make classification imprecise, but, as a general matter, it appears that 42 jurisdictions follow the Federal Rule, while 9 follow the Iowa Rule. Within both classifications there is a diversity of approaches. Nine jurisdictions that follow the Federal Rule have codified exceptions other than those listed in Federal Rule 606(b). See Appendix, infra. At least 16 jurisdictions, 11 of which follow the Federal Rule, have recognized an exception to the no-impeachment bar under the circumstances the Court faces here: juror testimony that racial bias played a part in deliberations. Ibid. According to the parties and amici, only one State other than Colorado has addressed this issue and declined to recognize an exception for racial bias. See Commonwealth v. Steele, 599 Pa. 341, 377–379, 961 A. 2d 786, 807–808 (2012). The federal courts, for their part, are governed by Federal Rule 606(b), but their interpretations deserve further comment. Various Courts of Appeals have had occasion to consider a racial bias exception and have reached different conclusions. Three have held or suggested there is a constitutional exception for evidence of racial bias. See United States v. Villar, 586 F. 3d 76, 87–88 (CA1 2009) (holding the Constitution demands a racial-bias exception); United States v. Henley, 238 F. 3d 1111, 1119–1121 (CA9 2001) (finding persuasive arguments in favor of an exception but not deciding the issue); Shillcutt v. Gagnon, 827 F. 2d 1155, 1158–1160 (CA7 1987) (observing that in some cases fundamental fairness could require an exception). One Court of Appeals has declined to find an exception, reasoning that other safeguards inherent in the trial process suffice to protect defendants’ constitutional interests. See United States v. Benally, 546 F. 3d 1230, 1240–1241 (CA10 2008). Another has suggested as much, holding in the habeas context that an exception for racial bias was not clearly established but indicating in dicta that no such exception exists. See Williams v. Price, 343 F. 3d 223, 237–239 (CA3 2003) (Alito, J.). And one Court of Appeals has held that evidence of racial bias is excluded by Rule 606(b), without addressing whether the Constitution may at times demand an exception. See Martinez v. Food City, Inc., 658 F. 2d 369, 373–374 (CA5 1981). C In addressing the scope of the common-law no-impeachment rule before Rule 606(b)’s adoption, the Reid and McDonald Courts noted the possibility of an exception to the rule in the “gravest and most important cases.” Reid, 12 How., at 366; McDonald, 238 U. S., at 269. Yet since the enactment of Rule 606(b), the Court has addressed the precise question whether the Constitution mandates an exception to it in just two instances. In its first case, Tanner, 483 U. S. 107 , the Court rejected a Sixth Amendment exception for evidence that some jurors were under the influence of drugs and alcohol during the trial. Id., at 125. Central to the Court’s reasoning were the “long-recognized and very substantial concerns” supporting “the protection of jury deliberations from intrusive inquiry.” Id., at 127. The Tanner Court echoed McDonald’s concern that, if attorneys could use juror testimony to attack verdicts, jurors would be “harassed and beset by the defeated party,” thus destroying “all frankness and freedom of discussion and conference.” 483 U. S., at 120 (quoting McDonald, supra, at 267–268). The Court was concerned, moreover, that attempts to impeach a verdict would “disrupt the finality of the process” and undermine both “jurors’ willingness to return an unpopular verdict” and “the community’s trust in a system that relies on the decisions of laypeople.” 483 U. S., at 120–121. The Tanner Court outlined existing, significant safeguards for the defendant’s right to an impartial and competent jury beyond post-trial juror testimony. At the outset of the trial process, voir dire provides an opportun-ity for the court and counsel to examine members of the venire for impartiality. As a trial proceeds, the court, counsel, and court personnel have some opportunity to learn of any juror misconduct. And, before the verdict, jurors themselves can report misconduct to the court. These procedures do not undermine the stability of a verdict once rendered. Even after the trial, evidence of misconduct other than juror testimony can be used to attempt to impeach the verdict. Id., at 127. Balancing these interests and safeguards against the defendant’s Sixth Amendment interest in that case, the Court affirmed the exclusion of affidavits pertaining to the jury’s inebri-ated state. Ibid. The second case to consider the general issue presented here was Warger, 574 U. S. ___. The Court again rejected the argument that, in the circumstances there, the jury trial right required an exception to the no-impeachment rule. Warger involved a civil case where, after the verdict was entered, the losing party sought to proffer evidence that the jury forewoman had failed to disclose prodefendant bias during voir dire. As in Tanner, the Court put substantial reliance on existing safeguards for a fair trial. The Court stated: “Even if jurors lie in voir dire in a way that conceals bias, juror impartiality is adequately assured by the parties’ ability to bring to the court’s attention any evidence of bias before the verdict is rendered, and to employ nonjuror evidence even after the verdict is rendered.” 574 U. S., at ___ (slip op., at 10). In Warger, however, the Court did reiterate that the no-impeachment rule may admit exceptions. As in Reid and McDonald, the Court warned of “juror bias so extreme that, almost by definition, the jury trial right has been abridged.” 574 U. S., at ___–___, n. 3 (slip op., at 10–11, n. 3). “If and when such a case arises,” the Court indicated it would “consider whether the usual safeguards are orare not sufficient to protect the integrity of the process.” Ibid. The recognition in Warger that there may be extreme cases where the jury trial right requires an exception to the no-impeachment rule must be interpreted in context as a guarded, cautious statement. This caution is warranted to avoid formulating an exception that might undermine the jury dynamics and finality interests the no-impeachment rule seeks to protect. Today, however, the Court faces the question that Reid, McDonald, and Warger left open. The Court must decide whether the Constitution requires an exception to the no-impeachment rule when a juror’s statements indicate that racial animus was a significant motivating factor in his or her finding of guilt. III It must become the heritage of our Nation to rise above racial classifications that are so inconsistent with our commitment to the equal dignity of all persons. This imperative to purge racial prejudice from the administration of justice was given new force and direction by the ratification of the Civil War Amendments. “[T]he central purpose of the Fourteenth Amendment was to eliminate racial discrimination emanating from official sources in the States.” McLaughlin v. Florida, 379 U. S. 184, 192 (1964) . In the years before and after the ratification of the Fourteenth Amendment, it became clear that racial discrimination in the jury system posed a particular threat both to the promise of the Amendment and to the integrity of the jury trial. “Almost immediately after the Civil War, the South began a practice that would continue for many decades: All-white juries punished black defendants particularly harshly, while simultaneously refusing to punish violence by whites, including Ku Klux Klan members, against blacks and Republicans.” Forman, Juries and Race in the Nineteenth Century, 113 Yale L. J. 895, 909–910 (2004). To take one example, just in the years 1865 and 1866, all-white juries in Texas decided a total of 500 prosecutions of white defendants charged with killing African-Americans. All 500 were acquitted. Id., at 916. The stark and unapologetic nature of race-motivated outcomes challenged the American belief that “the jury was a bulwark of liberty,” id., at 909, and prompted Congress to pass legislation to integrate the jury system and to bar persons from eligibility for jury service if they had conspired to deny the civil rights of African-Americans, id., at 920–930. Members of Congress stressed that the legislation was necessary to preserve the right to a fair trial and to guarantee the equal protection of the laws. Ibid. The duty to confront racial animus in the justice system is not the legislature’s alone. Time and again, this Court has been called upon to enforce the Constitution’s guarantee against state-sponsored racial discrimination in the jury system. Beginning in 1880, the Court interpreted the Fourteenth Amendment to prohibit the exclusion of jurors on the basis of race. Strauder v. West Virginia, 100 U. S. 303 –309 (1880). The Court has repeatedly struck down laws and practices that systematically exclude racial minorities from juries. See, e.g., Neal v. Delaware, 103 U. S. 370 (1881) ; Hollins v. Oklahoma, 295 U. S. 394 (1935) (per curiam); Avery v. Georgia, 345 U. S. 559 (1953) ; Hernandez v. Texas, 347 U. S. 475 (1954) ; Castaneda v. Partida, 430 U. S. 482 (1977) . To guard against discrimination in jury selection, the Court has ruled that no litigant may exclude a prospective juror on the basis of race. Batson v. Kentucky, 476 U. S. 79 (1986) ; Edmonson v. Leesville Concrete Co., 500 U. S. 614 (1991) ; Georgia v. McCollum, 505 U. S. 42 (1992) . In an effort to ensure that individuals who sit on juries are free of racial bias, the Court has held that the Constitution at times demands that defendants be permitted to ask questions about racial bias during voir dire. Ham v. South Carolina, 409 U. S. 524 (1973) ; Rosales-Lopez, 451 U. S. 182 ; Turner v. Murray, 476 U. S. 28 (1986) . The unmistakable principle underlying these precedents is that discrimination on the basis of race, “odious in all aspects, is especially pernicious in the administration of justice.” Rose v. Mitchell, 443 U. S. 545, 555 (1979) . The jury is to be “a criminal defendant’s fundamental ‘protection of life and liberty against race or color prejudice.’ ” McCleskey v. Kemp, 481 U. S. 279, 310 (1987) (quoting Strauder, supra, at 309). Permitting racial prejudice in the jury system damages “both the fact and the perception” of the jury’s role as “a vital check against the wrongful exercise of power by the State.” Powers v. Ohio, 499 U. S. 400, 411 (1991) ; cf. Aldridge v. United States, 283 U. S. 308, 315 (1931) ; Buck v. Davis, ante, at 22. IV A This case lies at the intersection of the Court’s decisions endorsing the no-impeachment rule and its decisions seeking to eliminate racial bias in the jury system. The two lines of precedent, however, need not conflict. Racial bias of the kind alleged in this case differs in critical ways from the compromise verdict in McDonald, the drug and alcohol abuse in Tanner, or the pro-defendant bias in Warger. The behavior in those cases is troubling and unacceptable, but each involved anomalous behavior from a single jury—or juror—gone off course. Jurors are presumed to follow their oath, cf. Penry v. Johnson, 532 U. S. 782, 799 (2001) , and neither history nor common experience show that the jury system is rife with mischief of these or similar kinds. To attempt to rid the jury of every irregularity of this sort would be to expose it to unrelenting scrutiny. “It is not at all clear . . . that the jury system could survive such efforts to perfect it.” Tanner, 483 U. S., at 120. The same cannot be said about racial bias, a familiar and recurring evil that, if left unaddressed, would risk systemic injury to the administration of justice. This Court’s decisions demonstrate that racial bias implicates unique historical, constitutional, and institutional concerns. An effort to address the most grave and serious statements of racial bias is not an effort to perfect thejury but to ensure that our legal system remains capable of coming ever closer to the promise of equal treat-ment under the law that is so central to a functioning democracy. Racial bias is distinct in a pragmatic sense as well. In past cases this Court has relied on other safeguards to protect the right to an impartial jury. Some of those safeguards, to be sure, can disclose racial bias. Voir dire at the outset of trial, observation of juror demeanor and conduct during trial, juror reports before the verdict, and nonjuror evidence after trial are important mechanisms for discovering bias. Yet their operation may be compromised, or they may prove insufficient. For instance, this Court has noted the dilemma faced by trial court judges and counsel in deciding whether to explore potential racial bias at voir dire. See Rosales-Lopez, supra; Ristaino v. Ross, 424 U. S. 589 (1976) . Generic questions about juror impartiality may not expose specific attitudes or biases that can poison jury deliberations. Yet more pointed questions “could well exacerbate whatever prejudice might exist without substantially aiding in exposing it.” Rosales-Lopez, supra, at 195 (Rehnquist, J., concurring in result). The stigma that attends racial bias may make it difficult for a juror to report inappropriate statements during the course of juror deliberations. It is one thing to accuse a fellow juror of having a personal experience that improperly influences her consideration of the case, as would have been required in Warger. It is quite another to call her a bigot. The recognition that certain of the Tanner safeguards may be less effective in rooting out racial bias than other kinds of bias is not dispositive. All forms of improper bias pose challenges to the trial process. But there is a sound basis to treat racial bias with added precaution. A constitutional rule that racial bias in the justice system must be addressed—including, in some instances, after the verdict has been entered—is necessary to prevent a systemic loss of confidence in jury verdicts, a confidence that is a central premise of the Sixth Amendment trial right. B For the reasons explained above, the Court now holds that where a juror makes a clear statement that indicates he or she relied on racial stereotypes or animus to convict a criminal defendant, the Sixth Amendment requires that the no-impeachment rule give way in order to permit the trial court to consider the evidence of the juror’s statement and any resulting denial of the jury trial guarantee. Not every offhand comment indicating racial bias or hostility will justify setting aside the no-impeachment bar to allow further judicial inquiry. For the inquiry to proceed, there must be a showing that one or more jurors made statements exhibiting overt racial bias that cast serious doubt on the fairness and impartiality of the jury’s deliberations and resulting verdict. To qualify, the statement must tend to show that racial animus was a significant motivating factor in the juror’s vote to convict. Whether that threshold showing has been satisfied is a matter committed to the substantial discretion of the trial court in light of all the circumstances, including the content and timing of the alleged statements and the reliability of the proffered evidence. The practical mechanics of acquiring and presenting such evidence will no doubt be shaped and guided by state rules of professional ethics and local court rules, both of which often limit counsel’s post-trial contact with jurors. See 27 C. Wright & V. Gold, Federal Practice and Procedure: Evidence §6076, pp. 580–583 (2d ed. 2007) (Wright); see also Variations of ABA Model Rules of Professional Conduct, Rule 3.5 (Sept. 15, 2016) (overview of state ethics rules); 2 Jurywork Systematic Techniques §13:18 (2016–2017) (overview of Federal District Court rules). These limits seek to provide jurors some protection when they return to their daily affairs after the verdict has been entered. But while a juror can always tell counsel they do not wish to discuss the case, jurors in some instances may come forward of their own accord. That is what happened here. In this case the alleged statements by a juror were egregious and unmistakable in their reliance on racial bias. Not only did juror H. C. deploy a dangerous racial stereotype to conclude petitioner was guilty and his alibi witness should not be believed, but he also encouraged other jurors to join him in convicting on that basis. Petitioner’s counsel did not seek out the two jurors’ allegations of racial bias. Pursuant to Colorado’s mandatory jury instruction, the trial court had set limits on juror contact and encouraged jurors to inform the court if anyone harassed them about their role in the case. Similar limits on juror contact can be found in other jurisdictions that recognize a racial-bias exception. See, e.g., Fla. Standard Jury Instrs. in Crim. Cases No. 4.2 (West 2016) (“Although you are at liberty to speak with anyone about your deliberations, you are also at liberty to refuse to speak to anyone”); Mass. Office of Jury Comm’r, Trial Juror’s Handbook (Dec. 2015) (“You are not required to speak with anyone once the trial is over. . . . If anyone tries to learn this confidential information from you, or if you feel harassed or embarrassed in any way, you should report it to the court . . . immediately”); N. J. Crim. Model Jury Charges, Non 2C Charges, Dismissal of Jury (2014) (“It will be up to each of you to decide whether to speak about your service as a juror”). With the understanding that they were under no obligation to speak out, the jurors approached petitioner’s counsel, within a short time after the verdict, to relay their concerns about H. C.’s statements. App. 77. A similar pattern is common in cases involving juror allegations of racial bias. See, e.g., Villar, 586 F. 3d, at 78 (juror e-mailed defense counsel within hours of the verdict); Kittle v. United States, 65 A. 3d 1144, 1147 (D. C. 2013) (juror wrote a letter to the judge the same day the court discharged the jury); Benally, 546 F. 3d, at 1231 (juror approached defense counsel the day after the jury announced its verdict). Pursuant to local court rules, petitioner’s counsel then sought and received permission from the court to contact the two jurors and obtain affidavits limited to recounting the exact statements made by H. C. that exhibited racial bias. While the trial court concluded that Colorado’s Rule 606(b) did not permit it even to consider the resulting affidavits, the Court’s holding today removes that bar. When jurors disclose an instance of racial bias as serious as the one involved in this case, the law must not wholly disregard its occurrence. C As the preceding discussion makes clear, the Court relies on the experiences of the 17 jurisdictions that have recognized a racial-bias exception to the no-impeachment rule—some for over half a century—with no signs of an increase in juror harassment or a loss of juror willingness to engage in searching and candid deliberations. The experience of these jurisdictions, and the experience of the courts going forward, will inform the proper exercise of trial judge discretion in these and related matters. This case does not ask, and the Court need not address, what procedures a trial court must follow when confronted with a motion for a new trial based on juror testimony of racial bias. See 27 Wright 575–578 (noting a divergence of authority over the necessity and scope of an evidentiary hearing on alleged juror misconduct). The Court also does not decide the appropriate standard for determining when evidence of racial bias is sufficient to require that the verdict be set aside and a new trial be granted. Compare, e.g., Shillcutt, 827 F. 2d, at 1159 (inquiring whether racial bias “pervaded the jury room”), with, e.g., Henley, 238 F. 3d, at 1120 (“One racist juror would be enough”). D It is proper to observe as well that there are standard and existing processes designed to prevent racial bias in jury deliberations. The advantages of careful voir dire have already been noted. And other safeguards deserve mention. Trial courts, often at the outset of the case and again in their final jury instructions, explain the jurors’ duty to review the evidence and reach a verdict in a fair and impartial way, free from bias of any kind. Some instructions are framed by trial judges based on their own learning and experience. Model jury instructions likely take into account these continuing developments and are common across jurisdictions. See, e.g., 1A K. O’Malley, J. Grenig, & W. Lee, Federal Jury Practice and Instructions, Criminal §10:01, p. 22 (6th ed. 2008) (“Perform these duties fairly. Do not let any bias, sympathy or prejudice that you may feel toward one side or the other influence your decision in any way”). Instructions may emphasize the group dynamic of deliberations by urging jurors to share their questions and conclusions with their colleagues. See, e.g., id., §20:01, at 841 (“It is your duty as jurors to consult with one another and to deliberate with one another with a view towards reaching an agreement if you can do so without violence to individual judgment”). Probing and thoughtful deliberation improves the likelihood that other jurors can confront the flawed nature of reasoning that is prompted or influenced by improper biases, whether racial or otherwise. These dynamics can help ensure that the exception is limited to rare cases. * * * The Nation must continue to make strides to overcome race-based discrimination. The progress that has already been made underlies the Court’s insistence that blatant racial prejudice is antithetical to the functioning of the jury system and must be confronted in egregious cases like this one despite the general bar of the no-impeachment rule. It is the mark of a maturing legal system that it seeks to understand and to implement the lessons of history. The Court now seeks to strengthen the broader principle that society can and must move forward by achieving the thoughtful, rational dialogue at the foundation of both the jury system and the free society that sustains our Constitution. The judgment of the Supreme Court of Colorado is reversed, and the case is remanded for further proceedings not inconsistent with this opinion. It is so ordered. APPENDIX Codified Exceptions in Addition to Those Enumerated in Fed. Rule Evid. 606(b) See Ariz. Rules Crim. Proc. 24.1(c)(3), (d) (2011) (exception for evidence of misconduct, including verdict by game of chance or intoxication); Idaho Rule Evid. 606(b) (2016) (game of chance); Ind. Rule Evid. 606(b)(2)(A) (Burns 2014) (drug or alcohol use); Minn. Rule Evid. 606(b) (2014) (threats of violence or violent acts); Mont. Rule Evid. 606(b) (2015) (game of chance); N. D. Rule Evid. 606(b)(2)(C) (2016–2017) (same); Tenn. Rule Evid. 606(b) (2016) (quotient verdict or game of chance); Tex. Rule Evid. 606(b)(2)(B) (West 2016) (rebutting claim juror was unqualified); Vt. Rule Evid. 606(b) (Cum. Supp. 2016) (juror communication with nonjuror); see also 27 C. Wright & V. Gold, Federal Practice and Procedure: Evidence §6071, p. 447, and n. 66 (2d ed. 2007); id., at 451, and n. 70; id., at 452, and n. 72. Judicially Recognized Exceptions for Evidence of Racial Bias See State v. Santiago, 245 Conn. 301, 323–340, 715 A. 2d 1, 14–22 (1998); Kittle v. United States, 65 A. 3d 1144, 1154–1556 (D. C. 2013); Fisher v. State, 690 A. 2d 917, 919–921, and n. 4 (Del. 1996) (Appendix to opinion), Powell v. Allstate Ins. Co., 652 So. 2d 354, 357–358 (Fla. 1995); Spencer v. State, 260 Ga. 640, 643–644, 398 S. E. 2d 179, 184–185 (1990); State v. Jackson, 81 Haw. 39, 48–49, 912 P. 2d 71, 80–81 (1996); Commonwealth v. Laguer, 410 Mass. 89, 97–98, 571 N. E. 2d 371, 376 (1991); State v. Callender, 297 N. W. 2d 744, 746 (Minn. 1980); Fleshner v. Pepose Vision Inst., P. C., 304 S. W. 3d 81, 87–90 (Mo. 2010); State v. Levitt, 36 N. J. 266, 271–273, 176 A. 2d 465, 467–468 (1961); People v. Rukaj, 123 App. Div. 2d 277, 280–281, 506 N. Y. S. 2d 677, 679–680 (1986); State v. Hidanovic, 2008 ND 66, ¶¶21–26, 747 N. W. 2d 463, 472–474; State v. Brown, 62 A. 3d 1099, 1110 (R. I. 2013); State v. Hunter, 320 S. C. 85, 88, 463 S. E. 2d 314, 316 (1995); Seattle v. Jackson, 70 Wash. 2d 733, 738, 425 P. 2d 385, 389 (1967); After Hour Welding, Inc. v. Laneil Management Co., 108 Wis. 2d 734, 739–740, 324 N. W. 2d 686, 690 (1982).

582.US.2016_16-399

Under the Civil Service Reform Act of 1978 (CSRA), the Merit Systems Protection Board (MSPB or Board) has the power to review certain serious personnel actions against federal employees. If an employee asserts rights under the CSRA only, MSPB decisions are subject to judicial review exclusively in the Federal Circuit. 5 U. S. C. §7703(b)(1). If the employee invokes only federal antidiscrimination law, the proper forum for judicial review is federal district court. See Kloeckner v. Solis, 568 U. S. 41 . An employee who complains of a serious adverse employment action and attributes the action, in whole or in part, to bias based on race, gender, age, or disability brings a “mixed case.” When the MSPB dismisses a mixed case on the merits or on procedural grounds, review authority lies in district court, not the Federal Circuit. Id., at 50, 56. This case concerns the proper forum for judicial review when the MSPB dismisses such a case for lack of jurisdiction. Anthony Perry received notice that he would be terminated from his employment at the U. S. Census Bureau for spotty attendance. Perry and the Bureau reached a settlement in which Perry agreed to a 30-day suspension and early retirement. The settlement also required Perry to dismiss discrimination claims he had filed separately with the Equal Employment Opportunity Commission (EEOC). After retiring, Perry appealed his suspension and retirement to the MSPB, alleging discrimination based on race, age, and disability, as well as retaliation by the Bureau for his prior discrimination complaints. The settlement, he maintained, did not stand in the way, because the Bureau had coerced him into signing it. But an MSPB administrative law judge (ALJ) determined that Perry had failed to prove that the settlement was coerced. Presuming Perry’s retirement to be voluntary, the ALJ dismissed his case. Because voluntary actions are not appealable to the MSPB, the ALJ observed, the Board lacked jurisdiction to entertain Perry’s claims. The MSPB affirmed, deeming Perry’s separation voluntary and therefore not subject to the Board’s jurisdiction. If dissatisfied with the MSPB’s ruling, the Board stated, Perry could seek judicial review in the Federal Circuit. Perry instead sought review in the D. C. Circuit, which, the parties later agreed, lacked jurisdiction. The D. C. Circuit held that the proper forum was the Federal Circuit and transferred the case there. Kloeckner did not control, the court concluded, because it addressed dismissals on procedural grounds, not jurisdictional grounds. Held: The proper review forum when the MSPB dismisses a mixed case on jurisdictional grounds is district court. Pp. 9–17. (a) The Government argues that employees must split their mixed claims, appealing MSPB nonappealability rulings to the Federal Circuit while repairing to the district court to adjudicate their discrimination claims. Perry counters that the district court alone can resolve his entire complaint. Perry advances the more sensible reading of the statutory prescriptions. Kloeckner announced a clear rule: “[M]ixed cases shall be filed in district court.” 568 U. S., at 50; see id., at 56. The key to district court review is the employee’s “clai[m] that an agency action appealable to the MSPB violates an antidiscrimination statute listed in §7702(a)(1).” Id., at 56 (emphasis added). Such a nonfrivolous allegation of jurisdiction suffices to establish district court jurisdiction. EEOC regulations are in accord, and several Courts of Appeals have similarly described mixed-case appeals as those alleging an adverse action subject to MSPB jurisdiction taken, in whole or in part, because of unlawful discrimination. Perry, who “complain[ed] of a personnel action serious enough to appeal to the MSPB” and “allege[d] that the [personnel] action was based on discrimination,” brought a mixed case, and district court jurisdiction was therefore proper. Pp. 9–12. (b) The Government’s proposed distinction—between MSPB merits and procedural decisions, on the one hand, and the Board’s jurisdictional rulings, on the other—has multiple infirmities. Had Congress wanted to bifurcate judicial review, sending merits and procedural decisions to district court and jurisdictional dismissals to the Federal Circuit, it could have said so. See Kloeckner, 568 U. S., at 52. The Government’s newly devised attempt to distinguish jurisdictional dismissals from procedural dismissals is a departure from its position in Kloeckner. Such a distinction, as both parties recognized in Kloeckner, would be perplexing and elusive. The distinction between jurisdiction and the merits is also not inevitably sharp, for the two inquiries may overlap. And because the MSPB may issue rulings on alternate or multiple grounds, some “jurisdictional,” others procedural or substantive, allocating judicial review authority based on a separate rule for jurisdictional rulings may prove unworkable in practice. Perry’s comprehension of the complex statutory text, in contrast, serves “[t]he CSRA’s objective of creating an integrated scheme of review[, which] would be seriously undermined” by “parallel litigation regarding the same agency action.” Elgin v. Department of Treasury, 567 U. S. 1 . Pp. 12–17. 829 F. 3d 760, reversed and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Alito, Sotomayor, and Kagan, JJ., joined. Gorsuch, J., filed a dissenting opinion, in which Thomas, J., joined.

This case concerns the proper forum for judicial review when a federal employee complains of a serious adverse employment action taken against him, one falling within the compass of the Civil Service Reform Act of 1978 (CSRA), 5 U. S. C. §1101 et seq., and attributes the action, in whole or in part, to bias based on race, gender, age, or disability, in violation of federal antidiscrimination laws. We refer to complaints of that order, descriptively, as “mixed cases.” In the CSRA, Congress created the Merit Systems Protection Board (MSPB or Board) to review certain serious personnel actions against federal employees. If an employee asserts rights under the CSRA only, MSPB decisions, all agree, are subject to judicial review exclusively in the Federal Circuit. §7703(b)(1). If the employee asserts no civil-service rights, invoking only federal antidiscrimination law, the proper forum for judicial review, again all agree, is a federal district court, see Kloeckner v. Solis, 568 U. S. 41, 46 (2012) ; the Federal Circuit, while empowered to review MSPB decisions on civil-service claims, §7703(b)(1)(A), lacks authority over claims arising under antidiscrimination laws, see §7703(c). When a complaint presents a mixed case, and the MSPB dismisses it, must the employee resort to the Federal Circuit for review of any civil-service issue, reserving claims under federal antidiscrimination law for discrete district court adjudication? If the MSPB dismisses a mixed case on the merits, the parties agree, review authority lies in district court, not in the Federal Circuit. In Kloeckner, 568 U. S., at 50, 56, we held, the proper review forum is also the district court when the MSPB dismisses a mixed case on procedural grounds, in Kloeckner itself, failure to meet a deadline for Board review set by the MSPB. We hold today that the review route remains the same when the MSPB types its dismissal of a mixed case as “jurisdictional.” As in Kloeckner, we are mindful that review rights should be read not to protract proceedings, increase costs, and stymie employees,[1] but to secure expeditious resolution of the claims employees present. See Elgin v. Department of Treasury, 567 U. S. 1, 15 (2012) (emphasizing need for “clear guidance about the proper forum for [an] employee’s [CSRA] claims”). Cf. Fed. Rule Civ. Proc. l. I A The CSRA “establishes a framework for evaluating personnel actions taken against federal employees.” Kloeckner v. Solis, 568 U. S. 41, 44 (2012) . For “particularly serious” actions, “for example, a removal from employment or a reduction in grade or pay,” “the affected employee has a right to appeal the agency’s decision to the MSPB.” Ibid. (citing §§1204, 7512, 7701). Such an appeal may present a civil-service claim only. Typically, the employee may allege that “the agency had insufficient cause for taking the action under the CSRA.” Id., at 44. An appeal to the MSPB, however, may also complain of adverse action taken, in whole or in part, because of discrimination prohibited by another federal statute, for example, Title VII of the Civil Rights Act of 1964, 42 U. S. C. §2000e et seq., or the Age Discrimination in Employment Act of 1967, 29 U. S. C. §621 et seq. See 5 U. S. C. §7702(a)(1); Kloeckner, 568 U. S., at 44. In Kloeckner, we explained, “[w]hen an employee complains of a personnel action serious enough to appeal to the MSPB and alleges that the action was based on discrimination, she is said (by pertinent regulation) to have brought a ‘mixed case.’ ” Ibid. (quoting 29 CFR §1614.302 (2012)). See also §1614.302(a)(2) (2016) (defining “mixed case appeal” as one in which an employee “alleges that an appealable agency action was effected, in whole or in part, because of discrimination”). For mixed cases, “[t]he CSRA and regulations of the MSPB and Equal Employment Opportunity Commission (EEOC) set out special procedures . . . different from those used when the employee either challenges a serious personnel action under the CSRA alone or attacks a less serious action as discriminatory.” Kloeckner, 568 U. S., at 44–45. As Kloeckner detailed, the CSRA provides diverse procedural routes for an employee’s pursuit of a mixed case. The employee “may first file a discrimination complaint with the agency itself,” in the agency’s equal employment opportunity (EEO) office, “much as an employee challenging a personnel practice not appealable to the MSPB could do.” Id., at 45 (citing 5 CFR §1201.154(a) (2012); 29 CFR §1614.302(b) (2012)); see §7702(a)(2). “If the agency [EEO office] decides against her, the employee may then either take the matter to the MSPB or bypass further administrative review by suing the agency in district court.” Kloeckner, 568 U. S., at 45 (citing 5 CFR §1201.154(b); 29 CFR §1614.302(d)(1)(i)); see §7702(a)(2). “Alternatively, the employee may initiate the process by bringing her case directly to the MSPB, forgoing the agency’s own system for evaluating discrimination charges.” Kloeckner, 568 U. S., at 45 (citing 5 CFR §1201.154(a); 29 CFR §1614.302(b)); see §7702(a)(1). Section 7702 prescribes appellate proceedings in actions involving discrimination. Defining the MSPB’s jurisdiction in mixed-case appeals that bypass an agency’s EEO office, §7702(a)(1) states in relevant part: “[I]n the case of any employee . . . who— “(A) has been affected by an action which the employee . . . may appeal to the [MSPB], and “(B) alleges that a basis for the action was discrimination prohibited by [specified antidiscrimination statutes], . . . “the Board shall, within 120 days of the filing of the appeal, decide both the issue of discrimination and the appealable action in accordance with the Board’s appellate procedures . . . .”[2] Section 7702(a)(2) similarly authorizes a mixed-case appeal to the MSPB from an agency EEO office’s decision. Then, “[i]f the MSPB upholds the personnel action (whether in the first instance or after the agency has done so), the employee again has a choice: She may request additional administrative process, this time with the EEOC, or else she may seek judicial review.” Kloeckner, 568 U. S., at 45 (citing §7702(a)(3), (b); 5 CFR §1201.161; 29 CFR §1614.303). Section 7703(b) designates the proper forum for judicial review of MSPB decisions. Section 7703(b)(1)(A) provides the general rule: “[A] petition to review a . . . final decision of the Board shall be filed in the United States Court of Appeals for the Federal Circuit.” Section 7703(b)(2) states the exception here relevant, governing “[c]ases of discrimination subject to the provisions of [§]7702.” See Kloeckner, 568 U. S., at 46 (“The ‘cases of discrimination’ in §7703(b)(2)’s exception . . . are mixed cases, in which an employee challenges as discriminatory a personnel action appealable to the MSPB.”). Such cases “shall be filed under [the enforcement sections of Title VII of the Civil Rights Act of 1964, the Age Discrimination in Employment Act, and the Fair Labor Standards Act of 1938, 29 U. S. C. §201 et seq.], as applicable.” §7703(b)(2). Those enforcement provisions “all authorize suit in federal district court.” Kloeckner, 568 U. S., at 46 (citing, inter alia, 42 U. S. C. §§2000e–16(c), 2000e–5(f); 29 U. S. C. §633a(c); §216(b)). Thus, if the MSPB decides against the employee on the merits of a mixed case, the statute instructs her to seek review in federal district court under the enforcement provision of the relevant antidiscrimination laws. §7703(b)(2); see Kloeckner, 568 U. S., at 56, n. 4.[3] Federal district court is also the proper forum for judicial review, we held in Kloeckner, when the MSPB dismissesa mixed case on procedural grounds. Id., at 50, 56. We rested that conclusion on this syllogism: “Under §7703(b)(2), ‘cases of discrimination subject to [§7702]’ shall be filed in district court.” Id., at 50 (alteration in original). Further, “[u]nder §7702(a)(1), [mixed cases qualify as] ‘cases of discrimination subject to [§7702].’ ” Ibid. (third alteration in original). Thus, “mixed cases shall be filed in district court.” Ibid. That syllogism, we held, holds true whether the dismissal rests on procedural grounds or on the merits, for “nowhere in the [CSRA’s] provisions on judicial review” is a distinction drawn between MSPB merits decisions and procedural rulings. Id., at 51. The instant case presents this question: Where doesan employee seek judicial review when the MSPB dis-misses her civil-service case alleging discrimination neither on the merits nor on a procedural ground, but for lack of jurisdiction? B Anthony Perry worked at the U. S. Census Bureau until 2012. 829 F. 3d 760, 762 (CADC 2016). In 2011, Perry received notice that he would be terminated because of spotty attendance. Ibid. Later that year, Perry and the Bureau reached a settlement in which Perry agreed to a 30-day suspension and early retirement. Ibid. The agreement required Perry to dismiss discrimination claims he had separately filed with the EEOC. Ibid. After retiring, Perry appealed his suspension and retirement to the MSPB. Ibid. He alleged discrimination on grounds of race, age, and disability, as well as retaliation by the Bureau for his prior discrimination complaints. Ibid. The settlement, he maintained, did not stand in the way, because the Bureau coerced him into signing it. Ibid. An MSPB administrative law judge (ALJ) eventually determined that Perry had failed to prove that the settlement was coerced. Perry v. Department of Commerce, No. DC–0752–12–0486–B–1 etc. (Dec. 23, 2013) (initial decision), App. to Pet. for Cert. 32a, 47a. Presuming Perry’s retirement to be voluntary, the ALJ dismissed his case. Id., at 33a, 47a. Voluntary actions are not appealable to the MSPB, the ALJ observed, hence, the ALJ concluded, the Board lacked jurisdiction to entertain Perry’s claims. Id., at 51a. The MSPB affirmed the ALJ’s decision. See Perry v. Department of Commerce, 2014 WL 5358308, *1 (Aug. 6, 2014) (final order). The settlement agreement, the Board recounted, provided that Perry would waive his Board appeal rights with respect to his suspension and retirement. Ibid. Because Perry did not prove that the agreement was involuntary, the Board determined (in accord with the ALJ) that his separation should be deemed voluntary, hence not an adverse action subject to the Board’s jurisdiction under §7702(a)(1). Id., at *3–*4. If dissatisfied with the MSPB’s ruling, the Board stated in its decision, Perry could seek judicial review in the Federal Circuit. Id., at *4. Perry instead filed a pro se petition for review in the D. C. Circuit. 829 F. 3d, at 763. The court ordered jurisdictional briefing and appointed counsel to argue forPerry. Ibid. By the time the court heard argument, the parties had agreed that the D. C. Circuit lacked jurisdiction, but disagreed on whether the proper forum for judicial review was the Federal Circuit, as the Government contended, or federal district court, as Perry maintained. Ibid. The D. C. Circuit held that the Federal Circuit had jurisdiction over Perry’s petition and transferred his case to that court under 28 U. S. C. §1631. 829 F. 3d, at 763. The court’s disposition was precedent-bound: In a prior decision, Powell v. Department of Defense, 158 F. 3d 597, 598 (1998), the D. C. Circuit had held that the Federal Circuit is the proper forum for judicial review of MSPB decisions dismissing mixed cases “on procedural or threshold grounds.” See 829 F. 3d, at 764, 767–768. Notably, Powell ranked as a “procedural or threshold matter” “the Board’s view of its jurisdiction.” 158 F. 3d, at 599 (internal quotation marks omitted). The D. C. Circuit rejected Perry’s argument that Powell was undermined by this Court’s intervening decision in Kloeckner, which held MSPB procedural dispositions of mixed cases reviewable in district court. 829 F. 3d, at 764–768. Kloeckner, the D. C. Circuit observed, repeatedly tied its decision to dismissals on “procedural grounds,” 568 U. S., at 44, 46, 49, 52, 54, 55. See 829 F. 3d, at 765. Jurisdictional dismissals differ from procedural dismissals, the D. C. Circuit concluded, given the CSRA’s reference to mixed cases as those “which the employee . . . may appeal to the [MSPB].” Id., at 766–767 (quoting §7702(a)(1)(A); emphasis added). A jurisdictional dismissal, the court said, rests on the Board’s determination that the employee may not appeal his case to the MSPB. Id., at 766–767. In contrast, a dismissal on procedural grounds, e.g., untimely resort to the MSPB, leaves the employee still “affected by an action which [she] may appeal tothe MSPB.” Ibid. (quoting §7702(a)(1)(A); alteration in original). We granted certiorari to review the D. C. Circuit’s decision, 580 U. S. ___ (2017), which accords with the Federal Circuit’s decision in Conforto v. Merit Systems Protection Bd., 713 F. 3d 1111 (2013). II Federal employees, the Government acknowledges, have a right to pursue claims of discrimination in violation of federal law in federal district court. Nor is there any doubt that the Federal Circuit lacks authority to adjudicate such claims. See §7703(c) (preserving “right to have the facts subject to trial de novo by the reviewing court” in any “case of discrimination” brought under §7703(b)(2)). The sole question here disputed: What procedural route may an employee in Perry’s situation take to gain judicial review of the MSPB’s jurisdictional disposition of a complaint that alleges adverse action taken under the CSRA in whole or in part due to discrimination proscribed by federal law? The Government argues, and the dissent agrees, that employees, situated as Perry is, must split their claims, appealing MSPB nonappealability rulings to the Federal Circuit while repairing to the district court for adjudication of their discrimination claims. As Perry sees it, one stop is all he need make. Exclusively competent to adjudicate “[c]ases of discrimination,” §7703(b)(2), the district court alone can resolve his entire complaint, Perry urges; the CSRA, he maintains, forces no bifurcation of his case. Section 7702(a)(1), the Government contends, marks a case as mixed only if the employee “has been affected by an action which the employee . . . may appeal to the [MSPB].” Brief for Respondent 15, 17–19, 21. An MSPB finding of nonappealability removes a case from that category, the Government asserts, and hence, from the purview of “[c]ases of discrimination” described in §7703(b)(2). Id., at 21. Only this reading of the CSRA’s provisions on judicial review—one ordering Federal Circuit review of any and all MSPB appealability determinations—the Government maintains, can ensure nationwide uniformity in answering questions arising under the CSRA. Id., at 26–32. Perry emphasizes in response that §7702(a)(1)(A)’s language, delineating cases in which an employee “has been affected by an action which the employee . . . may appeal to the [MSPB],” is not confined to cases an em-ployee may successfully appeal to the Board. Brief for Peti-tioner 19. The MSPB’s adverse ruling on the merits of his claim that the settlement was coerced, Perry argues, “did not retroactively divest the MSPB of jurisdiction to render that decision.” Id., at 21. The key consideration, according to Perry, is not what the MSPB determined about appealability; it is instead the nature of an employee’s claim that he had been “affected by an action [appealable] to the [MSPB]” (here, suspension for more than 14 days and involuntary removal, see §7512(1), (2)). See id., at 11, 23–24. Perry draws support for this argument from our recognition that “a party [may] establish jurisdiction at the outset of a case by means of a nonfrivolous assertion of jurisdictional elements,” Jerome B. Grubart, Inc. v. Great Lakes Dredge & Dock Co., 513 U. S. 527, 537 (1995) . See Brief for Petitioner 21–22. Perry, we hold, advances the more sensible reading of the statutory prescriptions. The Government’s procedure-jurisdiction distinction, we conclude, is no more tenable than “the merits-procedure distinction” we rejected in Kloeckner, 568 U. S., at 51. A As just noted, a nonfrivolous allegation of jurisdiction generally suffices to establish jurisdiction upon initiation of a case. See Jerome B. Grubart, Inc., 513 U. S., at 537. See also Bell v. Hood, 327 U. S. 678 –683 (1946) (To invoke federal-question jurisdiction, allegations in a complaint must simply be more than “insubstantial or frivolous,” and “[i]f the court does later exercise its jurisdiction to determine that the allegations in the complaint do not state a ground for relief, then dismissal of the case would be on the merits, not for want of jurisdiction.”). So too here: whether an employee “has been affected by anaction which [she] may appeal to the [MSPB],” §7702(a) (1)(A), turns on her well-pleaded allegations. Kloeckner, EEOC regulations, and Courts of Appeals’ decisions are corroborative. We announced a clear rule in Kloeckner: “[M]ixed cases shall be filed in district court.” 568 U. S., at 50. An employee brings a mixed case, we explained, when she “complains of a personnel action serious enough to appeal to the MSPB,” e.g., suspension for more than 14 days, §7512(2), “and alleges that the action was based on discrimination.” Id., at 44 (emphasis deleted). The key to district court review, we said, was the employee’s “clai[m] that an agency action appealable to the MSPB violates an antidiscrimination statute listed in §7702(a)(1).” Id., at 56 (emphasis added). EEOC regulations, see supra, at 3, are in accord: The defining feature of a “mixed case appeal,” those regulations instruct, is the employee’s “alleg[ation] that an appealable agency action was effected, in whole or in part, because of discrimination.” 29 CFR §1614.302(a)(2) (2016) (emphasis added). Several Courts of Appeals have similarly described mixed-case appeals as those alleging an adverse action subject to MSPB jurisdiction taken, in whole or in part, because of unlawful discrimination. See, e.g., Downey v. Runyon, 160 F. 3d 139, 143 (CA2 1998) (“Mixed appeals to the MSPB are those appeals alleging an appealable action affected in whole or in part by prohibited discrimination.” (emphasis added)); Powell, 158 F. 3d, at 597 (defining mixed-case appeal as “an appeal alleging both a Board-jurisdictional agency action and a claim of unlawful discrimination” (emphasis added)).See also Conforto, 713 F. 3d, at 1126–1127, n. 5 (Dyk, J., dissenting).[4] Because Perry “complain[ed] of a personnel action serious enough to appeal to the MSPB” (in his case, a 30-day suspension and involuntary removal, see supra, at 6; §7512(1), (2)) and “allege[d] that the [personnel] action was based on discrimination,” he brought a mixed case. Kloeckner, 568 U. S., at 44.[5] Judicial review of such a case lies in district court. Id., at 50, 56. B The Government rests heavily on a distinction between MSPB merits and procedural decisions, on the one hand, and the Board’s jurisdictional rulings, on the other.[6] The distinction has multiple infirmities. “If Congress had wanted to [bifurcate judicial review,] send[ing] merits decisions to district court and procedural dismissals to the Federal Circuit,” we observed in Kloeckner, “it could just have said so.” Id., at 52. The same observation could be made about bifurcating judicial review here, sending the MSPB’s merits and procedural decisions to district court, but its jurisdictional dismissals to the Federal Circuit.[7] The Government’s attempt to separate jurisdictional dismissals from procedural dismissals is newly devised. In Kloeckner, the Government agreed with the employee that there was “no basis” for a procedure-jurisdiction distinction. Brief for Respondent, O. T. 2012, No. 11–184, p. 25, n. 3; see Reply to Brief in Opposition, O. T. 2012, No. 11–184, pp. 1–2 (stating employee’s agreement with the Government that procedural and jurisdictional dismissals should travel together). Issues of both kinds, the Government there urged, should go to the Federal Circuit. Drawing such a distinction, the Government observed, would be “difficult and unpredictable.” Brief in Opposition in Kloeckner, O. T. 2012, No. 11–184, p. 15 (internal quotation marks omitted). Now, in light of our holding in Kloeckner that procedural dismissals should go to district court, the Government has changed course, contending that MSPB procedural and jurisdictional dismissals should travel different paths.[8] A procedure-jurisdiction distinction for purposes of determining the court in which judicial review lies, as both parties recognized in Kloeckner, would be perplexing and elusive. If a 30-day suspension followed by termination becomes nonappealable to the MSPB when the Board credits a release signed by the employee, one may ask why a determination that the employee complained of such adverse actions (suspension and termination) too late, i.e., after a Board-set deadline, does not similarly render the complaint nonappealable. In both situations, the Board disassociates itself from the case upon making a threshold determination. This Court, like others, we note, has sometimes wrestled over the proper characterization of timeliness questions. Compare Bowles v. Russell, 551 U. S. 205 –211, 215 (2007) (timely filing of notice of appeal in civil cases is “jurisdictional”), with id., at 217–219 (Souter, J., dissenting) (timeliness of notice of appeal is a procedural issue). Just as the proper characterization of a question as jurisdictional rather than procedural can be slippery, the distinction between jurisdictional and merits issues is not inevitably sharp, for the two inquiries may overlap. See Shoaf v. Department of Agriculture, 260 F. 3d 1336, 1341 (CA Fed. 2001) (“recogniz[ing] that the MSPB’s jurisdiction and the merits of an alleged involuntary separation are inextricably intertwined” (internal quotation marks omitted)). This case fits that bill. The MSPB determined that it lacked jurisdiction over Perry’s civil-service claims on the ground that he voluntarily released those claimsby entering into a valid settlement with his employing agency, the Census Bureau. See App. to Pet. for Cert. 27a.[9] But the validity of the settlement is at the heart of the dispute on the merits of Perry’s complaint. In essence, the MSPB ruled that it lacked jurisdiction because Perry’s claims fail on the merits. See Shoaf, 260 F. 3d, at 1341 (If it is established that an employee’s “resignation or retirement was involuntary and thus tantamount to forced removal,” then “not only [does the Board] ha[ve] jurisdiction, but also the employee wins on the merits and is entitled to reinstatement.” (internal quotation marks omitted)). See also Conforto, 713 F. 3d, at 1126 (Dyk, J., dissenting) (“[I]t cannot be that [the Federal Circuit] lack[s] jurisdiction to review the ‘merits’ of mixed cases but nevertheless may review ‘jurisdictional’ issues that are identical to the merits . . . .”).[10] Distinguishing between MSPB jurisdictional rulings and the Board’s procedural or substantive rulings for purposes of allocating judicial review authority between district court and the Federal Circuit is problematic for a further reason: In practice, the distinction may be unworkable. The MSPB sometimes rules on alternate grounds, one typed “jurisdictional,” another either procedural or substantive. See, e.g., Davenport v. Postal Service, 97 MSPR 417 (2004) (dismissing “for lack of jurisdiction and as untimely filed” (emphasis added)). To which court does appeal lie? Or, suppose that the Board addresses a complaint that encompasses multiple claims, dismissing some for want of jurisdiction, others on procedural or substantive grounds. See, e.g., Donahue v. Postal Service, 2006 WL 859448, *1, *3 (ED Pa., Mar. 31, 2006). Tellingly, the Government is silent on the proper channeling of appeals in such cases. Desirable as national uniformity may be,[11] it should not override the expense, delay, and inconvenience of requiring employees to sever inextricably related claims, resorting to two discrete appellate forums, in order to safeguard their rights. Perry’s comprehension of the complex statutory text, we are persuaded, best serves “[t]he CSRA’s objective of creating an integrated scheme of review[, which] would be seriously undermined” by “parallel litigation regarding the same agency action.” Elgin, 567 U. S., at 14. See also United States v. Fausto, 484 U. S. 439 –445 (1988).[12] Perry asks us not to “tweak” the statute, see post, at 1, but to read it sensibly, i.e., to refrain from reading into it the appeal-splitting bifurcation sought by the Government. Accordingly, we hold: (1) the Federal Circuit is the proper review forum when the MSPB disposes of complaints arising solely under the CSRA; and (2) in mixed cases, such as Perry’s, in which the employee (or former employee) complains of serious adverse action prompted, in whole or in part, by the employing agency’s violation of federal antidiscrimination laws, the district court is the proper forum for judicial review. * * * For the reasons stated, the judgment of the United States Court of Appeals for the District of Columbia Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Many CSRA claimants proceed pro se. See MSPB, Congressional Budget Justification FY 2017, p. 14 (2016) (“Generally, at least half or more of the appeals filed with the [MSPB] are from pro se appellants . . . .”). 2 If the MSPB fails to render a “judicially reviewable action” within 120 days, an employee may, “at any time after . . . the 120th day,” “file a civil action [in district court] to the same extent and in the same manner as provided in” the federal antidiscrimination laws invoked by the employee. §7702(e)(1). 3 Our decision in Kloeckner v. Solis, 568 U. S. 41 (2012) , did not merely assume that the civil-service component of mixed cases travels to district court. See id., at 56, n. 4 (“If the MSPB rejects on the merits a complaint alleging that an agency violated the CSRA as well as an antidiscrimination law, the suit will come to district court for a decision on both questions.” (emphasis added)). But see post, at 9–10. Characteristic of “mixed cases,” the employee in Kloeckner complained of adverse action taken, at least in part, because of discrimination. See 568 U. S., at 47. The Board dismissed that case, not for any flaw under antidiscrimination law, but because the employee missed a deadline set by the MSPB. See id., at 47–48. 4 Our interpretation is also consistent with another CSRA provision, §7513(d), which provides that “[a]n employee against whom an action is taken under this section is entitled to appeal to the . . . Board.” Because the “entitle[ment] to appeal” conferred in §7513(d) must be determined before an appeal is filed, such a right cannot depend on the outcome of the appeal. 5 If, as the dissent and the Government argue, see post, at 8–10; Brief for Respondent 19–26, 33–35, Perry’s case is not “mixed,” one can only wonder what kind of case it is, surely not one asserting rights under the CSRA only, or one invoking only antidiscrimination law. See supra, at 1–2. This is, of course, a paradigm mixed case: Perry alleges serious personnel actions (suspension and forced retirement) caused in whole or in part by prohibited discrimination. So did the employee in Kloeckner. She alleged that her firing (a serious personnel action) was based on discrimination. See 568 U. S., at 47. Thus Perry, like Kloeckner, well understood what the term “mixed case” means. 6 Notably, the dissent ventures no support for the principal argument made by the Government, i.e., that MSPB jurisdictional dispositions belong in the Federal Circuit, procedural and merits dispositions, in district court. 7 As Judge Dyk, dissenting in Conforto v. Merit Systems Protection Bd., 713 F. 3d 1111 (CA Fed. 2013), pointed out: “[W]here Congress intended to distinguish between different types of Board decisions, it did so expressly.” Id., at 1124, n. 1 (citing §3330b(b) (“An election under this section may not be made . . . after the [MSPB] has issued a judicially reviewable decision on the merits of the appeal.” (emphasis added)); §7703(a)(2) (“The Board shall be named respondent in any proceeding brought pursuant to this subsection, unless the employee . . . seeks review of a final order or decision on the merits . . . .” (emphasis added))). 8 This is not the first time the Government has changed its position. Before the Federal Circuit in Ballentine v. Merit Systems Protection Bd., 738 F. 2d 1244 (1984), the Government moved to transfer to district court an appeal challenging a jurisdictional dismissal by the MSPB. See id., at 1245. The Government argued that “even a question of the Board’s jurisdiction to hear an attempted mixed case appeal must be addressed by a district court.” Id., at 1247 (internal quotation marks omitted). Rejecting the Government’s position, the Federal Circuit concluded that it could review MSPB decisions on “procedural or threshold matters, not related to the merits of a discrimination claim.” Ibid. In Kloeckner, we disapproved the Federal Circuit’s holding with respect to MSPB procedural dismissals. 568 U. S., at 50, 56. Today we disapprove Ballentine’s holding with respect to jurisdictional dismissals, thereby adopting precisely the position advanced by the Government in that case. 9 In civil litigation, a release is an affirmative defense to a plaintiff’s claim for relief, not something the plaintiff must anticipate and negate in her pleading. See Fed. Rule Civ. Proc. 8(c)(1) (listing among affirmative defenses “release” and “waiver”); Newton v. Rumery, 480 U. S. 386, 391 (1987) . In that light, the MSPB’s jurisdiction should be determined by the adverse actions Perry asserts, suspension and forced retirement; the settlement releasing Perry’s claims would figure as a defense to his complaint, it would not enter into the determination whether the Board has jurisdiction over his claims. 10 If a reviewing court “agree[d] with the Board’s assessment,” then Perry would indeed have “lost his chance to pursue his . . . discrimination claim[s],” post, at 3, for those claims would have been defeated had he voluntarily submitted to the agency’s action. 11 In Kloeckner, we rejected the Government’s national uniformity argument. See 568 U. S., at 55–56, n. 4. “When Congress passed the CSRA, the Federal Circuit did not exist,” we observed, so uniformity did not then figure in Congress’ calculus. Id., at 56, n. 4. Moreover, even under the Government’s reading, “many cases involving federal employment issues [would be resolved] in district court. If the MSPB rejects on the merits a complaint alleging that an agency violated the CSRA as well as an antidiscrimination law, the suit will come to district court for a decision on both questions.” Ibid. 12 In both Elgin v. Department of Treasury, 567 U. S. 1 (2012) , and United States v. Fausto, 484 U. S. 439 (1988) , we rejected employees’ attempts to divide particular issues or claims among review forums. In Elgin, a federal employee opted not to seek review of an MSPB ALJ’s decision, either before the full Board or in the Federal Circuit; he instead brought in District Court, in the first instance, a constitutional challenge to an agency personnel action. 567 U. S., at 7–8. We concluded that an employee with civil-service claims must follow the CSRA’s procedures and may not bring a standalone constitutional challenge in district court. Id., at 8. In Fausto, a federal employee with CSRA claims filed an action in the United States Claims Court under the Back Pay Act of 1966. 484 U. S., at 443. We determined that the employee could not bring his action under the Back Pay Act because the CSRA provided “the comprehensive and integrated review scheme.” See id., at 454. Contrary to the dissent’s suggestion, see post, at 10, neither case indicated that the Federal Circuit, as opposed to district court, is the preferred forum for judicial review of all CSRA claims. Rather, both decisions emphasized the benefits of an integrated review scheme and the problems associated with bifurcating consideration of a single matter in different forums. See 567 U. S., at 13–14; 484 U. S., at 444–445. It is the dissent’s insistence on bifurcated review, therefore, that “Elgin and Fausto warned against,” post, at 10.

580.US.2016_15-777

Section 289 of the Patent Act makes it unlawful to manufacture or sell an “article of manufacture” to which a patented design or a colorable imitation thereof has been applied and makes an infringer liable to the patent holder “to the extent of his total profit.” 35 U. S. C. §289. As relevant here, a jury found that various smartphones manufactured by petitioners (collectively, Samsung) infringed design patents owned by respondent Apple Inc. that covered a rectangular front face with rounded edges and a grid of colorful icons on a black screen. Apple was awarded $399 million in damages—Samsung’s entire profit from the sale of its infringing smartphones. The Federal Circuit affirmed the damages award, rejecting Samsung’s argument that damages should be limited because the relevant articles of manufacture were the front face or screen rather than the entire smartphone. The court reasoned that such a limit was not required because the components of Samsung’s smartphones were not sold separately to ordinary consumers and thus were not distinct articles of manufacture. Held: In the case of a multicomponent product, the relevant “article of manufacture” for arriving at a §289 damages award need not be the end product sold to the consumer but may be only a component of that product. Pp. 4–9. (a) The statutory text resolves the issue here. An “article of manufacture,” which is simply a thing made by hand or machine, encompasses both a product sold to a consumer and a component of that product. This reading is consistent with §171(a) of the Patent Act, which makes certain “design[s] for an article of manufacture” eligible for design patent protection, and which has been understood by the Patent Office and the courts to permit a design patent that extends to only a component of a multicomponent product, see, e.g., Ex parte Adams, 84 Off. Gaz. Pat. Office 311; Application of Zahn, 617 F. 2d 261, 268 (CCPA). This reading is also consistent with the Court’s reading of the term “manufacture” in §101, which makes “any new and useful . . . manufacture” eligible for utility patent protection. See Diamond v. Chakrabarty, 447 U. S. 303 . Pp. 4–7. (b) Because the term “article of manufacture” is broad enough to embrace both a product sold to a consumer and a component of that product, whether sold separately or not, the Federal Circuit’s narrower reading cannot be squared with §289’s text. Absent adequate briefing by the parties, this Court declines to resolve whether the relevant article of manufacture for each design patent at issue here is the smartphone or a particular smartphone component. Doing so is not necessary to resolve the question presented, and the Federal Circuit may address any remaining issues on remand. Pp. 7–8. 786 F. 3d 983, reversed and remanded. Sotomayor, J., delivered the opinion for a unanimous Court.

Section 289 of the Patent Act provides a damages rem-edy specific to design patent infringement. A person who manufactures or sells “any article of manufacture to which [a patented] design or colorable imitation has been applied shall be liable to the owner to the extent of his totalprofit.” 35 U. S. C. §289. In the case of a design for a single-component product, such as a dinner plate, the product is the “article of manufacture” to which the design has been applied. In the case of a design for a multicomponent product, such as a kitchen oven, identifying the “article of manufacture” to which the design has been applied is a more difficult task. This case involves the infringement of designs for smartphones. The United States Court of Appeals for the Federal Circuit identified the entire smartphone as the only permissible “article of manufacture” for the purpose of calculating §289 damages because consumers could not separately purchase components of the smartphones. The question before us is whether that reading is consistent with §289. We hold that it is not. I A The federal patent laws have long permitted those who invent designs for manufactured articles to patent their designs. See Patent Act of 1842, §3, 5Stat. 543–544. Patent protection is available for a “new, original and ornamental design for an article of manufacture.” 35 U. S. C. §171(a). A patentable design “gives a peculiar or distinctive appearance to the manufacture, or article to which it may be applied, or to which it gives form.” Gorham Co. v. White, 14 Wall. 511, 525 (1872). This Court has explained that a design patent is infringed “if, in the eye of an ordinary observer, giving such attention as a purchaser usually gives, two designs are substantially the same.” Id., at 528. In 1885, this Court limited the damages available for design patent infringement. The statute in effect at the time allowed a holder of a design patent to recover “the actual damages sustained” from infringement. Rev. Stat. §4919. In Dobson v. Hartford Carpet Co., 114 U. S. 439 (1885) , the lower courts had awarded the holders of design patents on carpets damages in the amount of “the entire profit to the [patent holders], per yard, in the manufacture and sale of carpets of the patented designs, and not merely the value which the designs contributed to the carpets.” Id., at 443. This Court reversed the damages award and construed the statute to require proof that the profits were “due to” the design rather than other aspects of the carpets. Id., at 444; see also Dobson v. Dornan, 118 U. S. 10, 17 (1886) (“The plaintiff must show what profits or damages are attributable to the use of the infringing design”). In 1887, in response to the Dobson cases, Congress enacted a specific damages remedy for design patent infringement. See S. Rep. No. 206, 49th Cong., 1st Sess., 1–2 (1886); H. R. Rep. No. 1966, 49th Cong., 1st Sess., 1–2 (1886). The new provision made it unlawful to manufacture or sell an article of manufacture to which a patented design or a colorable imitation thereof had been applied. An act to amend the law relating to patents, trademarks, and copyright, §1, 24Stat. 387. It went on to make a design patent infringer “liable in the amount of” $250 or “the total profit made by him from the manufacture or sale . . . of the article or articles to which the design, or color-able imitation thereof, has been applied.” Ibid. The Patent Act of 1952 codified this provision in §289. 66Stat. 813. That codified language now reads, in relevant part: “Whoever during the term of a patent for a design, without license of the owner, (1) applies the patented design, or any colorable imitation thereof, to any article of manufacture for the purpose of sale, or (2) sells or exposes for sale any article of manufacture to which such design or colorable imitation has been applied shall be liable to the owner to the extent of his total profit, but not less than $250 . . . .” 35 U. S. C. §289. B Apple Inc. released its first-generation iPhone in 2007. The iPhone is a smartphone, a “cell phone with a broad range of other functions based on advanced computing capability, large storage capacity, and Internet connectiv-ity.” Riley v. California, 573 U. S. ___, ___ (2014) (slip op., at 2). Apple secured many design patents in connection with the release. Among those patents were the D618,677 patent, covering a black rectangular front face with rounded corners, the D593,087 patent, covering a rectangular front face with rounded corners and a raised rim, and the D604,305 patent, covering a grid of 16 colorful icons on a black screen. App. 530–578. Samsung Electronics Co., Samsung Electronics America, Inc., and Samsung Telecommunications America, LLC (Samsung), also manufacture smartphones. After Apple released its iPhone, Samsung released a series of smartphones that resembled the iPhone. Id., at 357–358. Apple sued Samsung in 2011, alleging, as relevant here, that various Samsung smartphones infringed Apple’s D593,087, D618,677, and D604,305 design patents. A jury found that several Samsung smartphones did infringe those patents. See id., at 273–276. All told, Apple was awarded $399 million in damages for Samsung’s design patent infringement, the entire profit Samsung made from its sales of the infringing smartphones. See id., at 277–280, 348–350. The Federal Circuit affirmed the design patent infringement damages award.[1] In doing so, it rejected Samsung’s argument “that the profits awarded should have been limited to the infringing ‘article of manufacture’ ”—for example, the screen or case of the smartphone—“not the entire infringing product”—the smartphone. 786 F. 3d 983, 1002 (2015). It reasoned that “limit[ing] the dam-ages” award was not required because the “innards of Samsung’s smartphones were not sold separately from their shells as distinct articles of manufacture to ordinary purchasers.” Ibid. We granted certiorari, 577 U. S. ___ (2016), and now reverse and remand. II Section 289 allows a patent holder to recover the total profit an infringer makes from the infringement. It does so by first prohibiting the unlicensed “appli[cation]” of a “patented design, or any colorable imitation thereof, to any article of manufacture for the purpose of sale” or the unlicensed sale or exposure to sale of “any article of manufacture to which [a patented] design or colorable imitation has been applied.” 35 U. S. C. §289. It then makes a person who violates that prohibition “liable to the owner to the extent of his total profit, but not less than $250.” Ibid. “Total,” of course, means all. See American Heritage Dictionary 1836 (5th ed. 2011) (“[t]he whole amount of something; the entirety”). The “total profit” for which §289 makes an infringer liable is thus all of the profit made from the prohibited conduct, that is, from the manufacture or sale of the “article of manufacture to which [the patented] design or colorable imitation has been applied.” Arriving at a damages award under §289 thus involves two steps. First, identify the “article of manufacture” to which the infringed design has been applied. Second, calculate the infringer’s total profit made on that article of manufacture. This case requires us to address a threshold matter: the scope of the term “article of manufacture.” The only question we resolve today is whether, in the case of a multicomponent product, the relevant “article of manufacture” must always be the end product sold to the consumer or whether it can also be a component of that product. Under the former interpretation, a patent holder will always be entitled to the infringer’s total profit from the end product. Under the latter interpretation, a patent holder will sometimes be entitled to the infringer’s total profit from a component of the end product.[2] A The text resolves this case. The term “article of manufacture,” as used in §289, encompasses both a product sold to a consumer and a component of that product. “Article of manufacture” has a broad meaning. An “article” is just “a particular thing.” J. Stormonth, A Dictionary of the English Language 53 (1885) (Stormonth); see also American Heritage Dictionary, at 101 (“[a]n individual thing or element of a class; a particular object or item”). And “manufacture” means “the conversion of raw materials by the hand, or by machinery, into articles suitable for the use of man” and “the articles so made.” Stormonth 589; see also American Heritage Dictionary, at 1070 (“[t]he act, craft, or process of manufacturing products, especially on a large scale” or “[a] product that is manufactured”). An article of manufacture, then, is sim-ply a thing made by hand or machine. So understood, the term “article of manufacture” is broad enough to encompass both a product sold to a consumer as well as a component of that product. A component of a product, no less than the product itself, is a thing made by hand or machine. That a component may be integrated into a larger product, in other words, does not put it outside the category of articles of manufacture. This reading of article of manufacture in §289 is consistent with 35 U. S. C. §171(a), which makes “new, original and ornamental design[s] for an article of manufacture” eligible for design patent protection.[3] The Patent Office and the courts have understood §171 to permit a design patent for a design extending to only a component of a multicomponent product. See, e.g., Ex parte Adams, 84 Off. Gaz. Pat. Office 311 (1898) (“The several articles of manufacture of peculiar shape which when combined produce a machine or structure having movable parts may each separately be patented as a design . . . ”); Application of Zahn, 617 F. 2d 261, 268 (CCPA 1980) (“Section 171 authorizes patents on ornamental designs for articles of manufacture. While the design must be embodied in some articles, the statute is not limited to designs for complete articles, or ‘discrete’ articles, and certainly not to articles separately sold . . . ”). This reading is also consistent with 35 U. S. C. §101, which makes “any new and useful . . . manufacture . . . or any new and useful improvement thereof” eligible for utility patent protection. Cf. 8 D. Chisum, Patents §23.03[2], pp. 23–12 to 23–13 (2014) (noting that “article of manufacture” in §171 includes “what would be considered a ‘manufacture’ within the meaning of Section 101”). “[T]his Court has read the term ‘manufacture’ in §101 . . . to mean ‘the production of articles for use from raw or prepared materials by giving to these materials new forms, qualities, properties, or combinations, whether by hand-labor or by machinery.’ ” Diamond v. Chakrabarty, 447 U. S. 303, 308 (1980) (quoting American Fruit Growers, Inc. v. Brogdex Co., 283 U. S. 1, 11 (1931) ). The broad term includes “the parts of a machine considered sepa-rately from the machine itself.” 1 W. Robinson, The Law of Patents for Useful Inventions §183, p. 270 (1890). B The Federal Circuit’s narrower reading of “article of manufacture” cannot be squared with the text of §289. The Federal Circuit found that components of the infringing smartphones could not be the relevant article of manufacture because consumers could not purchase those components separately from the smartphones. See 786 F. 3d, at 1002 (declining to limit a §289 award to a component of the smartphone because “[t]he innards of Samsung’s smartphones were not sold separately from their shells as distinct articles of manufacture to ordinary purchasers”); see also Nordock, Inc. v. Systems Inc., 803 F. 3d 1344, 1355 (CA Fed. 2015) (declining to limit a §289 award to a design for a “ ‘lip and hinge plate’ ” because it was “welded together” with a leveler and “there was no evidence” it was sold “separate[ly] from the leveler as a complete unit”). But, for the reasons given above, the term “article of manufacture” is broad enough to embrace both a product sold to a consumer and a component of that product, whether sold separately or not. Thus, reading “article of manufacture” in §289 to cover only an end product sold to a consumer gives too narrow a meaning to the phrase. The parties ask us to go further and resolve whether, for each of the design patents at issue here, the relevant article of manufacture is the smartphone, or a particular smartphone component. Doing so would require us to set out a test for identifying the relevant article of manufacture at the first step of the §289 damages inquiry and to parse the record to apply that test in this case. The United States as amicus curiae suggested a test, see Brief for United States as Amicus Curiae 27–29, but Samsung and Apple did not brief the issue. We decline to lay out a test for the first step of the §289 damages inquiry in the absence of adequate briefing by the parties. Doing so is not necessary to resolve the question presented in this case, and the Federal Circuit may address any remaining issues on remand. III The judgment of the United States Court of Appeals for the Federal Circuit is therefore reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Samsung raised a host of challenges on appeal related to other claims in the litigation between Apple and Samsung. The Federal Circuit affirmed in part—with respect to the design patent infringement finding, the validity of two utility patent claims, and the design and utility patent infringement damages awards—and reversed and remanded in part—with respect to trade dress dilution. Only the design patent infringement award is at issue here. 2 In its petition for certiorari and in its briefing, Samsung challenged the decision below on a second ground. It argued that 35 U. S. C. §289 contains a causation requirement, which limits a §289 damages award to the total profit the infringer made because of the infringement. Samsung abandoned this theory at argument, and so we do not address it. See Tr. of Oral Arg. 6. 3 As originally enacted, the provision protected “any new and original design for a manufacture.” §3, 5Stat. 544. The provision listed examples, including a design “worked into or worked on, or printed or painted or cast or otherwise fixed on, any article of manufacture” and a “shape or configuration of any article of manufacture.” Ibid. A streamlined version enacted in 1902 protected “any new, original, and ornamental design for an article of manufacture.” Ch. 783, 32Stat. 193. The Patent Act of 1952 retained that language. See §171, 66Stat. 813.

582.US.2016_15-1039

The Biologics Price Competition and Innovation Act of 2009 (BPCIA or Act) provides an abbreviated pathway for obtaining Food and Drug Administration (FDA) approval of a drug that is biosimilar to an already licensed biological product (reference product). 42 U. S. C. §262(k). It also provides procedures for resolving patent disputes between biosimilar manufacturers (applicants) and manufacturers of reference products (sponsors). §262(l). The Act treats the mere submission of a biosimilar application as an “artificial” act of infringement, enabling parties to bring patent infringement actions at certain points in the application process even if the applicant has not committed a traditional act of patent infringement. See 35 U. S. C. §§271(e)(2)(C)(i), (ii). Under §262(l)(2)(A), an applicant seeking FDA approval of a biosimilar must provide its application and manufacturing information to the sponsor within 20 days of the date the FDA notifies the applicant that it has accepted the application for review. This triggers an exchange of information between the applicant and sponsor designed to create lists of relevant patents and flesh out potential legal arguments. §262(l)(3). The BPCIA then channels the parties into two phases of patent litigation. In the first, the parties collaborate to identify patents on the lists for immediate litigation. The second phase—triggered when the applicant, pursuant to §262(l)(8)(A), gives the sponsor notice at least 180 days before commercially marketing the biosimilar—involves any listed patents not litigated in the first phase. The applicant has substantial control over the timing and scope of both phases of litigation. Failure to comply with these procedural requirements may lead to two consequences relevant here. Under §262(l)(9)(C), if an applicant fails to provide its application and manufacturing information to the sponsor under §262(l)(2)(A), then the sponsor, but not the applicant, may immediately bring an action “for a declaration of infringement, validity, or enforceability of any patent that claims the biological product or a use of the biological product.” And under §262(l)(9)(B), if an applicant provides the application and manufacturing information but fails to complete a subsequent step in the process, the sponsor, but not the applicant, may bring a declaratory-judgment action with respect to any patent included on the sponsor’s list of relevant patents. Neupogen is a filgrastim product marketed by Amgen, which claims to hold patents on methods of manufacturing and using filgrastim. Sandoz sought FDA approval to market a biosimilar filgrastim product under the brand name Zarxio, with Neupogen as the reference product. A day after the FDA informed Sandoz that its application had been accepted for review, Sandoz notified Amgen that it had submitted an application and that it intended to market Zarxio immediately upon receiving FDA approval. It later informed Amgen that it did not intend to provide the application and manufacturing information required by §262(l)(2)(A) and that Amgen could sue immediately for infringement under §262(l)(9)(C). Amgen sued Sandoz for patent infringement and also asserted that Sandoz engaged in “unlawful” conduct in violation of California’s unfair competition law. This latter claim was predicated on two alleged violations of the BPCIA: Sandoz’s failure to provide its application and manufacturing information under §262(l)(2)(A), and its provision of notice of commercial marketing under §262(l)(8)(A) prior to obtaining licensure from the FDA. Amgen sought injunctions to enforce both BPCIA requirements. Sandoz counterclaimed for declaratory judgments that the asserted patent was invalid and not infringed and that it had not violated the BPCIA. While the case was pending, the FDA licensed Zarxio, and Sandoz provided Amgen a further notice of commercial marketing. The District Court subsequently granted partial judgment on the pleadings to Sandoz on its BPCIA counterclaims and dismissed Amgen’s unfair competition claims with prejudice. The Federal Circuit affirmed in part, vacated in part, and remanded. The court affirmed the dismissal of Amgen’s state-law claim based on Sandoz’s alleged violation of §262(l)(2)(A), holding that Sandoz did not violate the BPCIA in failing to disclose its application and manufacturing information and that the BPCIA provides the exclusive remedies for failure to comply with this requirement. The court also held that under §262(l)(8)(A) an applicant must provide notice of commercial marketing after obtaining licensure, and that this requirement is mandatory. It thus enjoined Sandoz from marketing Zarxio until 180 days after the date it provided its second notice. Held: Section 262(l)(2)(A) is not enforceable by injunction under federal law, but the Federal Circuit on remand should determine whether a state-law injunction is available. An applicant may provide notice under §262(l)(8)(A) prior to obtaining licensure. Pp. 10–18. (a) Section 262(l)(2)(A)’s requirement that an applicant provide the sponsor with its application and manufacturing information is not enforceable by an injunction under federal law. The Federal Circuit reached the proper result on this point, but its reasoning was flawed. It cited §271(e)(4), which expressly provides the “only remedies” for an act of artificial infringement. In light of this language, the court reasoned that no remedy other than those specified in the text—such as an injunction to compel the applicant to provide its application and manufacturing information—was available. The problem with this reasoning is that Sandoz’s failure to disclose was not an act of artificial infringement remediable under §271(e)(4). Submitting an application constitutes an act of artificial infringement; failing to disclose the application and manufacturing information required by §262(l)(2)(A) does not. Another provision, §262(l)(9)(C), provides a remedy for an applicant’s failure to turn over its application and manufacturing information. It authorizes the sponsor, but not the applicant, to bring an immediate declaratory-judgment action for artificial infringement, thus vesting in the sponsor the control that the applicant would otherwise have exercised over the scope and timing of the patent litigation and depriving the applicant of the certainty it could have obtained by bringing a declaratory-judgment action prior to marketing its product. The presence of this remedy, coupled with the absence of any other textually specified remedies, indicates that Congress did not intend sponsors to have access to injunctive relief, at least as a matter of federal law, to enforce the disclosure requirement. See Great-West Life & Annuity Ins. Co. v. Knudson, 534 U. S. 204 . Statutory context further confirms that Congress did not authorize courts to enforce §262(l)(2)(A) by injunction. Pp. 10–13. (b) The Federal Circuit should determine on remand whether an injunction is available under state law to enforce §262(l)(2)(A). Whether Sandoz’s conduct was “unlawful” under California’s unfair competition statute is a question of state law, and the Federal Circuit thus erred in attempting to answer that question by referring only to the BPCIA. There is no dispute about how the federal scheme actually works on the facts of this case: Sandoz failed to disclose the requisite information under §262(l)(2)(A), and was accordingly subject to the consequence specified in §262(l)(9)(C). As a result, there is nothing to decide on this point as a matter of federal law. The court on remand should determine whether California law would treat noncompliance with §262(l)(2)(A) as “unlawful,” and whether the BPCIA pre-empts any additional state-law remedy for failure to comply with §262(l)(2)(A). Pp. 13–15. (c) An applicant may provide notice of commercial marketing before obtaining a license. Section 262(l)(8)(A) states that the applicant “shall provide notice to the reference product sponsor not later than 180 days before the date of the first commercial marketing of the biological product licensed under subsection (k).” Because the phrase “of the biological product licensed under subsection (k)” modifies “commercial marketing” rather than “notice,” “commercial marketing” is the point in time by which the biosimilar must be “licensed.” Accordingly, the applicant may provide notice either before or after receiving FDA approval. Statutory context confirms that §262(l)(8)(A) contains a single timing requirement (180 days before marketing), rather than the two requirements posited by the Federal Circuit (after licensing, and 180 days before marketing). “Had Congress intended to” impose two timing requirements in §262(l)(8)(A), “it presumably would have done so expressly as it did in the” adjacent provision, §262(l)(8)(B). Russello v. United States, 464 U. S. 16 . Amgen’s contrary arguments are unpersuasive, and its various policy arguments cannot overcome the statute’s plain language. Pp. 15–18. 794 F. 3d 1347, vacated in part, reversed in part, and remanded. Thomas, J., delivered the opinion for a unanimous Court. Breyer, J., filed a concurring opinion.Notes 1 Together with No. 15–1195, Amgen Inc. et al. v. Sandoz Inc., also on certiorari to the same court.

These cases involve 42 U. S. C. §262(l), which was enacted as part of the Biologics Price Competition and Innovation Act of 2009 (BPCIA), 124Stat. 808. The BPCIA governs a type of drug called a biosimilar, which is a biologic product that is highly similar to a biologic product that has already been approved by the Food and Drug Administration (FDA). Under §262(l), an applicant that seeks FDA approval of a biosimilar must provide its application materials and manufacturing information to the manufacturer of the corresponding biologic within 20 days of the date the FDA notifies the applicant that it has accepted the application for review. The applicant then must give notice to the manufacturer at least 180 days before marketing the biosimilar commercially. The first question presented by these cases is whether the requirement that an applicant provide its application and manufacturing information to the manufacturer of the biologic is enforceable by injunction. We conclude that an injunction is not available under federal law, but we remand for the court below to decide whether an injunction is available under state law. The second question is whether the applicant must give notice to the manufac-turer after, rather than before, obtaining a license from the FDA for its biosimilar. We conclude that an applicant may provide notice before obtaining a license. I The complex statutory scheme at issue in these cases establishes processes both for obtaining FDA approval of biosimilars and for resolving patent disputes between manufacturers of licensed biologics and manufacturers of biosimilars. Before turning to the questions presented, we first explain the statutory background. A A biologic is a type of drug derived from natural, biological sources such as animals or microorganisms. Biologics thus differ from traditional drugs, which are typically synthesized from chemicals.[1] A manufacturer of a biologic may market the drug only if the FDA has licensed it pursuant to either of two review processes set forth in §262. The default pathway for approval, used for new biologics, is set forth in §262(a). Under that subsection, the FDA may license a new biologic if, among other things, the manufacturer demonstrates that it is “safe, pure, and potent.” §262(a)(2)(C)(i)(I). In addition to this default route, the statute also prescribes an alternative, abbreviated route for FDA approval of biosimilars, which is set forth in §262(k). To obtain approval through the BPCIA’s abbreviated process, the manufacturer of a biosimilar (applicant) does not need to show that the product is “safe, pure, and potent.” Instead, the applicant may piggyback on the showing made by the manufacturer (sponsor) of a previously licensed biologic (reference product). See §262(k)(2)(A)(iii). An applicant must show that its product is “highly similar” to the reference product and that there are no “clinically meaningful differences” between the two in terms of “safety, purity, and potency.” §§262(i)(2)(A), (B); see also §262(k)(2)(A)(i)(I). An applicant may not submit an application until 4 years after the reference product is first licensed, and the FDA may not license a biosimilar until 12 years after the reference product is first licensed. §§262(k)(7)(A), (B). As a result, the manufacturer of a new biologic enjoys a 12-year period when its biologic may be marketed without competition from biosimilars. B A sponsor may hold multiple patents covering the biologic, its therapeutic uses, and the processes used to manufacture it. Those patents may constrain an applicant’s ability to market its biosimilar even after the expiration of the 12-year exclusivity period contained in §262(k)(7)(A). The BPCIA facilitates litigation during the period preceding FDA approval so that the parties do not have to wait until commercial marketing to resolve their patent disputes. It enables the parties to bring infringement actions at certain points in the application process, even if the applicant has not yet committed an act that would traditionally constitute patent infringement. See 35 U. S. C. §271(a) (traditionally infringing acts include making, using, offering to sell, or selling any patented invention within the United States without authority to do so). Specifically, it provides that the mere submission of a biosimilar application constitutes an act of infringement. §§271(e)(2)(C)(i), (ii). We will refer to this kind of preapproval infringement as “artificial” infringement. Section 271(e)(4) provides remedies for artificial infringement, including injunctive relief and damages. C The BPCIA sets forth a carefully calibrated scheme for preparing to adjudicate, and then adjudicating, claims of infringement. See 42 U. S. C. §262(l). When the FDA accepts an application for review, it notifies the applicant, who within 20 days “shall provide” to the sponsor a copy of the application and information about how the biosimilar is manufactured. §262(l)(2)(A). The applicant also “may provide” the sponsor with any additional information that it requests. §262(l)(2)(B). These disclosures enable the sponsor to evaluate the biosimilar for possible infringement of patents it holds on the reference product (i.e., the corresponding biologic). §262(l)(1)(D). The information the applicant provides is subject to strict confidentiality rules, enforceable by injunction. See §262(l)(1)(H). The first question presented by these cases is whether §262(l)(2)(A)’s requirement—that the applicant provide its application and manufacturing information to the sponsor—is itself enforceable by injunction. After the applicant makes the requisite disclosures, the parties exchange information to identify relevant patents and to flesh out the legal arguments that they might raise in future litigation. Within 60 days of receiving the application and manufacturing information, the sponsor “shall provide” to the applicant “a list of patents” for which it believes it could assert an infringement claim if a person without a license made, used, offered to sell, sold, or imported “the biological product that is the subject of the [biosimilar] application.” §262(l)(3)(A)(i). The sponsor must also identify any patents on the list that it would be willing to license. §262(l)(3)(A)(ii). Next, within 60 days of receiving the sponsor’s list, the applicant may provide to the sponsor a list of patents that the applicant believes are relevant but that the sponsor omitted from its own list, §262(l)(3)(B)(i), and “shall provide” to the sponsor reasons why it could not be held liable for infringing the relevant patents, §262(l)(3)(B)(ii). The applicant may argue that the relevant patents are invalid, unenforceable, or not infringed, or the applicant may agree not to market the biosimilar until a particular pat-ent has expired. Ibid. The applicant must also respond to the sponsor’s offers to license particular patents. §262(l)(3)(B)(iii). Then, within 60 days of receiving the applicant’s responses, the sponsor “shall provide” to the applicant its own arguments concerning infringement, enforceability, and validity as to each relevant patent. §262(l)(3)(C). Following this exchange, the BPCIA channels the parties into two phases of patent litigation. In the first phase, the parties collaborate to identify patents that they would like to litigate immediately. The second phase is triggered by the applicant’s notice of commercial marketing and involves any patents that were included on the parties’ §262(l)(3) lists but not litigated in the first phase. At the outset of the first phase, the applicant and the sponsor must negotiate to determine which patents included on the §262(l)(3) lists will be litigated immediately. See §§262(l)(4)(A), (l)(6). If they cannot agree, then they must engage in another list exchange. §262(l)(4)(B). The applicant “shall notify” the sponsor of the number of pat-ents it intends to list for litigation, §262(l)(5)(A), and, within five days, the parties “shall simultaneously exchange” lists of the patents they would like to litigate immediately. §262(l)(5)(B)(i). This process gives the applicant substantial control over the scope of the first phase of litigation: The number of patents on the sponsor’s list is limited to the number contained in the applicant’s list, though the sponsor always has the right to list at least one patent. §262(l)(5)(B)(ii). The parties then proceed to litigate infringement with respect to the patents they agreed to litigate or, if they failed to agree, the patents contained on the lists they simultaneously exchanged under §262(l)(5). §§262(l)(6)(A), (B). Section 271(e)(2)(C)(i) facilitates this first phase of litigation by making it an act of artificial infringement, with respect to any patent included on the parties’ §262(l)(3) lists, to submit an application for a license from the FDA. The sponsor “shall bring an action” in court within 30 days of the date of agreement or the simultaneous list exchange. §§262(l)(6)(A), (B). If the sponsor brings a timely action and prevails, it may obtain a rem-edy provided by §271(e)(4). The second phase of litigation involves patents that were included on the original §262(l)(3) lists but not litigated in the first phase (and any patents that the sponsor acquired after the §262(l)(3) exchange occurred and added to the lists, see §262(l)(7)). The second phase is commenced by the applicant’s notice of commercial marketing, which the applicant “shall provide” to the sponsor “not later than 180 days before the date of the first commercial marketing of the biological product licensed under subsection (k).” §262(l)(8)(A). The BPCIA bars any declaratory judgment action prior to this notice. §262(l)(9)(A) (prohibiting, in situations where the parties have complied with each step of the BPCIA process, either the sponsor or the applicant from seeking a “declaration of infringement, validity, or enforceability of any patent” that was included on the §262(l)(3) lists but not litigated in the first phase “prior to the date notice is received under paragraph (8)(A)”). Because the applicant (subject to certain constraints) chooses when to begin commercial marketing and when to give notice, it wields substantial control over the timing of the second phase of litigation. The second question presented is whether notice is effective if an appli-cant provides it prior to the FDA’s decision to license the biosimilar. In this second phase of litigation, either party may sue for declaratory relief. See §262(l)(9)(A). In addition, prior to the date of first commercial marketing, the sponsor may “seek a preliminary injunction prohibiting the [biosimilar] applicant from engaging in the commercial manufacture or sale of [the biosimilar] until the court decides the issue of patent validity, enforcement, and infringement with respect to any patent that” was included on the §262(l)(3) lists but not litigated in the first phase. §262(l)(8)(B). D If the parties comply with each step outlined in the BPCIA, they will have the opportunity to litigate the relevant patents before the biosimilar is marketed. To encourage parties to comply with its procedural requirements, the BPCIA includes various consequences for failing to do so. Two of the BPCIA’s remedial provisions are at issue here. Under §262(l)(9)(C), if an applicant fails to provide its application and manufacturing information to the sponsor—thus effectively pretermitting the entire two-phase litigation process—then the sponsor, but not the applicant, may immediately bring an action “for a declaration of infringement, validity, or enforceability of any patent that claims the biological product or a use of the biological product.” Section 271(e)(2)(C)(ii) facilitates this action by making it an artificial act of infringement, with respect to any patent that could have been included on the §262(l)(3) lists, to submit a biosimilar application. Similarly, when an applicant provides the application and manufacturing information but fails to complete a subsequent step, §262(l)(9)(B) provides that the sponsor, but not the applicant, may bring a declaratory-judgment action with respect to any patent included on the sponsor’s §262(l)(3)(A) list of patents (as well as those it acquired later and added to the list). As noted, it is an act of artificial infringement, with respect to any patent on the §262(l)(3) lists, to submit an application to the FDA. See §271(e)(2)(C)(i). II These cases concern filgrastim, a biologic used to stimulate the production of white blood cells. Amgen, the respondent in No. 15–1039 and the petitioner in No. 15–1195, has marketed a filgrastim product called Neupogen since 1991 and claims to hold patents on methods of manufacturing and using filgrastim. In May 2014, Sandoz, the petitioner in No. 15–1039 and the respondent in No. 15–1195, filed an application with the FDA seeking approval to market a filgrastim biosimilar under the brand name Zarxio, with Neupogen as the reference product. The FDA informed Sandoz on July 7, 2014, that it had accepted the application for review. One day later, Sandoz notified Amgen both that it had submitted an application and that it intended to begin marketing Zarxio immediately upon receiving FDA approval, which it expected in the first half of 2015. Sandoz later confirmed that it did not intend to provide the requisite application and manufacturing information under §262(l)(2)(A) and informed Amgen that Amgen could sue for infringement immediately under §262(l)(9)(C). In October 2014, Amgen sued Sandoz for patent infringement. Amgen also asserted two claims under California’s unfair competition law, which prohibits “any unlawful . . . business act or practice.” Cal. Bus. & Prof. Code Ann. §17200 (West 2008). A “business act or practice” is “unlawful” under the unfair competition law if it violates a rule contained in some other state or federal statute. Rose v. Bank of America, N. A., 57 Cal. 4th 390, 396, 304 P. 3d 181, 185 (2013). Amgen alleged that Sandoz engaged in “unlawful” conduct when it failed to provide its application and manufacturing information under §262(l)(2)(A), and when it provided notice of commercial marketing under §262(l)(8)(A) before, rather than after, the FDA licensed its biosimilar. Amgen sought injunctions to enforce both requirements. Sandoz counterclaimed for declaratory judgments that the asserted pat-ent was invalid and not infringed and that it had not violated the BPCIA. While the case was pending in the District Court, the FDA licensed Zarxio, and Sandoz provided Amgen a further notice of commercial marketing. The District Court subsequently granted partial judgment on the pleadings to Sandoz on its BPCIA counterclaims and dismissed Amgen’s unfair competition claims with prejudice. 2015 WL 1264756, *7–*9 (ND Cal., Mar. 19, 2015). After the District Court entered final judgment as to these claims, Amgen appealed to the Federal Circuit, which granted an injunction pending appeal against the commercial marketing of Zarxio. A divided Federal Circuit affirmed in part, vacated in part, and remanded. First, the court affirmed the dismissal of Amgen’s state-law claim based on Sandoz’s alleged violation of §262(l)(2)(A). It held that Sandoz did not violate the BPCIA in failing to disclose its application and manufacturing information. It further held that the remedies contained in the BPCIA are the exclusive remedies for an applicant’s failure to comply with §262(l)(2)(A). 794 F. 3d 1347, 1357, 1360 (2015). Second, the court held that an applicant may provide effective notice of commercial marketing only after the FDA has licensed the biosimilar. Id., at 1358. Accord-ingly, the 180-day clock began after Sandoz’s second, post-licensure notice. The Federal Circuit further concluded that the notice requirement is mandatory and extended its injunction pending appeal to bar Sandoz from marketing Zarxio until 180 days after the date it provided its second notice. Id., at 1360–1361. We granted Sandoz’s petition for certiorari, No. 15–1039, and Amgen’s conditional cross-petition for certiorari, No. 15–1195, and consolidated the cases. 580 U. S. ___ (2017). III The first question we must answer is whether §262(l)(2)(A)’s requirement that an applicant provide the sponsor with its application and manufacturing information is enforceable by an injunction under either federal or state law. A We agree with the Federal Circuit that an injunction under federal law is not available to enforce §262(l)(2)(A), though for slightly different reasons than those provided by the court below. The Federal Circuit held that “ 42 U. S. C. §262(l)(9)(C) and 35 U. S. C. §271(e) expressly provide the only remedies” for a violation of §262(l)(2)(A), 794 F. 3d, at 1357, and neither of those provisions authorizes a court to compel compliance with §262(l)(2)(A). In concluding that the remedies specified in the BPCIA are exclusive, the Federal Circuit relied primarily on §271(e)(4), which states that it provides “ ‘the only remedies which may be granted by a court for an act of [artificial] infringement.’ ” Id., at 1356 (emphasis deleted). The flaw in the Federal Circuit’s reasoning is that Sandoz’s failure to disclose its application and manufacturing information was not an act of artificial infringement, and thus was not remediable under §271(e)(4). Submitting an application constitutes an act of artificial infringement. See §§271(e)(2)(C)(i), (ii) (“It shall be an act of infringement to submit . . . an application seeking approval of a biological product”). Failing to disclose the application and manufacturing information under §262(l)(2)(A) does not. In reaching the opposite conclusion, the Federal Circuit relied on §271(e)(2)(C)(ii), which states that “[i]t shall be an act of infringement to submit[,] if the applicant for the application fails to provide the application and information required under [§262(l)(2)(A)], an application seeking approval of a biological product for a patent that could be identified pursuant to [§262(l)(3)(A)(i)].” (Emphasis added.) The court appeared to conclude, based on the italicized language, that an applicant’s noncompliance with §262(l)(2)(A) is an element of the act of artificial infringement (along with the submission of the application). 794 F. 3d, at 1356. We disagree. The italicized language merely assists in identifying which patents will be the subject of the artificial infringement suit. It does not define the act of artificial infringement itself. This conclusion follows from the structure of §271(e)(2)(C). Clause (i) of §271(e)(2)(C) defines artificial infringement in the situation where the parties proceed through the list exchange process and the patents subject to suit are those contained in the §262(l)(3) lists, as supplemented under §262(l)(7). That clause provides that it is an act of artificial infringement to submit, “with respect to a patent that is identified in the list of patents described in [§262(l)(3)] (including as provided under [§262(l)(7)]), an application seeking approval of a biological product.” (Emphasis added.) Clause (ii) of §271(e)(2)(C), in contrast, defines artificial infringement in the situation where an applicant fails to disclose its application and manufacturing information altogether and the parties never prepare the §262(l)(3) lists. That clause provides that the submission of the application represents an act of artificial infringement with respect to any patent that could have been included on the lists. In this way, the two clauses of §271(e)(2)(C) work in tandem. They both treat submission of the application as the act of artificial infringement for which §271(e)(4) provides the remedies. And they both identify the patents subject to suit, although by different means depending on whether the applicant disclosed its application and manufacturing information under §262(l)(2)(A). If the applicant made the disclosures, clause (i) applies; if it did not, clause (ii) applies. In neither instance is the applicant’s failure to provide its application and manufacturing information an element of the act of artificial infringement, and in neither instance does §271(e)(4) provide a remedy for that failure. See Brief for Amgen Inc. et al. 66–67 (conceding both points). A separate provision of §262, however, does provide a remedy for an applicant’s failure to turn over its application and manufacturing information. When an applicant fails to comply with §262(l)(2)(A), §262(l)(9)(C) authorizes the sponsor, but not the applicant, to bring an immediate declaratory-judgment action for artificial infringement as defined in §271(e)(2)(C)(ii). Section 262(l)(9)(C) thus vests in the sponsor the control that the applicant would otherwise have exercised over the scope and timing of the patent litigation. It also deprives the applicant of the certainty that it could have obtained by bringing a declaratory-judgment action prior to marketing its product. The remedy provided by §262(l)(9)(C) excludes all other federal remedies, including injunctive relief. Where, as here, “a statute expressly provides a remedy, courts must be especially reluctant to provide additional remedies.” Karahalios v. Federal Employees, 489 U. S. 527, 533 (1989) . The BPCIA’s “carefully crafted and detailed enforcement scheme provides strong evidence that Congress did not intend to authorize other remedies that it simply forgot to incorporate expressly.” Great-West Life & Annu-ity Ins. Co. v. Knudson, 534 U. S. 204, 209 (2002) (internal quotation marks omitted). The presence of §262(l)(9)(C), coupled with the absence of any other textually specified remedies, indicates that Congress did not intend sponsors to have access to injunctive relief, at least as a matter of federal law, to enforce the disclosure requirement. Statutory context further confirms that Congress did not authorize courts to enforce §262(l)(2)(A) by injunction. Section 262(l)(1)(H) provides that “the court shall consider immediate injunctive relief to be an appropriate and necessary remedy for any violation or threatened violation” of the rules governing the confidentiality of information disclosed under §262(l). We assume that Congress acted intentionally when it provided an injunctive remedy for breach of the confidentiality requirements but not for breach of §262(l)(2)(A)’s disclosure requirement. Cf. Touche Ross & Co. v. Redington, 442 U. S. 560, 572 (1979) (“[W]hen Congress wished to provide a private damage remedy, it knew how to do so and did so expressly”).[2] Accordingly, the Federal Circuit properly declined to grant an injunction under federal law. B The Federal Circuit rejected Amgen’s request for an injunction under state law for two reasons. First, it interpreted California’s unfair competition law not to provide a remedy when the underlying statute specifies an “expressly . . . exclusive” remedy. 794 F. 3d, at 1360 (citing Cal. Bus. & Prof. Code Ann. §17205; Loeffler v. Target Corp., 58 Cal. 4th 1081, 1125–1126, 324 P. 3d 50, 76 (2014)). It further held that §271(e)(4), by its text, “provides ‘the only remedies’ ” for an applicant’s failure to disclose its application and manufacturing information. 794 F. 3d, at 1360 (quoting §271(e)(4)). The court thus concluded that no state remedy was available for Sandoz’s alleged violation of §262(l)(2)(A) under the terms of California’s unfair competition law. This state-law holding rests on an incorrect interpretation of federal law. As we have explained, failure to comply with §262(l)(2)(A) is not an act of artificial infringement. Because §271(e)(4) provides remedies only for artificial infringement, it provides no remedy at all, much less an “expressly . . . exclusive” one, for Sandoz’s failure to comply with §262(l)(2)(A). Second, the Federal Circuit held in the alternative that Sandoz’s failure to disclose its application and manufacturing information was not “unlawful” under California’s unfair competition law. In the court’s view, when an applicant declines to provide its application and manufacturing information to the sponsor, it takes a path “expressly contemplated by” §262(l)(9)(C) and §271(e)(2)(C)(ii) and thus does not violate the BPCIA. 794 F. 3d, at 1357, 1360. In their briefs before this Court, the parties frame this issue as whether the §262(l)(2)(A) requirement is mandatory in all circumstances, see Brief for Amgen Inc. et al. 58, or merely a condition precedent to the information exchange process, see Reply Brief for Sandoz Inc. 33. If it is only a condition precedent, then an applicant effectively has the option to withhold its application and manufacturing information and does not commit an “unlawful” act in doing so. We decline to resolve this particular dispute definitively because it does not present a question of federal law. The BPCIA, standing alone, does not require a court to decide whether §262(l)(2)(A) is mandatory or conditional; the court need only determine whether the applicant supplied the sponsor with the information required under §262(l)(2)(A). If the applicant failed to provide that information, then the sponsor, but not the applicant, could bring an immediate declaratory-judgment action pursuant to §262(l)(9)(C). The parties in these cases agree—as did the Federal Circuit—that Sandoz failed to comply with §262(l)(2)(A), thus subjecting itself to that consequence. There is no dispute about how the federal scheme actually works, and thus nothing for us to decide as a matter of federal law. The mandatory or conditional nature of the BPCIA’s requirements matters only for purposes of California’s unfair competition law, which penalizes “unlawful” conduct. Whether Sandoz’s conduct was “unlawful” under the unfair competition law is a state-law question, and the court below erred in attempting to answer that question by referring to the BPCIA alone. On remand, the Federal Circuit should determine whether California law would treat noncompliance with §262(l)(2)(A) as “unlawful.” If the answer is yes, then the court should proceed to determine whether the BPCIA pre-empts any additional remedy available under state law for an applicant’s failure to comply with §262(l)(2)(A) (and whether Sandoz has forfeited any pre-emption defense, see 794 F. 3d, at 1360, n. 5). The court is also of course free to address the pre-emption question first by assuming that a remedy under state law exists. IV The second question at issue in these cases is whether an applicant must provide notice after the FDA licenses its biosimilar, or if it may also provide effective notice before licensure. Section 262(l)(8)(A) states that the applicant “shall provide notice to the reference product sponsor not later than 180 days before the date of the first commercial marketing of the biological product licensed under subsection (k).” The Federal Circuit held that an applicant’s biosimilar must already be “licensed” at the time the applicant gives notice. 794 F. 3d, at 1358. We disagree. The applicant must give “notice” at least 180 days “before the date of the first commercial marketing.” “[C]ommercial marketing,” in turn, must be “of the biological product licensed under subsection (k).” §262(l)(8)(A). Because this latter phrase modifies “commercial marketing” rather than “notice,” “commercial marketing” is the point in time by which the biosimilar must be “licensed.” The statute’s use of the word “licensed” merely reflects the fact that, on the “date of the first commercial marketing,” the product must be “licensed.” See §262(a)(1)(A). Accordingly, the applicant may provide notice either before or after receiving FDA approval. Statutory context confirms this interpretation. Section 262(l)(8)(A) contains a single timing requirement: The applicant must provide notice at least 180 days prior to marketing its biosimilar. The Federal Circuit, however, interpreted the provision to impose two timing requirements: The applicant must provide notice after the FDA licenses the biosimilar and at least 180 days before the applicant markets the biosimilar. An adjacent provision expressly sets forth just that type of dual timing requirement. See §262(l)(8)(B) (“After receiving notice under subparagraph (A) and before such date of the first commercial marketing of such biological product, the reference product sponsor may seek a preliminary injunction” (emphasis added)). But Congress did not use that structure in §262(l)(8)(A). “Had Congress intended to” impose two timing requirements in §262(l)(8)(A), “it presumably would have done so expressly as it did in the immediately following” subparagraph. Russello v. United States, 464 U. S. 16, 23 (1983) . We are not persuaded by Amgen’s arguments to the contrary. Amgen points out that other provisions refer to “ ‘the biological product that is the subject of’ ” the application, rather than the “ ‘biological product licensed under subsection (k).’ ” Brief for Amgen Inc. et al. 28 (emphasis added). In its view, this variation “is a strong textual indication that §262(l)(8)(A), unlike the other provisions, refers to a product that has already been ‘licensed’ by the FDA.” Ibid. Amgen’s interpretation is not necessary to harmonize Congress’ use of the two different phrases. The provision upon which Amgen primarily relies (and that is generally illustrative of the other provisions it cites) requires the applicant to explain why the sponsor’s patents are “ ‘invalid, unenforceable, or will not be infringed by the commercial marketing of the biological product that is the subject of the subsection (k) application.’ ” Id., at 29–30 (quoting §262(l)(3)(B)(ii)(I); emphasis deleted). This provision uses the phrase “subject of the subsection (k) application” rather than “product licensed under subsection (k)” because the applicant can evaluate validity, enforceability, and infringement with respect to the biosimilar only as it exists when the applicant is conducting the evaluation, which it does before licensure. The applicant cannot make the same evaluation with respect to the biosimilar as it will exist after licensure, because the biosimilar’s specifications may change during the application process. See, e.g., 794 F. 3d, at 1358. In contrast, nothing in §262(l)(8)(A) turns on the precise status or characteristics of the biosimilar application. Amgen also advances a host of policy arguments that prelicensure notice is undesirable. See Brief for Amgen Inc. et al. 35–42. Sandoz and the Government, in turn, respond with their own bevy of arguments that Amgen’s concerns are misplaced and that prelicensure notice affirmatively furthers Congress’ intent. See Brief for Sandoz Inc. 39–42, 56; Brief for United States as Amicus Curiae 28–29. The plausibility of the contentions on both sides illustrates why such disputes are appropriately addressed to Congress, not the courts. Even if we were persuaded that Amgen had the better of the policy arguments, those arguments could not overcome the statute’s plain language, which is our “primary guide” to Congress’ preferred policy. McFarland v. Scott, 512 U. S. 849, 865 (1994) (Thomas, J., dissenting). In sum, because Sandoz fully complied with §262(l)(8)(A) when it first gave notice (before licensure) in July 2014, the Federal Circuit erred in issuing a federal injunction prohibiting Sandoz from marketing Zarxio until 180 days after licensure. Furthermore, because Amgen’s request for state-law relief is predicated on its argument that the BPCIA forbids prelicensure notice, its claim under California’s unfair competition law also fails. We accordingly reverse the Federal Circuit’s judgment as to the notice provision. * * * For the foregoing reasons, the judgment of the Court of Appeals is vacated in part and reversed in part, and the cases are remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 FDA, What Are “Biologics” Questions and Answers (Aug. 5, 2015), http://www.fda.gov/aboutfda/centersoffices/officeofmedicalproductsandtobacco /cber /ucm133077.htm (as last visited June 6, 2017). 2 In holding that §262(l)(9)(C) represents the exclusive remedy for an applicant’s failure to provide its application and manufacturing information, we express no view on whether a district court could take into account an applicant’s violation of §262(l)(2)(A) (or any other BPCIA procedural requirement) in deciding whether to grant a preliminary injunction under 35 U. S. C. §271(e)(4)(B) or §283 against marketing the biosimilar. See Winter v. Natural Resources Defense Council, Inc., 555 U. S. 7, 20 (2008) (court should consider “balance of equities” in deciding whether to grant a preliminary injunction).

580.US.2016_15-927

In 2003, petitioners (collectively, SCA) notified respondents (collectively, First Quality) that their adult incontinence products infringed an SCA patent. First Quality responded that its own patent antedated SCA’s patent and made it invalid. In 2004, SCA sought reexamination of its patent in light of First Quality’s patent, and in 2007, the Patent and Trademark Office confirmed the SCA patent’s validity. SCA sued First Quality for patent infringement in 2010. The District Court granted summary judgment to First Quality on the grounds of equitable estoppel and laches. While SCA’s appeal was pending, this Court held that laches could not preclude a claim for damages incurred within the Copyright Act’s 3-year limitations period. Petrella v. Metro-Goldwyn-Mayer, Inc., 572 U. S. ___, ___. A Federal Circuit panel nevertheless affirmed the District Court’s laches holding based on Circuit precedent, which permitted laches to be asserted against a claim for damages incurred within the Patent Act’s 6-year limitations period, 35 U. S. C. §286. The en banc court reheard the case in light of Petrella and reaffirmed the original panel’s laches holding. Held: Laches cannot be invoked as a defense against a claim for damages brought within §286’s 6-year limitations period. Pp. 3–16. (a) Petrella’s holding rested on both separation-of-powers principles and the traditional role of laches in equity. A statute of limitations reflects a congressional decision that timeliness is better judged by a hard and fast rule instead of a case-specific judicial determination. Applying laches within a limitations period specified by Congress would give judges a “legislation-overriding” role that exceeds the Judiciary’s power. 572 U. S., at ___. Moreover, applying laches within a limitations period would clash with the gap-filling purpose for which the defense developed in the equity courts. Pp. 3–5. (b) Petrella’s reasoning easily fits §286. There, the Court found in the Copyright Act’s language a congressional judgment that a claim filed within three years of accrual cannot be dismissed on timeliness grounds. 572 U. S., at ___. By that same logic, §286 of the Patent Act represents Congress’s judgment that a patentee may recover damages for any infringement committed within six years of the filing of the claim. First Quality contends that this case differs from Petrella because a true statute of limitations runs forward from the date a cause of action accrues, whereas §286’s limitations period runs backward from the filing of the complaint. However, Petrella repeatedly characterized the Copyright Act’s limitations period as running backward from the date the suit was filed. First Quality also contends that a true statute of limitations begins to run when the plaintiff discovers a cause of action, which is not the case with §286’s limitations period, but ordinarily, a statute of limitations begins to run on the date that the claim accrues, not when the cause of action is discovered. Pp. 5–8. (c) The Federal Circuit based its decision on the idea that §282 of the Patent Act, which provides for “defenses in any action involving the validity or infringement of a patent,” creates an exception to §286 by codifying laches as such a defense, and First Quality argues that laches is a defense within §282(b)(1) based on “unenforceability.” Even assuming that §282(b)(1) incorporates a laches defense of some dimension, it does not necessarily follow that the defense may be invoked to bar a claim for damages incurred within the period set out in §286. Indeed, it would be exceedingly unusual, if not unprecedented, if Congress chose to include in the Patent Act both a statute of limitations for damages and a laches provision applicable to a damages claim. Neither the Federal Circuit, nor any party, has identified a single federal statute that provides such dual protection against untimely claims. Pp. 8–9. (d) The Federal Circuit and First Quality rely on lower court patent cases decided before the 1952 Patent Act to argue that §282 codified a pre-1952 practice of permitting laches to be asserted against damages claims. But the most prominent feature of the relevant legal landscape at that time was the well-established rule that laches cannot be invoked to bar a claim for damages incurred within a limitations period specified by Congress. In light of this rule, which Petrella confirmed and restated, 572 U. S., at ___, nothing less than a broad and unambiguous consensus of lower court decisions could support the inference that §282(b)(1) codifies a very different patent-law-specific rule. Pp. 9–10. (e) The Federal Circuit and First Quality rely on three types of cases: (1) pre-1938 equity cases; (2) pre-1938 claims at law; and (3) cases decided after the merger of law and equity in 1938. None of these establishes a broad, unambiguous consensus in favor of applying laches to damages claims in the patent context. Many of the pre-1938 equity cases do not even reveal whether the plaintiff asked for damages, and of the cases in which damages were sought, many merely suggest in dicta that laches might limit damages. The handful of cases that apply laches against a damages claim are too few to establish a settled, national consensus. In any event, the most that can possibly be gathered from a pre-1938 equity case is that laches could defeat a damages claim in an equity court, not that the defense could entirely prevent a patentee from recovering damages. Similarly, even if all three pre-1938 cases at law cited by First Quality squarely held that laches could be applied to a damages claim within the limitations period, that number would be insufficient to overcome the presumption that Congress legislates against the background of general common-law principles. First Quality argues that the small number of cases at law should not count against its position because there were few patent cases brought at law after 1870, but it is First Quality’s burden to show that Congress departed from the traditional common-law rule. As for the post-1938 patent case law, there is scant evidence supporting First Quality’s claim that courts continued to apply laches to damages claims after the merger of law and equity. Only two Courts of Appeals held that laches could bar a damages claim, and that does not constitute a settled, uniform practice of applying laches to damages claims. Pp. 11–15. (f) First Quality’s additional arguments are unconvincing and do not require extended discussion. It points to post-1952 Court of Appeals decisions holding that laches can be invoked as a defense against a damages claim, but nothing that Congress has done since 1952 has altered §282’s meaning. As for the various policy arguments presented here, this Court cannot overrule Congress’s judgment based on its own policy views. Pp. 15–16. 807 F. 3d 1311, vacated in part and remanded. Alito, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Thomas, Ginsburg, Sotomayor, and Kagan, JJ., joined. Breyer, J., filed a dissenting opinion.

We return to a subject that we addressed in Petrella v. Metro-Goldwyn-Mayer, Inc., 572 U. S. ___ (2014): the relationship between the equitable defense of laches and claims for damages that are brought within the time allowed by a statute of limitations. In Petrella, we held that laches cannot preclude a claim for damages incurred within the Copyright Act’s 3-year limitations period. Id., at ___ (slip op., at 1). “[L]aches,” we explained, “cannot be invoked to bar legal relief” “[i]n the face of a statute of limitations enacted by Congress.” Id., at ___ (slip op., at 13). The question in this case is whether Petrella’s reasoning applies to a similar provision of the Patent Act, 35 U. S. C. §286. We hold that it does. I Petitioners SCA Hygiene Products Aktiebolag and SCA Personal Care, Inc. (collectively, SCA), manufacture and sell adult incontinence products. In October 2003, SCA sent a letter to respondents (collectively, First Quality), alleging that First Quality was making and selling products that infringed SCA’s rights under U. S. Patent No. 6,375,646 B1 (’646 patent). App. 54a. First Quality responded that one of its patents—U. S. Patent No. 5,415,649 (Watanabe patent)—antedated the ’646 patent and revealed “the same diaper construction.” Id., at 53a. As a result, First Quality maintained, the ’646 patent was invalid and could not support an infringement claim. Ibid. SCA sent First Quality no further correspondence regarding the ’646 patent, and First Quality proceeded to develop and market its products. In July 2004, without notifying First Quality, SCA asked the Patent and Trademark Office (PTO) to initiate a reexamination proceeding to determine whether the ’646 patent was valid in light of the Watanabe patent. Id., at 49a–51a. Three years later, in March 2007, the PTO issued a certificate confirming the validity of the ’646 patent. In August 2010, SCA filed this patent infringement action against First Quality. First Quality moved for summary judgment based on laches and equitable estoppel, and the District Court granted that motion on both grounds. 2013 WL 3776173, *12 (WD Ky., July 16, 2013). SCA appealed to the Federal Circuit, but before the Federal Circuit panel issued its decision, this Court de-cided Petrella. The panel nevertheless held, based on a Fed-eral Circuit precedent, A. C. Aukerman Co. v. R. L. Chaides Constr. Co., 960 F. 2d 1020 (1992) (en banc), that SCA’s claims were barred by laches.[1] The Federal Circuit then reheard the case en banc in order to reconsider Aukerman in light of Petrella. But in a 6-to-5 decision, the en banc court reaffirmed Aukerman’s holding that laches can be asserted to defeat a claim for damages incurred within the 6-year period set out in the Patent Act. As it had in Aukerman, the en banc court concluded that Congress, in enacting the Patent Act, had “codified a laches defense” that “barred recovery of legal remedies.” 807 F. 3d 1311, 1323–1329 (2015). Judge Hughes, joined by four other judges, dissented.[2] Id., at 1337–1342 (opinion concurring in part and dissenting in part). We granted certiorari. 578 U. S. ___ (2016). II Laches is “a defense developed by courts of equity” to protect defendants against “unreasonable, prejudicial delay in commencing suit.” Petrella, supra, at ___, ___ (slip op., at 1, 12). See also 1 D. Dobbs, Law of Remedies §2.3(5), p. 89 (2d ed. 1993) (Dobbs) (“The equitable doctrine of laches bars the plaintiff whose unreasonable delay in prosecuting a claim or protecting a right has worked a prejudice to the defendant”). Before the separate systems of law and equity were merged in 1938, the ordinary rule was that laches was available only in equity courts.[3] See County of Oneida v. Oneida Indian Nation of N. Y., 470 U. S. 226 , n. 16 (1985). This case turns on the application of the defense to a claim for damages, a quintessential legal remedy. We discussed this subject at length in Petrella. Petrella arose out of a copyright dispute relating to the film Raging Bull. 572 U. S., at ___ (slip op., at 8). The Copyright Act’s statute of limitations requires a copyright holder claiming infringement to file suit “within three years after the claim accrued.” 17 U. S. C. §507(b). In Petrella, the plaintiff sought relief for alleged acts of infringement that accrued within that 3-year period, but the lower courts nevertheless held that laches barred her claims. See 695 F. 3d 946 (CA9 2012). We reversed, holding that laches cannot defeat a damages claim brought within the period prescribed by the Copyright Act’s statute of limitations. Petrella, 572 U. S., at ___–___ (slip op., at 11–14). And in so holding, we spoke in broad terms. See id., at ___ (slip op., at 13) (“[I]n the face of a statute of limitations enacted by Congress, laches cannot be invoked to bar legal relief”). Petrella’s holding rested on both separation-of-powers principles and the traditional role of laches in equity. Laches provides a shield against untimely claims, id., at ___ (slip op., at 19), and statutes of limitations serve a similar function. When Congress enacts a statute of limitations, it speaks directly to the issue of timeliness and provides a rule for determining whether a claim is timely enough to permit relief. Id., at ___ (slip op., at 11). The enactment of a statute of limitations necessarily reflects a congressional decision that the timeliness of covered claims is better judged on the basis of a generally hard and fast rule rather than the sort of case-specific judicial determination that occurs when a laches defense is asserted. Therefore, applying laches within a limitations period specified by Congress would give judges a “legislation-overriding” role that is beyond the Judiciary’s power. Id., at ___ (slip op., at 14). As we stressed in Petrella, “courts are not at liberty to jettison Congress’ judgment on the timeliness of suit.” Id., at ___ (slip op., at 1). Applying laches within the limitations period would also clash with the purpose for which the defense developed in the equity courts. As Petrella recounted, the “principal application” of laches “was, and remains, to claims of an equitable cast for which the Legislature has provided no fixed time limitation.” Id., at ___ (slip op., at 12); see also R. Weaver, E. Shoben, & M. Kelly, Principles of Remedies Law 21 (2d ed. 2011); 1 Dobbs §2.4(4), at 104; 1 J. Story, Commentaries on Equity Jurisprudence §55(a), p. 73 (2d ed. 1839). Laches is a gap-filling doctrine, and where there is a statute of limitations, there is no gap to fill.[4] Petrella, supra, at ___ (slip op., at 14); see also 1 Dobbs §2.4(4), at 108 (“[I]f the plaintiff has done only what she is permitted to do by statute, and has not misled the defendant [so as to trigger equitable estoppel], the basis for barring the plaintiff seems to have disappeared”). With Petrella’s principles in mind, we turn to the present dispute. III A Although the relevant statutory provisions in Petrella and this case are worded differently, Petrella’s reasoning easily fits the provision at issue here. As noted, the statute in Petrella precludes a civil action for copyright infringement “unless it is commenced within three years after the claim accrued.” 17 U. S. C. §507(b). We saw in this language a congressional judgment that a claim filed within three years of accrual cannot be dismissed on timeliness grounds. 572 U. S., at ___ (slip op., at 11); see also id., at ___–___ (slip op., at 14–15). The same reasoning applies in this case. Section 286 of the Patent Act provides: “Except as otherwise provided by law, no recovery shall be had for any infringement committed more than six years prior to the filing of the complaint or counterclaim for infringement in the action.” By the logic of Petrella, we infer that this provision represents a judgment by Congress that a patentee may recover damages for any infringement committed within six years of the filing of the claim. B First Quality contends that this case differs from Petrella because §286 of the Patent Act is not a true statute of limitations. A true statute of limitations, we are told, “runs forward from the date a cause of action accrues,” but §286 “runs backward from the time of suit.” Brief for Respondents 41. Petrella cannot reasonably be distinguished on this ground. First Quality thinks it critical that §286 “runs backward from the time of suit,” Brief for Respondents 41, but Petrella described the Copyright Act’s statute of limitations in almost identical terms. We said that this provision “allows plaintiffs . . . to gain retrospective relief running only three years back from the date the complaint was filed.” 572 U. S., at ___ (slip op., at 6–7) (emphasis added). See also id., at ___ (slip op., at 11) (“[A] successful plaintiff can gain retrospective relief only three years back from the time of suit”). And we described the Copyright Act’s statute of limitations as “a three-year look-back limitations period.” Id., at ___ (slip op., at 4). First Quality contends that the application of a true statute of limitations, like the defense of laches (but unlike §286), takes into account the fairness of permitting the adjudication of a particular plaintiff’s claim. First Quality argues as follows: “When Congress enacts [a true statute of limitations], it can be viewed as having made a considered judgment about how much delay may occur after a plaintiff knows of a cause of action (i.e., after accrual) before the plaintiff must bring suit—thus potentially leaving no room for judges to evaluate the reasonableness of a plaintiff’s delay on a case-by-case basis under laches.” Brief for Respondents 42. According to First Quality, §286 of the Patent Act is different because it “turns only on when the infringer is sued, regardless of when the pat-entee learned of the infringement.” Ibid. This argument misunderstands the way in which statutes of limitations generally work. First Quality says that the accrual of a claim, the event that triggers the running of a statute of limitations, occurs when “a plaintiff knows of a cause of action,” ibid., but that is not ordinarily true. As we wrote in Petrella, “[a] claim ordinarily accrues ‘when [a] plaintiff has a complete and present cause of action.’ ” 572 U. S., at ___ (slip op., at 4); see Graham County Soil & Water Conservation Dist. v. United States ex rel. Wilson, 545 U. S. 409 –419 (2005). While some claims are subject to a “discovery rule” under which the limitations period begins when the plaintiff discovers or should have discovered the injury giving rise to the claim, that is not a universal feature of statutes of limitations. See, e.g., ibid. (limitations period in 31 U. S. C. §3731(b)(1) begins to run when the cause of action accrues); TRW Inc. v. Andrews, 534 U. S. 19, 28 (2001) (same with regard to 15 U. S. C. §1681p). And in Petrella, we specifically noted that “we have not passed on the question” whether the Copyright Act’s statute of limitations is governed by such a rule. 572 U. S., at ___, n. 4 (slip op., at 4, n. 4). For these reasons, Petrella cannot be dismissed as applicable only to what First Quality regards as true statutes of limitations. At least for present purposes, nothing depends on this debatable taxonomy. Compare Automobile Workers v. Hoosier Cardinal Corp., 383 U. S. 696, 704 (1966) (describing §286 as “enacting a uniform period of limitations”); 1 Dobbs §2.4(4), at 107, and n. 33 (same), with A. Stucki Co. v. Buckeye Steel Castings Co., 963 F. 2d 360, 363, n. 3 (CA Fed. 1992) (Section 286 “is not, strictly speaking, a statute of limitations”); Standard Oil Co. v. Nippon Shokubai Kagaku Co., Ltd., 754 F. 2d 345, 348 (CA Fed. 1985) (“[Section] 286 cannot properly be called a ‘statute of limitations’ in the sense that it defeats the right to bring suit”). C The Federal Circuit based its decision on a different footing. Section 286 of the Patent Act begins with the phrase “[e]xcept as otherwise provided by law,” and according to the Federal Circuit, §282 of the Act is a provision that provides otherwise. In its view, §282 creates an exception to §286 by codifying laches as a defense to all patent infringement claims, including claims for damages suffered within §286’s 6-year period. 807 F. 3d, at 1329–1330. Section 282(b), which does not specifically mention laches, provides in relevant part as follows: “The following shall be defenses in any action involving the validity or infringement of a patent and shall be pleaded: “(1) Noninfringement, absence of liability for infringement or unenforceability.” The en banc majority below never identified which word or phrase in §282 codifies laches as a defense, but First Quality argues that laches falls within §282(b)(1) because laches is a defense based on “unenforceability.” Brief for Respondents 28–33. SCA disputes this interpretation of §282(b)(1), arguing that laches does not make a patent categorically unenforceable. Reply Brief 6–8; see Aukerman, 960 F. 2d, at 1030 (“Recognition of laches as a defense . . . does not affect the general enforceability of the patent against others”). We need not decide this question. Even if we assume for the sake of argument that §282(b)(1) incorporates a laches defense of some dimension, it does not necessarily follow that this defense may be invoked to bar a claim for damages incurred within the period set out in §286. Indeed, it would be exceedingly unusual, if not unprecedented, if Congress chose to include in the Patent Act both a statute of limitations for damages and a laches provision applicable to a damages claim. Neither the Federal Circuit, nor First Quality, nor any of First Quality’s amici has identified a single federal statute that provides such dual protection against untimely claims. D In holding that Congress codified a damages-limiting laches defense, the Federal Circuit relied on patent cases decided by the lower courts prior to the enactment of the Patent Act. After surveying these cases, the Federal Circuit concluded that by 1952 there was a well-established practice of applying laches to such damages claims and that Congress, in adopting §282, must have chosen to codify such a defense in §282(b)(1). 807 F. 3d, at 1321–1329. First Quality now presses a similar argument. We have closely examined the cases on which the Federal Circuit and First Quality rely, and we find that they are insufficient to support the suggested interpretation of the Patent Act. The most prominent feature of the relevant legal landscape at the time of enactment of the Patent Act was the well-established general rule, often repeated by this Court, that laches cannot be invoked to bar a claim for damages incurred within a limitations period specified by Congress. See Holmberg v. Armbrecht, 327 U. S. 392, 395 (1946) (“If Congress explicitly puts a limit upon the time for enforcing a right which it created, there is an end of the matter”); United States v. Mack, 295 U. S. 480, 489 (1935) (“Laches within the term of the statute of limitations is no defense at law”); Wehrman v. Conklin, 155 U. S. 314, 326 (1894) (“Though a good defense in equity, laches is no defense at law. If the plaintiff at law has brought his action within the period fixed by the statute of limitations, no court can deprive him of his right to proceed”); Cross v. Allen, 141 U. S. 528, 537 (1891) (“So long as the demands secured were not barred by the statute of limitations, there could be no laches in prosecuting a suit”). Petrella confirmed and restated this long-standing rule. 572 U. S., at ___ (slip op., at 12) (“[T]his Court has cautioned against invoking laches to bar legal relief”). If Congress examined the relevant legal landscape when it adopted 35 U. S. C. §282, it could not have missed our cases endorsing this general rule. The Federal Circuit and First Quality dismiss the significance of this Court’s many reiterations of the general rule because they were not made in patent cases. But as the dissenters below noted, “[p]atent law is governed by the same common-law principles, methods of statutory interpretation, and procedural rules as other areas of civil litigation.” 807 F. 3d, at 1333 (opinion of Hughes, J.). In light of the general rule regarding the relationship between laches and statutes of limitations, nothing less than a broad and unambiguous consensus of lower court decisions could support the inference that §282(b)(1) codifies a very different patent-law-specific rule. No such consensus is to be found.[5] IV The pre-1952 cases on which First Quality relies fall into three groups: (1) cases decided by equity courts before 1938; (2) cases decided by law courts before 1938; and (3) cases decided after the merger of equity and law in 1938. We will discuss each group separately. A Pre-1938 equity cases The pre-1938 equity cases are unpersuasive for several, often overlapping reasons. Many do not even reveal whether the plaintiff asked for damages. Indeed, some say nothing at all about the form of relief that was sought, see, e.g., Cummings v. Wilson & Willard Mfg. Co., 4 F. 2d 453 (CA9 1925), and others state only that the plaintiff wanted an accounting of profits, e.g., Westco-Chippewa Pump Co. v. Delaware Elec. & Supply Co., 64 F. 2d 185, 186 (CA3 1933); Wolf Mineral Process Corp. v. Minerals Separation North Am. Corp., 18 F. 2d 483, 484 (CA4 1927). The equitable remedy of an accounting, however, was not the same as damages. The remedy of damages seeks to compensate the victim for its loss, whereas the remedy of an accounting, which Congress abolished in the patent context in 1946,[6] sought disgorgement of ill-gotten profits. See Birdsall v. Coolidge, 93 U. S. 64 –69 (1876); 1 Dobbs §4.3(5), at 611 (“Accounting holds the defendant liable for his profits, not for damages”); A. Walker, Patent Laws §573, p. 401 (1886) (distinguishing between the two remedies); G. Curtis, Law of Patents §341(a), p. 461 (4th ed. 1873); 2 J. Pomeroy, Treatise on Equitable Remedies §568, p. 977 (1905). First Quality argues that courts sometimes used the term “accounting” imprecisely to refer to both an accounting of profits and a calculation of damages, Brief for Respondents 19–20, but even if that is true, this loose usage shows only that a reference to “accounting” might refer to damages. For that reason, the Federal Circuit did not rely on cases seeking only an accounting, 807 F. 3d, at 1326, n. 7, and we likewise exclude such cases from our analysis. Turning to the cases that actually refer to damages, we note that many of the cases merely suggest in dicta that laches might limit recovery of damages. See, e.g., Hartford-Empire Co. v. Swindell Bros., 96 F. 2d 227, 233, modified on reh’g, 99 F. 2d 61 (CA4 1938). Such dicta “settles nothing.” Jama v. Immigration and Customs Enforcement, 543 U. S. 335 , n. 12 (2005). See also Hartford Underwriters Ins. Co. v. Union Planters Bank, N. A., 530 U. S. 1 –10 (2000); Metropolitan Stevedore Co. v. Rambo, 515 U. S. 291, 300 (1995) . As for the cases in which laches was actually held to bar a claim for damages, e.g., Wolf, Sayer & Heller v. United States Slicing Mach. Co., 261 F. 195, 197–198 (CA7 1919); A. R. Mosler & Co. v. Lurie, 209 F. 364, 369–370 (CA2 1913), these cases are too few to establish a settled, national consensus. See Hartford Underwriters, supra, at 10. Moreover, the most that can possibly be gathered from a pre-1938 equity case is that laches could defeat a damages claim in an equity court, not that the defense could en-tirely prevent a patentee from recovering damages. Before 1870, a patentee wishing to obtain both an injunction against future infringement and damages for past infringement was required to bring two suits, one in an equity court (where injunctive relief but not damages was available), and one in a court of law (where damages but not injunctive relief could be sought). See Beauchamp, The First Patent Litigation Explosion, 125 Yale L. J. 848, 913–914 (2016). To rectify this situation, Congress en-acted a law in 1870 authorizing equity courts to award dam-ages in patent-infringement actions. Rev. Stat. §4921. And although statutes of limitations did not generally apply in equity, Congress in 1897 enacted a statute that, like the current §286, imposed a 6-year limitations period for damages claims and made that statute applicable in both law and equity. §6, 29Stat. 694. Pointing to cases decided between 1897 and 1938 in which an equity court permitted a defendant in an infringement case to invoke the defense of laches, First Quality contends that Congress, aware of these cases, assumed that the 1952 Act would likewise allow a defendant in an infringement case to claim laches with respect to a claim for damages occurring within a limitations period. This argument overlooks the fact that a patentee, during the period in question, could always sue for damages in law, where the equitable doctrine of laches did not apply, and could thus avoid any possible laches defense. Thus, accepting First Quality’s argument would not return patentees to the position they held from 1897 to 1938. Instead, it would go much further and permit laches entirely to defeat claims like SCA’s.[7] B Pre-1938 claims at law First Quality cites three Court of Appeals cases in which laches was raised in a proceeding at law and in which, according to First Quality, the defense was held to bar a damages claim. See Universal Coin Lock Co. v. American Sanitary Lock Co., 104 F. 2d 781 (CA7 1939); Banker v. Ford Motor Co., 69 F. 2d 665 (CA3 1934); Ford v. Huff, 296 F. 652 (CA5 1924). But even if all of these cases squarely held that laches could be applied to a damages claim at law within the limitations period, they would still constitute only a handful of decisions out of the corpus of pre-1952 patent cases, and that would not be enough to overcome the presumption that Congress legislates against the background of general common-law principles. See H. McClintock, Handbook of the Principles of Equity §28, p. 75 (2d ed. 1948) (“The majority of the courts which have considered the question have refused to enjoin an action at law on the ground of the laches of the plaintiff at law”). In any event, these cases, like the equity cases, offer minimal support for First Quality’s position. Not one of these cases even mentions the statute of limitations. One of the three, Ford, is not even a patent infringement case; it is a breach-of-contract case arising out of a patent dispute, 296 F., at 654, and it is unclear whether the ground for decision was laches or equitable estoppel. See 807 F. 3d, at 1340 (opinion of Hughes, J.). Another, Universal Coin, applied laches to a legal damages claim without any analysis of the propriety of doing so. 104 F. 2d, at 783. First Quality protests that the paucity of supporting cases at law should not count against its argument since very few patent-infringement cases were brought at law after 1870. Brief for Respondents 25–26. But the fact remains that it is First Quality’s burden to show that Congress departed from the traditional common-law rule highlighted in our cases.[8] C Post-merger cases First Quality claims that courts continued to apply laches to damages claims after the merger of law and equity in 1938, but First Quality’s evidence is scant. During this period, two Courts of Appeals stated in dicta that laches could bar legal damages claims. See Chicago Pneumatic Tool Co. v. Hughes Tool Co., 192 F. 2d 620, 625 (CA10 1951); Shaffer v. Rector Well Equip. Co., 155 F. 2d 344, 347 (CA5 1946). And two others actually held that laches could bar a damages claim. See, e.g., Brennan v. Hawley Prods. Co., 182 F. 2d 945, 948 (CA7 1950); Lukens Steel Co. v. American Locomotive Co., 197 F. 2d 939, 941 (CA2 1952) (alternative holding). This does not constitute a settled, uniform practice of applying laches to damages claims. After surveying the pre-1952 case law, we are not convinced that Congress, in enacting §282 of the Patent Act, departed from the general rule regarding the application of laches to damages suffered within the time for filing suit set out in a statute of limitations. V First Quality’s additional arguments do not require extended discussion. First Quality points to post-1952 Court of Appeals decisions holding that laches can be invoked as a defense against a damages claim. Noting that Congress has amended §282 without altering the “ ‘unenforceability’ ” language that is said to incorporate a laches defense, First Quality contends that Congress has implicitly ratified these decisions. Brief for Respondents 35–36. We reject this argument. Nothing that Congress has done since 1952 has altered the meaning of §282. See Central Bank of Denver, N. A. v. First Interstate Bank of Denver, N. A., 511 U. S. 164, 186 (1994) ; West Virginia Univ. Hospitals, Inc. v. Casey, 499 U. S. 83 , and n. 7 (1991). First Quality and its supporting amici also make various policy arguments, but we cannot overrule Congress’s judgment based on our own policy views. We note, however, as we did in Petrella, that the doctrine of equitableestoppel provides protection against some of the problems that First Quality highlights, namely, unscrupulous patentees inducing potential targets of infringement suits to invest in the production of arguably infringing products. 572 U. S., at ___ (slip op., at 19). Indeed, the Federal Circuit held that there are genuine disputes of material fact as to whether equitable estoppel bars First Quality’s claims in this very case. See 807 F. 3d, at 1333. * * * Laches cannot be interposed as a defense against damages where the infringement occurred within the period prescribed by §286. The judgment of the Court of Appeals is vacated in part, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 The panel reversed the District Court’s holding on equitable estoppel, concluding that there are genuine disputes of material fact relating to that defense. 767 F. 3d 1339, 1351 (2014). 2 The dissenting judges concurred in the portion of the majority opinion relating to the application of laches to equitable relief. 807 F. 3d, at 1333, n. 1 (opinion of Hughes, J.); see also id., at 1331–1333 (majority opinion). We do not address that aspect of the Federal Circuit’s judgment. Nor do we address the Federal Circuit’s reversal of the District Court’s equitable estoppel holding. Id., at 1333 (reinstating original panel holding on equitable estoppel). 3 “The federal courts always had equity powers as well as law power, but they operated, until the Federal Rules of Civil Procedure, by distinctly separating equity cases and even had separate equity rules.” 1 Dobbs §2.6(1), at 148, n. 2; see also Gulfstream Aerospace Corp. v. Mayacamas Corp., 485 U. S. 271, 279 (1988) . It is in this sense that we refer in this opinion to federal courts as equity or law courts. 4 The dissent argues that there is a “gap” in the statutory scheme because the Patent Act’s statute of limitations might permit a patentee to wait until an infringing product has become successful before suing for infringement. Post, at 2–3 (opinion of Breyer, J.). We rejected a version of this argument in Petrella, 572 U. S., at ___–___ (slip op., at 16–17), and we do so here. The dissent’s argument implies that, insofar as the lack of a laches defense could produce policy outcomes judges deem undesirable, there is a “gap” for laches to fill, notwithstanding the presence of a statute of limitations. That is precisely the kind of “legislation-overriding” judicial role that Petrella rightly disclaimed. Id., at ___ (slip op., at 14). 5 Because we conclude that First Quality fails to show that there was a special laches rule in the patent context, we need not address whether it is ever reasonable to assume that Congress legislated against the background of a lower court consensus rather than the contrary decisions of this Court. Cf. 807 F. 3d, at 1338 (opinion of Hughes, J.) (“For even if there were differing views in the lower [federal] courts, it would be nearly impossible to conclude that there was a uniform understanding of the common law that was inconsistent with Supreme Court precedent. In our judicial system, the Supreme Court's understanding is controlling”). 6 See 60Stat. 778; see also Aro Mfg. Co. v. Convertible Top Replacement Co., 377 U. S. 476, 505 (1964) . 7 The dissent misunderstands this point and thinks that we dismiss the relevance of the equity cases because they applied laches “to equitable claims without statutes of limitations.” Post, at 5. But we are well aware that a statute of limitations applied in equity when these cases arose. See supra, at 13. 8 For the same reason, the dissent misses the mark when it demands that we cite cases “holding that laches could not bar a patent claim for damages.” Post, at 8.

582.US.2016_15-1191

The Immigration and Nationality Act provides the framework for acquisition of U. S. citizenship from birth by a child born abroad, when one parent is a U. S. citizen and the other a citizen of another nation. Applicable to married couples, the main rule in effect at the time here relevant, 8 U. S. C. §1401(a)(7) (1958 ed.), required the U. S.-citizen parent to have ten years’ physical presence in the United States prior to the child’s birth, “at least five of which were after attaining” age 14. The rule is made applicable to unwed U. S.-citizen fathers by §1409(a), but §1409(c) creates an exception for an unwed U. S.-citizen mother, whose citizenship can be transmitted to a child born abroad if she has lived continuously in the United States for just one year prior to the child’s birth. Respondent Luis Ramón Morales-Santana, who has lived in the United States since he was 13, asserts U. S. citizenship at birth based on the U. S. citizenship of his biological father, José Morales. José moved to the Dominican Republic 20 days short of his 19th birthday, therefore failing to satisfy §1401(a)(7)’s requirement of five years’ physical presence after age 14. There, he lived with the Dominican woman who gave birth to Morales-Santana. José accepted parental responsibility and included Morales-Santana in his household; he married Morales-Santana’s mother and his name was then added to hers on Morales-Santana’s birth certificate. In 2000, the Government sought to remove Morales-Santana based on several criminal convictions, ranking him as alien because, at his time of birth, his father did not satisfy the requirement of five years’ physical presence after age 14. An immigration judge rejected Morales-Santana’s citizenship claim and ordered his removal. Morales-Santana later moved to reopen the proceedings, asserting that the Government’s refusal to recognize that he derived citizenship from his U. S.-citizen father violated the Constitution’s equal protection guarantee. The Board of Immigration Appeals denied the motion, but the Second Circuit reversed. Relying on this Court’s post-1970 construction of the equal protection principle as it bears on gender-based classifications, the court held unconstitutional the differential treatment of unwed mothers and fathers. To cure this infirmity, the Court of Appeals held that Morales-Santana derived citizenship through his father, just as he would were his mother the U. S. citizen. Held: 1. The gender line Congress drew is incompatible with the Fifth Amendment’s requirement that the Government accord to all persons “the equal protection of the laws.” Pp. 6–23. (a) Morales-Santana satisfies the requirements for third-party standing in seeking to vindicate his father’s right to equal protection. José Morales’ ability to pass citizenship to his son easily satisfies the requirement that the third party have a “ ‘close’ relationship with the person who possesses the right.” Kowalski v. Tesmer, 543 U. S. 125 . And José’s death many years before the current controversy arose is “a ‘hindrance’ to [José’s] ability to protect his own interests.” Ibid. Pp. 6–7. (b) Sections 1401 and 1409 date from an era when the Nation’s lawbooks were rife with overbroad generalizations about the way men and women are. Today, such laws receive the heightened scrutiny that now attends “all gender-based classifications,” J. E. B. v. Alabama ex rel. T. B., 511 U. S. 127 , including laws granting or denying benefits “on the basis of the sex of the qualifying parent,” Califano v. Westcott, 443 U. S. 76 . Prescribing one rule for mothers, another for fathers, §1409 is of the same genre as the classifications declared unconstitutional in Westcott; Reed v. Reed, 404 U. S. 71 –77; Frontiero v. Richardson, 411 U. S. 677 –691; Weinberger v. Wiesenfeld, 420 U. S. 636 –653; and Califano v. Goldfarb, 430 U. S. 199 –207. A successful defense therefore requires an “ ‘exceedingly persuasive justification.’ ” United States v. Virginia, 518 U. S. 515 . Pp. 7–9. (c) The Government must show, at least, that its gender-based “ ‘classification serves “important governmental objectives and that the discriminatory means employed” are “substantially related to [achieving] those objectives.” ’ ” Virginia, 518 U. S., at 533. The classification must serve an important governmental interest today, for “new insights and societal understandings can reveal unjustified inequality . . . that once passed unnoticed and unchallenged.” Obergefell v. Hodges, 576 U. S. ___, ___. Pp. 9–14. (1) At the time §1409 was enacted as part of the Nationality Act of 1940 (1940 Act), two once habitual, but now untenable, assumptions pervaded the Nation’s citizenship laws and underpinned judicial and administrative rulings: In marriage, husband is dominant, wife subordinate; unwed mother is the sole guardian of a nonmarital child. In the 1940 Act, Congress codified the mother-as-sole-guardian perception for unmarried parents. According to the stereotype, a residency requirement was justified for unwed citizen fathers, who would care little about, and have scant contact with, their nonmarital children. Unwed citizen mothers needed no such prophylactic, because the alien father, along with his foreign ways, was presumptively out of the picture. Pp. 9–13. (2) For close to a half century, this Court has viewed with suspicion laws that rely on “overbroad generalizations about the different talents, capacities, or preferences of males and females.” Virginia, 518 U. S., at 533. No “important [governmental] interest” is served by laws grounded, as §1409(a) and (c) are, in the obsolescing view that “unwed fathers [are] invariably less qualified and entitled than mothers” to take responsibility for nonmarital children. Caban v. Mohammed, 441 U. S. 380 . In light of this equal protection jurisprudence, §1409(a) and (c)’s discrete duration-of-residence requirements for mothers and fathers are anachronistic. Pp. 13–14. (d) The Government points to Fiallo v. Bell, 430 U. S. 787 ; Miller v. Albright, 523 U. S. 420 ; and Nguyen v. INS, 533 U. S. 53 , for support. But Fiallo involved entry preferences for alien children; the case did not present a claim of U. S. citizenship. And Miller and Nguyen addressed a paternal-acknowledgment requirement well met here, not the length of a parent’s prebirth residency in the United States. Pp. 14–16. (e) The Government’s suggested rationales for §1409(a) and (c)’s gender-based differential do not survive heightened scrutiny. Pp. 16–23. (1) The Government asserts that Congress sought to ensure that a child born abroad has a strong connection to the United States. The statute, the Government suggests, bracketed an unwed U. S.-citizen mother with a married couple in which both parents are U. S. citizens because she is the only legally recognized parent at birth; and aligned an unwed U. S.-citizen father with a married couple, one spouse a citizen, the other, an alien, because of the competing national influence of the alien mother. This rationale conforms to the long-held view that unwed fathers care little about their children. And the gender-based means scarcely serve the suggested congressional interest. Citizenship may be transmitted to children who have no tie to the United States so long as their U. S.-citizen mother was continuously present in the United States for one year at any point in her life prior to the child’s birth; but it may not be transmitted by a U. S.-citizen father who falls a few days short of meeting §1401(a)(7)’s longer physical-presence requirements, even if he acknowledges paternity on the day the child is born and raises the child in the United States. Pp. 17–19. (2) The Government also maintains that Congress wished to reduce the risk of statelessness for the foreign-born child of a U. S. citizen. But congressional hearings and reports offer no support for the assertion that a statelessness concern prompted the diverse physical-presence requirements. Nor has the Government shown that the risk of statelessness disproportionately endangered the children of unwed U. S.-citizen mothers. Pp. 19–23. 2. Because this Court is not equipped to convert §1409(c)’s exception for unwed U. S.-citizen mothers into the main rule displacing §§1401(a)(7) and 1409(a), it falls to Congress to select a uniform prescription that neither favors nor disadvantages any person on the basis of gender. In the interim, §1401(a)(7)’s current requirement should apply, prospectively, to children born to unwed U. S.-citizen mothers. The legislature’s intent, as revealed by the statute at hand, governs the choice between the two remedial alternatives: extending favorable treatment to the excluded class or withdrawing favorable treatment from the favored class. Ordinarily, the preferred rule is to extend favorable treatment. Westcott, 443 U. S., at 89–90. Here, however, extension to fathers of §1409(c)’s favorable treatment for mothers would displace Congress’ general rule, the longer physical-presence requirements of §§1401(a)(7) and 1409 applicable to unwed U. S.-citizen fathers and U. S.-citizen parents, male as well as female, married to the child’s alien parent. Congress’ “ ‘commitment to th[is] residual policy’ ” and “ ‘the degree of potential disruption of the statutory scheme that would occur by extension as opposed to abrogation,’ ” Heckler v. Mathews, 465 U. S. 728 , n. 5, indicate that Congress would likely have abrogated §1409(c)’s special exception, preferring to preserve “the importance of residence in this country as the talisman of dedicated attachment,” Rogers v. Bellei, 401 U. S. 815 . Pp. 23–28. 804 F. 3d 520, affirmed in part, reversed in part, and remanded. Ginsburg, J., delivered the opinion of the Court, in which Roberts, C. J., and Kennedy, Breyer, Sotomayor, and Kagan, JJ., joined. Thomas, J., filed an opinion concurring in the judgment in part, in which Alito, J., joined. Gorsuch, J., took no part in the consideration or decision of the case.

This case concerns a gender-based differential in the law governing acquisition of U. S. citizenship by a child born abroad, when one parent is a U. S. citizen, the other, a citizen of another nation. The main rule appears in 8 U. S. C. §1401(a)(7) (1958 ed.), now §1401(g) (2012 ed.). Applicable to married couples, §1401(a)(7) requires a period of physical presence in the United States for the U. S.-citizen parent. The requirement, as initially prescribed, was ten years’ physical presence prior to the child’s birth, §601(g) (1940 ed.); currently, the requirement is five years prebirth, §1401(g) (2012 ed.). That main rule is rendered applicable to unwed U. S.-citizen fathers by §1409(a). Congress ordered an exception, however, for unwed U. S.-citizen mothers. Contained in §1409(c), the exception allows an unwed mother to transmit her citizenship to a child born abroad if she has lived in the United States for just one year prior to the child’s birth. The respondent in this case, Luis Ramón Morales-Santana, was born in the Dominican Republic when his father was just 20 days short of meeting §1401(a)(7)’s physical-presence requirement. Opposing removal to the Dominican Republic, Morales-Santana asserts that the equal protection principle implicit in the Fifth Amendment[1] entitles him to citizenship stature. We hold that the gender line Congress drew is incompatible with the requirement that the Government accord to all persons “the equal protection of the laws.” Nevertheless, we cannot convert §1409(c)’s exception for unwed mothers into the main rule displacing §1401(a)(7) (covering married couples) and §1409(a) (covering unwed fathers). We must therefore leave it to Congress to select, going forward, a physical-presence requirement (ten years, one year, or some other period) uniformly applicable to all children born abroad with one U. S.-citizen and one alien parent, wed or unwed. In the interim, the Government must ensure that the laws in question are administered in a manner free from gender-based discrimination. I A We first describe in greater detail the regime Congress constructed. The general rules for acquiring U. S. citizenship are found in 8 U. S. C. §1401, the first section in Chapter 1 of Title III of the Immigration and Nationality Act (1952 Act or INA), §301, 66Stat. 235–236. Section 1401 sets forth the INA’s rules for determining who “shall be nationals and citizens of the United States at birth” by establishing a range of residency and physical-presence requirements calibrated primarily to the parents’ nationality and the child’s place of birth. §1401(a) (1958 ed.); §1401 (2012 ed.). The primacy of §1401 in the statutory scheme is evident. Comprehensive in coverage, §1401 provides the general framework for the acquisition of citizenship at birth. In particular, at the time relevant here,[2] §1401(a)(7) provided for the U. S. citizenship of “a person born outside the geographical limits of the United States and its outlying possessions of parents one of whom is an alien, and the other a citizen of the United States who, prior to the birth of such person, was physically present in the United States or its outlying possessions for a period or periods totaling not less than ten years, at least five of which were after attaining the age of fourteen years: Provided, That any periods of honorable service in the Armed Forces of the United States by such citizen parent may be included in computing the physical presence requirements of this paragraph.” Congress has since reduced the duration requirement to five years, two after age 14. §1401(g) (2012 ed.).[3] Section 1409 pertains specifically to children with unmarried parents. Its first subsection, §1409(a), incorporates by reference the physical-presence requirements of §1401, thereby allowing an acknowledged unwed citizen parent to transmit U. S. citizenship to a foreign-born child under the same terms as a married citizen parent. Section 1409(c)—a provision applicable only to unwed U. S.-citizen mothers—states an exception to the physical-presence requirements of §§1401 and 1409(a). Under §1409(c)’s exception, only one year of continuous physical presence is required before unwed mothers may pass citizenship to their children born abroad. B Respondent Luis Ramón Morales-Santana moved to the United States at age 13, and has resided in this country most of his life. Now facing deportation, he asserts U. S. citizenship at birth based on the citizenship of his biological father, José Morales, who accepted parental responsibility and included Morales-Santana in his household. José Morales was born in Guánica, Puerto Rico, on March 19, 1900. Record 55–56. Puerto Rico was then, as it is now, part of the United States, see Puerto Rico v. Sanchez Valle, 579 U. S. ___, ___–___ (2016) (slip op., at 2–4); 8 U. S. C. §1101(a)(38) (1958 ed.) (“The term United States . . . means the continental United States, Alaska, Hawaii, Puerto Rico, Guam, and the [U. S.] Virgin Islands.” (internal quotation marks omitted)); §1101(a)(38) (2012 ed.) (similar), and José became a U. S. citizen under the Organic Act of Puerto Rico, ch. 145, §5, 39Stat. 953 (a predecessor to 8 U. S. C. §1402). After living in Puerto Rico for nearly two decades, José left his childhood home on February 27, 1919, 20 days short of his 19th birthday, therefore failing to satisfy §1401(a)(7)’s requirement of five years’ physical presence after age 14. Record 57, 66. He did so to take up employment as a builder-mechanic for a U. S. company in the then-U. S.-occupied Dominican Republic. Ibid.[4] By 1959, José attested in a June 21, 1971 affidavit presented to the U. S. Embassy in the Dominican Republic, he was living with Yrma Santana Montilla, a Dominican woman he would eventually marry. Id., at 57. In 1962, Yrma gave birth to their child, respondent Luis Morales-Santana. Id., at 166–167. While the record before us reveals little about Morales-Santana’s childhood, the Dominican archives disclose that Yrma and José married in 1970, and that José was then added to Morales-Santana’s birth certificate as his father. Id., at 163–164, 167. José also related in the same affidavit that he was then saving money “for the susten[ance] of [his] family” in anticipation of undergoing surgery in Puerto Rico, where members of his family still resided. Id., at 57. In 1975, when Morales-Santana was 13, he moved to Puerto Rico, id., at 368, and by 1976, the year his father died, he was attending public school in the Bronx, a New York City borough, id., at 140, 369.[5] C In 2000, the Government placed Morales-Santana in removal proceedings based on several convictions for offenses under New York State Penal Law, all of them rendered on May 17, 1995. Id., at 426. Morales-Santana ranked as an alien despite the many years he lived in the United States, because, at the time of his birth, his father did not satisfy the requirement of five years’ physical presence after age 14. See supra, at 3–4, and n. 3. An immigration judge rejected Morales-Santana’s claim to citizenship derived from the U. S. citizenship of his father, and ordered Morales-Santana’s removal to the Dominican Republic. Record 253, 366; App. to Pet. for Cert. 45a–49a. In 2010, Morales-Santana moved to reopen the proceedings, asserting that the Government’s refusal to recognize that he derived citizenship from his U. S.-citizen father violated the Constitution’s equal protection guarantee. See Record 27, 45. The Board of Immigration Appeals (BIA) denied the motion. App. to Pet. for Cert. 8a, 42a–44a. The United States Court of Appeals for the Second Circuit reversed the BIA’s decision. 804 F. 3d 520, 524 (2015). Relying on this Court’s post-1970 construction of the equal protection principle as it bears on gender-based classifications, the court held unconstitutional the differential treatment of unwed mothers and fathers. Id., at 527–535. To cure the constitutional flaw, the court further held that Morales-Santana derived citizenship through his father, just as he would were his mother the U. S. citizen. Id., at 535–538. In so ruling, the Second Circuit declined to follow the conflicting decision of the Ninth Circuit in United States v. Flores-Villar, 536 F. 3d 990 (2008), see 804 F. 3d, at 530, 535, n. 17. We granted certiorariin Flores-Villar, but ultimately affirmed by an equally divided Court. Flores-Villar v. United States, 564 U. S. 210 (2011) ( per curiam). Taking up Morales-Santana’s request for review, 579 U. S. ___ (2016), we consider the matter anew. II Because §1409 treats sons and daughters alike, Morales-Santana does not suffer discrimination on the basis of his gender. He complains, instead, of gender-based discrimination against his father, who was unwed at the time of Morales-Santana’s birth and was not accorded the right an unwed U. S.-citizen mother would have to transmit citizenship to her child. Although the Government does not contend otherwise, we briefly explain why Morales-Santana may seek to vindicate his father’s right to the equal protection of the laws.[6] Ordinarily, a party “must assert his own legal rights” and “cannot rest his claim to relief on the legal rights . . . of third parties.” Warth v. Seldin, 422 U. S. 490, 499 (1975) . But we recognize an exception where, as here, “the party asserting the right has a close relationship with the person who possesses the right [and] there is a hindrance to the possessor’s ability to protect his own interests.” Kowalski v. Tesmer, 543 U. S. 125, 130 (2004) (quoting Powers v. Ohio, 499 U. S. 400, 411 (1991) ). José Morales’ ability to pass citizenship to his son, respondent Morales-Santana, easily satisfies the “close relationship” requirement. So, too, is the “hindrance” requirement well met. José Morales’ failure to assert a claim in his own right “stems from disability,” not “disinterest,” Miller v. Albright, 523 U. S. 420, 450 (1998) (O’Connor, J., concurring in judgment), for José died in 1976, Record 140, many years before the current controversy arose. See Hodel v. Irving, 481 U. S. 704 –712, 723, n. 7 (1987) (children and their guardians may assert Fifth Amendment rights of deceased relatives). Morales-Santana is thus the “obvious claimant,” see Craig v. Boren, 429 U. S. 190, 197 (1976) , the “best available proponent,” Singleton v. Wulff, 428 U. S. 106, 116 (1976) , of his father’s right to equal protection. III Sections 1401 and 1409, we note, date from an era when the lawbooks of our Nation were rife with overbroad generalizations about the way men and women are. See, e.g., Hoyt v. Florida, 368 U. S. 57, 62 (1961) (women are the “center of home and family life,” therefore they can be “relieved from the civic duty of jury service”); Goesaert v. Cleary, 335 U. S. 464, 466 (1948) (States may draw “a sharp line between the sexes”). Today, laws of this kind are subject to review under the heightened scrutiny that now attends “all gender-based classifications.” J. E. B. v. Alabama ex rel. T. B., 511 U. S. 127, 136 (1994) ; see, e.g., United States v. Virginia, 518 U. S. 515 –556 (1996) (state-maintained military academy may not deny admission to qualified women). Laws granting or denying benefits “on the basis of the sex of the qualifying parent,” our post-1970 decisions affirm, differentiate on the basis of gender, and therefore attract heightened review under the Constitution’s equal protection guarantee. Califano v. Westcott, 443 U. S. 76, 84 (1979) ; see id., at 88–89 (holding unconstitutional provision of unemployed-parent benefits exclusively to fathers). Accord Califano v. Goldfarb, 430 U. S. 199 –207 (1977) (plurality opinion) (holding unconstitutional a Social Security classification that denied widowers survivors’ benefits available to widows); Weinberger v. Wiesenfeld, 420 U. S. 636 –653 (1975) (holding unconstitutional a Social Security classification that excluded fathers from receipt of child-in-care benefits available to mothers); Frontiero v. Richardson, 411 U. S. 677 –691 (1973) (plurality opinion) (holding unconstitutional exclusion of married female officers in the military from benefits automatically accorded married male officers); cf. Reed v. Reed, 404 U. S. 71 –77 (1971) (holding unconstitutional a probate-code preference for a father over a mother as administrator of a deceased child’s estate).[7] Prescribing one rule for mothers, another for fathers, §1409 is of the same genre as the classifications we declared unconstitutional in Reed, Frontiero, Wiesenfeld, Goldfarb, and Westcott. As in those cases, heightened scrutiny is in order. Successful defense of legislation that differentiates on the basis of gender, we have reiterated, requires an “exceedingly persuasive justification.” Virginia, 518 U. S., at 531 (internal quotation marks omitted); Kirchberg v. Feenstra, 450 U. S. 455, 461 (1981) (internal quotation marks omitted). A The defender of legislation that differentiates on the basis of gender must show “at least that the [challenged] classification serves important governmental objectives and that the discriminatory means employed are substantially related to the achievement of those objectives.” Virginia, 518 U. S., at 533 (quoting Mississippi Univ. for Women v. Hogan, 458 U. S. 718, 724 (1982) ; alteration in original); see Tuan Anh Nguyen v. INS, 533 U. S. 53, 60, 70 (2001) . Moreover, the classification must substantially serve an important governmental interest today, for “in interpreting the [e]qual [p]rotection [guarantee], [we have] recognized that new insights and societal understandings can reveal unjustified inequality . . . that once passed unnoticed and unchallenged.” Obergefell v. Hodges, 576 U. S. ___, ___ (2015) (slip op., at 20). Here, the Government has supplied no “exceedingly persuasive justification,” Virginia, 518 U. S., at 531 (internal quotation marks omitted), for §1409(a) and (c)’s “gender-based” and “gender-biased” disparity, Westcott, 443 U. S., at 84 (internalquotation marks omitted). 1 History reveals what lurks behind §1409. Enacted in the Nationality Act of 1940 (1940 Act), see 54Stat. 1139–1140, §1409 ended a century and a half of congressional silence on the citizenship of children born abroad to unwed parents.[8] During this era, two once habitual, but now untenable, assumptions pervaded our Nation’s citizenship laws and underpinned judicial and administrative rulings: In marriage, husband is dominant, wife subordinate; unwed mother is the natural and sole guardian of a nonmarital child. Under the once entrenched principle of male dominance in marriage, the husband controlled both wife and child. “[D]ominance [of] the husband,” this Court observed in 1915, “is an ancient principle of our jurisprudence.” Mackenzie v. Hare, 239 U. S. 299, 311 (1915) .[9] See generally Brief for Professors of History et al. as Amici Curiae 4–15. Through the early 20th century, a male citizen automatically conferred U. S. citizenship on his alien wife. Act of Feb. 10, 1855, ch. 71, §2, 10Stat. 604; see Kelly v. Owen, 7 Wall. 496, 498 (1869) (the 1855 Act “confers the privileges of citizenship upon women married to citizens of the United States”); C. Bredbenner, A Nationality of Her Own:Women, Marriage, and the Law of Citizenship 15–16, 20–21 (1998). A female citizen, however, was incapable of conferring citizenship on her husband; indeed, she was subject to expatriation if she married an alien.[10] The family of a citizen or a lawfully admitted permanent resident enjoyed statutory exemptions from entry requirements, but only if the citizen or resident was male. See, e.g., Act of Mar. 3, 1903, ch. 1012, §37, 32Stat. 1221 (wives and children entering the country to join permanent-resident aliens and found to have contracted contagious diseases during transit shall not be deported if the diseases were easily curable or did not present a danger to others); S. Rep. No. 1515, 81st Cong., 2d Sess., 415–417 (1950) (wives exempt from literacy and quota requirements). And from 1790 until 1934, the foreign-born child of a married couple gained U. S. citizenship only through the father.[11] For unwed parents, the father-controls tradition never held sway. Instead, the mother was regarded as the child’s natural and sole guardian. At common law, the mother, and only the mother, was “bound to maintain [a nonmarital child] as its natural guardian.” 2 J. Kent, Commentaries on American Law *215–*216 (8th ed. 1854); see Nguyen, 533 U. S., at 91–92 (O’Connor, J., dissenting). In line with that understanding, in the early 20th century, the State Department sometimes permitted unwed mothers to pass citizenship to their children, despite the absence of any statutory authority for the practice. See Hearings on H. R. 6127 before the House Committee on Immigration and Naturalization, 76th Cong., 1st Sess., 43, 431 (1940) (hereinafter 1940 Hearings); 39 Op. Atty. Gen. 397, 397–398 (1939); 39 Op. Atty. Gen. 290, 291 (1939). See also Collins, Illegitimate Borders: Jus Sanguinis Citizenship and the Legal Construction of Family, Race, and Nation, 123 Yale L. J. 2134, 2199–2205 (2014) (hereinafter Collins). In the 1940 Act, Congress discarded the father-controls assumption concerning married parents, but codified the mother-as-sole-guardian perception regarding unmarried parents. The Roosevelt administration, which proposed §1409, explained: “[T]he mother [of a nonmarital child] stands in the place of the father . . . [,] has a right tothe custody and control of such a child as against the putative father, and is bound to maintain it as its natu-ral guardian.” 1940 Hearings 431 (internal quotation marks omitted). This unwed-mother-as-natural-guardian notion renders §1409’s gender-based residency rules understandable. Fearing that a foreign-born child could turn out “more alien than American in character,” the administration believed that a citizen parent with lengthy ties to the United States would counteract the influence of the alien parent. Id., at 426–427. Concern about the attachment of foreign-born children to the United States explains the treatment of unwed citizen fathers, who, according to the familiar stereotype, would care little about, and have scant contact with, their nonmarital children. For unwed citizen mothers, however, there was no need for a prolonged residency prophylactic: The alien father, who might transmit foreign ways, was presumptively out of the picture. See id., at 431; Collins 2203 (in “nearly uniform view” of U. S. officials, “almost invariably,” the mother alone “concern[ed] herself with [a nonmarital] child” (internal quotation marks omitted)). 2 For close to a half century, as earlier observed, see supra, at 7–8, this Court has viewed with suspicion laws that rely on “overbroad generalizations about the different talents, capacities, or preferences of males and females.” Virginia, 518 U. S., at 533; see Wiesenfeld, 420 U. S., at 643, 648. In particular, we have recognized that if a “statutory objective is to exclude or ‘protect’ members of one gender” in reliance on “fixed notions concerning [that gender’s] roles and abilities,” the “objective itself is illegitimate.” Mississippi Univ. for Women, 458 U. S., at 725. In accord with this eventual understanding, the Court has held that no “important [governmental] interest” is served by laws grounded, as §1409(a) and (c) are, in the obsolescing view that “unwed fathers [are] invariably less qualified and entitled than mothers” to take responsibility for nonmarital children. Caban v. Mohammed, 441 U. S. 380, 382, 394 (1979) .[12] Overbroad generalizations of that order, the Court has come to comprehend, have a constraining impact, descriptive though they may be of the way many people still order their lives.[13] Laws according or denying benefits in reliance on “[s]tereotypes about women’s domestic roles,” the Court has observed, may “creat[e] a self-fulfilling cycle of discrimination that force[s] women to continue to assume the role of primary family caregiver.” Nevada Dept. of Human Resources v. Hibbs, 538 U. S. 721, 736 (2003) . Correspondingly, such laws may disserve men who exercise responsibility for raising their children. See ibid. In light of the equal protection jurisprudence this Court has developed since 1971, see Virginia, 518 U. S., at 531–534, §1409(a) and (c)’s discrete duration-of-residence requirements for unwed mothers and fathers who have accepted parental responsibility is stunningly anachronistic. B In urging this Court nevertheless to reject Morales-Santana’s equal protection plea, the Government cites three decisions of this Court: Fiallo v. Bell, 430 U. S. 787 (1977) ; Miller v. Albright, 523 U. S. 420 ; and Nguyen v. INS, 533 U. S. 53 . None controls this case. The 1952 Act provision at issue in Fiallo gave special immigration preferences to alien children of citizen (or lawful-permanent-resident) mothers, and to alien unwed mothers of citizen (or lawful-permanent-resident) children. 430 U. S., at 788–789, and n. 1. Unwed fathers and their children, asserting their right to equal protection, sought the same preferences. Id., at 791. Applying minimal scrutiny (rational-basis review), the Court upheld the provision, relying on Congress’ “exceptionally broad power” to admit or exclude aliens. Id., at 792, 794.[14] This case, however, involves no entry preference for aliens. Morales-Santana claims he is, and since birth has been, a U. S. citizen. Examining a claim of that order, the Court has not disclaimed, as it did in Fiallo, the application of an exacting standard of review. See Nguyen, 533 U. S., at 60–61, 70; Miller, 523 U. S., at 434–435, n. 11 (opinion of Stevens, J.). The provision challenged in Miller and Nguyen as violative of equal protection requires unwed U. S.-citizen fathers, but not mothers, to formally acknowledge parenthood of their foreign-born children in order to transmit their U. S. citizenship to those children. See §1409(a)(4) (2012 ed.).[15] After Miller produced no opinion for the Court, see 523 U. S., at 423, we took up the issue anew in Nguyen. There, the Court held that imposing a paternal-acknowledgment requirement on fathers was a justifiable, easily met means of ensuring the existence of a biological parent-child relationship, which the mother establishes by giving birth. See 533 U. S., at 62–63. Morales-Santana’s challenge does not renew the contest over §1409’s paternal-acknowledgment requirement (whether the current version or that in effect in 1970), and the Government does not dispute that Morales-Santana’s father, by marrying Morales-Santana’s mother, satisfied that requirement. Unlike the paternal-acknowledgment requirement at issue in Nguyen and Miller, the physical-presence requirements now before us relate solely to the duration of the parent’s prebirth residency in the United States, notto the parent’s filial tie to the child. As the Court of Appeals observed in this case, a man needs no more time in the United States than a woman “in order to have assimilated citizenship-related values to transmit to [his]child.” 804 F. 3d, at 531. And unlike Nguyen’s parental-acknowledgment requirement, §1409(a)’s age-calibrated physical-presence requirements cannot fairly be described as “minimal.” 533 U. S., at 70. C Notwithstanding §1409(a) and (c)’s provenance in traditional notions of the way women and men are, the Government maintains that the statute serves two important objectives: (1) ensuring a connection between the child to become a citizen and the United States and (2) preventing “statelessness,” i.e., a child’s possession of no citizenship at all. Even indulging the assumption that Congress intended §1409 to serve these interests, but see supra, at 9–13, neither rationale survives heightened scrutiny. 1 We take up first the Government’s assertion that §1409(a) and (c)’s gender-based differential ensures that a child born abroad has a connection to the United States of sufficient strength to warrant conferral of citizenship at birth. The Government does not contend, nor could it, that unmarried men take more time to absorb U. S. values than unmarried women do. See supra, at 16. Instead, it presents a novel argument, one it did not advance in Flores-Villar.[16] An unwed mother, the Government urges, is the child’s only “legally recognized” parent at the time of childbirth. Brief for Petitioner 9–10, 28–32.[17] An unwed citizen father enters the scene later, as a second parent. A longer physical connection to the United States is warranted for the unwed father, the Government maintains, because of the “competing national influence” of the alien mother. Id., at 9–10. Congress, the Government suggests, designed the statute to bracket an unwed U. S.-citizen mother with a married couple in which both parents are U. S. citizens,[18] and to align an unwed U. S.-citizen father with a married couple, one spouse a citizen, the other, an alien. Underlying this apparent design is the assumption that the alien father of a nonmarital child born abroad to a U. S.-citizen mother will not accept parental responsibility. For an actual affiliation between alien father and nonmarital child would create the “competing national influence” that, according to the Government, justifies imposing on unwed U. S.-citizen fathers, but not unwed U. S.-citizen mothers, lengthy physical-presence requirements. Hardly gender neutral, see id., at 9, that assumption conforms to the long-held view that unwed fathers care little about, indeed are strangers to, their children. See supra, at 9–13. Lump characterization of that kind, however, no longer passes equal protection inspection. See supra, at 13–14, and n. 13. Accepting, arguendo, that Congress intended the diverse physical-presence prescriptions to serve an interest in ensuring a connection between the foreign-born nonmarital child and the United States, the gender-based means scarcely serve the posited end. The scheme permits the transmission of citizenship to children who have no tie to the United States so long as their mother was a U. S. citizen continuously present in the United States for one year at any point in her life prior to the child’s birth. The transmission holds even if the mother marries the child’s alien father immediately after the child’s birth and never returns with the child to the United States. At the same time, the legislation precludes citizenship transmission by a U. S.-citizen father who falls a few days short of meeting §1401(a)(7)’s longer physical-presence requirements, even if the father acknowledges paternity on the day of the child’s birth and raises the child in the United States.[19] One cannot see in this driven-by-gender scheme the close means-end fit required to survive heightened scrutiny. See, e.g., Wengler v. Druggists Mut. Ins. Co., 446 U. S. 142 –152 (1980) (holding unconstitutional state workers’ compensation death-benefits statute presuming widows’ but not widowers’ dependence on their spouse’s earnings); Westcott, 443 U. S., at 88–89. 2 The Government maintains that Congress established the gender-based residency differential in §1409(a) and (c) to reduce the risk that a foreign-born child of a U. S. citizen would be born stateless. Brief for Petitioner 33. This risk, according to the Government, was substantially greater for the foreign-born child of an unwed U. S.-citizen mother than it was for the foreign-born child of an unwed U. S.-citizen father. Ibid. But there is little reason to believe that a statelessness concern prompted the diverse physical-presence requirements. Nor has the Government shown that the risk of statelessness disproportionately endangered the children of unwed mothers. As the Court of Appeals pointed out, with one exception,[20] nothing in the congressional hearings and reports on the 1940 and 1952 Acts “refer[s] to the problem of statelessness for children born abroad.” 804 F. 3d, at 532–533. See Collins 2205, n. 283 (author examined “many hundreds of pre-1940 administrative memos . . . defend[ing] or explain[ing] recognition of the nonmarital foreign-born children of American mothers as citizens”; of the hundreds, “exactly one memo by a U. S. official . . . mentions the risk of statelessness for the foreign-born nonmarital children of American mothers as a concern”). Reducing the incidence of statelessness was the express goal of other sections of the 1940 Act. See 1940 Hearings 430 (“stateless[ness]” is “object” of section on foundlings). The justification for §1409’s gender-based dichotomy, however, was not the child’s plight, it was the mother’s role as the “natural guardian” of a nonmarital child. See supra, at 9–13; Collins 2205 (“[T]he pronounced gender asymmetry of the Nationality Act’s treatment of nonmarital foreign-born children of American mothers and fathers was shaped by contemporary maternalist norms regarding the mother’s relationship with her nonmarital child—and the father’s lack of such a relationship.”). It will not do to “hypothesiz[e] or inven[t]” governmental purposes for gender classifications “post hoc in response to litigation.” Virginia, 518 U. S., at 533, 535–536. Infecting the Government’s risk-of-statelessness argument is an assumption without foundation. “[F]oreign laws that would put the child of the U. S.-citizen mother at risk of statelessness (by not providing for the child to acquire the father’s citizenship at birth),” the Government asserts, “would protect the child of the U. S.-citizen father against statelessness by providing that the child would take his mother’s citizenship.” Brief for Petitioner 35. The Government, however, neglected to expose this supposed “protection” to a reality check. Had it done so, it would have recognized the formidable impediments placed by foreign laws on an unwed mother’s transmission of citizenship to her child. See Brief for Scholars on Statelessness as Amici Curiae 13–22, A1–A15. Experts who have studied the issue report that, at the time relevant here, in “at least thirty countries,” citizen mothers generally could not transmit their citizenship to nonmarital children born within the mother’s country. Id., at 14; see id., at 14–17. “[A]s many as forty-five countries,” they further report, “did not permit their female citizens to assign nationality to a nonmarital child born outside the subject country with a foreign father.” Id., at 18; see id., at 18–21. In still other countries, they also observed, there was no legislation in point, leaving the nationality of nonmarital children uncertain. Id., at 21–22; see Sandifer, A Comparative Study of Laws Relating to Nationality at Birth and to Loss of Nationality, 29 Am. J. Int’l L. 248, 256, 258 (1935) (of 79 nations studied, about half made no specific provision for the nationality of nonmarital children). Taking account of the foreign laws actually in force, these experts concluded, “the risk of parenting stateless children abroad was, as of [1940 and 1952], and remains today, substantial for unmarried U. S. fathers, a risk perhaps greater than that for unmarried U. S. mothers.” Brief for Scholars on Statelessness as Amici Curiae 9–10; see id., at 38–39. One can hardly characterize as gender neutral a scheme allegedly attending to the risk of statelessness for children of unwed U. S.-citizen mothers while ignoring the same risk for children of unwed U. S.-citizen fathers. In 2014, the United Nations High Commissioner for Refugees (UNHCR) undertook a ten-year project to eliminate statelessness by 2024. See generally UNHCR, Ending Statelessness Within 10 Years, online at http://www.unhcr.org/en-us/protection/statelessness/546217229/special-report-ending-statelessness-10-years.html (all Internet materials as last visited June 9, 2017). Cognizant that discrimination against either mothers or fathers in citizenship and nationality laws is a major cause of statelessness, the Commissioner has made a key component of its project the elimination of gender discrimination in such laws. UNHCR, The Campaign To End Statelessness: April 2016 Update 1 (referring to speech of UNHCR “highlight[ing] the issue of gender discrimination in the nationality laws of 27 countries—a major cause of statelessness globally”), online at http://www.unhcr.org/ibelong/wp-content / uploads / Campaign-Update-April-2016.pdf;UNHCR, Background Note on Gender Equality, Nationality Laws and Statelessness 2016, p. 1 (“Ensuring gender equality in nationality laws can mitigate the risks of statelessness.”), online at http://www.refworld.org/docid/56de83ca4.html. In this light, we cannot countenance risk of statelessness as a reason to uphold, rather than strike out, differential treatment of unmarried women andmen with regard to transmission of citizenship to their children. In sum, the Government has advanced no “exceedingly persuasive” justification for §1409(a) and (c)’s gender-specific residency and age criteria. Those disparate criteria, we hold, cannot withstand inspection under a Constitution that requires the Government to respect the equal dignity and stature of its male and female citizens.[21] IV While the equal protection infirmity in retaining a longer physical-presence requirement for unwed fathers than for unwed mothers is clear, this Court is not equipped to grant the relief Morales-Santana seeks, i.e., extending to his father (and, derivatively, to him) the benefit of the one-year physical-presence term §1409(c) reserves for unwed mothers. There are “two remedial alternatives,” our decisions instruct, Westcott, 443 U. S., at 89 (quoting Welsh v. United States, 398 U. S. 333, 361 (1970) (Harlan, J., concurringin result)), when a statute benefits one class (in this case, unwed mothers and their children), as §1409(c) does, and excludes another from the benefit (here, unwed fathers and their children). “[A] court may either declare [the statute] a nullity and order that its benefits not extend to the class that the legislature intended to benefit, or it may extend the coverage of the statute to include those who are aggrieved by exclusion.” Westcott, 443 U. S., at 89 (quoting Welsh, 398 U. S., at 361 (opinion of Harlan, J.)).[22] “[W]hen the ‘right invoked is that to equal treatment,’ the appropriate remedy is a mandate of equal treatment, a result that can be accomplished by withdrawal of benefits from the favored class as well as by extension of benefits to the excluded class.” Heckler v. Mathews, 465 U. S. 728, 740 (1984) (quoting Iowa-Des Moines Nat. Bank v. Bennett, 284 U. S. 239, 247 (1931) ; emphasis deleted). “How equality is accomplished . . . is a matter on which the Constitution is silent.” Levin v. Commerce Energy, Inc., 560 U. S. 413 –427 (2010).[23] The choice between these outcomes is governed by the legislature’s intent, as revealed by the statute at hand. See id., at 427 (“On finding unlawful discrimination, . . . courts may attempt, within the bounds of their institutional competence, to implement what the legislature would have willed had it been apprised of the constitutional infirmity.”). See also Ayotte v. Planned Parenthood of Northern New Eng., 546 U. S. 320, 330 (2006) (“the touchstone for any decision about remedy is legislative intent”).[24] Ordinarily, we have reiterated, “extension, rather than nullification, is the proper course.” Westcott, 443 U. S., at 89. Illustratively, in a series of cases involving federal financial assistance benefits, the Court struck discriminatory exceptions denying benefits to discrete groups, which meant benefits previously denied were extended. See, e.g., Goldfarb, 430 U. S., at 202–204, 213–217 (plurality opinion) (survivors’ benefits), aff’g 396 F. Supp. 308, 309 (EDNY 1975) (per curiam); Jimenez v. Weinberger, 417 U. S. 628 –631, and n. 2, 637–638 (1974) (disability benefits); Department of Agriculture v. Moreno, 413 U. S. 528 –530, 538 (1973) (food stamps); Frontiero, 411 U. S., at 678–679, and n. 2, 691, and n. 25 (plurality opinion) (military spousal benefits). Here, however, the discriminatory exception consists of favorable treatment for a discrete group (a shorter physical-presence requirement for unwed U. S.-citizen mothers giving birth abroad). Following the same approach as in those benefits cases—striking the discriminatory exception—leads here to extending the general rule of longer physical-presence requirements to cover the previously favored group. The Court has looked to Justice Harlan’s concurring opinion in Welsh v. United States, 398 U. S., at 361–367, in considering whether the legislature would have struck an exception and applied the general rule equally to all, or instead, would have broadened the exception to cure the equal protection violation. In making this assessment, a court should “ ‘measure the intensity of commitment to the residual policy’ ”—the main rule, not the exception—“ ‘and consider the degree of potential disruption of the statutory scheme that would occur by extension as opposed to abrogation.’ ” Heckler, 465 U. S., at 739, n. 5 (quoting Welsh, 398 U. S., at 365 (opinion of Harlan, J.)). The residual policy here, the longer physical-presence requirement stated in §§1401(a)(7) and 1409, evidences Congress’ recognition of “the importance of residence in this country as the talisman of dedicated attachment.” Rogers v. Bellei, 401 U. S. 815, 834 (1971) ; see Weedin v. Chin Bow, 274 U. S. 657 –666 (1927) (Congress “attached more importance to actual residence in the United States as indicating a basis for citizenship than it did to descent. . . . [T]he heritable blood of citizenship was thus associated unmistakeably with residence within the country which was thus recognized as essential to full citizenship.” (internal quotation marks omitted)). And the potential for “disruption of the statutory scheme” is large. For if §1409(c)’s one-year dispensation were extended to unwed citizen fathers, would it not be irrational to retain the longer term when the U. S.-citizen parent is married? Disadvantageous treatment of marital children in comparison to nonmarital children is scarcely a purpose one can sensibly attribute to Congress.[25] Although extension of benefits is customary in federal benefit cases, see supra, at 23–24, n. 22, 25, all indicators in this case point in the opposite direction.[26] Put to the choice, Congress, we believe, would have abrogated §1409(c)’s exception, preferring preservation of the general rule.[27] V The gender-based distinction infecting §§1401(a)(7) and 1409(a) and (c), we hold, violates the equal protection principle, as the Court of Appeals correctly ruled. For the reasons stated, however, we must adopt the remedial course Congress likely would have chosen “had it been apprised of the constitutional infirmity.” Levin, 560 U. S., at 427. Although the preferred rule in the typical case is to extend favorable treatment, see Westcott, 443 U. S., at 89–90, this is hardly the typical case.[28] Extension here would render the special treatment Congress prescribed in §1409(c), the one-year physical-presence requirement for U. S.-citizen mothers, the general rule, no longer an exception. Section 1401(a)(7)’s longer physical-presence requirement, applicable to a substantial majority of children born abroad to one U. S.-citizen parent and one foreign-citizen parent, therefore, must hold sway.[29] Going forward, Congress may address the issue and settle on a uniform prescription that neither favors nor disadvantages any person on the basis of gender. In the interim,as the Government suggests, §1401(a)(7)’s now-five-year requirement should apply, prospectively, to children born to unwed U. S.-citizen mothers. See Brief for Petitioner 12, 51; Reply Brief 19, n. 3. * * * The judgment of the Court of Appeals for the Second Circuit is affirmed in part and reversed in part, and the case is remanded for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 As this case involves federal, not state, legislation, the applicable equality guarantee is not the Fourteenth Amendment’s explicit Equal Protection Clause, it is the guarantee implicit in the Fifth Amendment’s Due Process Clause. See Weinberger v. Wiesenfeld, 420 U. S. 636 , n. 2 (1975) (“[W]hile the Fifth Amendment contains no equal protection clause, it does forbid discrimination that is so unjustifiable as to be violative of due process. This Court’s approach to Fifth Amendment equal protection claims has always been precisely the same as to equal protection claims under the Fourteenth Amendment.” (citations and internal quotation marks omitted; alteration in original)). 2 Unless otherwise noted, references to 8 U. S. C. §§1401 and 1409 are to the 1958 edition of the U. S. Code, the version in effect when respondent Morales-Santana was born. Section 1409(a) and (c) have retained their numbering; §1401(a)(7) has become §1401(g). 3 The reduction affects only children born on or after November 14, 1986. §8(r), 102Stat. 2619; see §§12–13, 100Stat. 3657. Because Morales-Santana was born in 1962, his challenge is to the ten-years, five-after-age-14 requirement applicable at the time of his birth. 4 See generally B. Calder, The Impact of Intervention: The Dominican Republic During the U. S. Occupation of 1916–1924, pp. 17, 204–205 (1984) (describing establishment of a U. S. military government in the Dominican Republic in 1916, and plans, beginning in late 1920, for withdrawal). 5 There is no question that Morales-Santana himself satisfied the five-year residence requirement that once conditioned a child’s acquisition of citizenship under §1401(a)(7). See §1401(b). 6 We explain why Morales-Santana has third-party standing in view of the Government’s opposition to such standing in Flores-Villar v. United States, 564 U. S. 210 (2011) (per curiam ). See Brief for United States, O. T. 2010, No. 09–5801, pp. 10–14. 7 See Gunther, In Search of Evolving Doctrine on a Changing Court: A Model for a Newer Equal Protection, 86 Harv. L. Rev. 1, 34 (1972) (“It is difficult to understand [Reed] without an assumption that some special sensitivity to sex as a classifying factor entered into the analysis. . . . Only by importing some special suspicion of sex-related means . . . can the [Reed] result be made entirely persuasive.”). 8 The provision was first codified in 1940 at 8 U. S. C. §605, see §205, 54Stat. 1139–1140, and recodified in 1952 at §1409, see §309, 66Stat. 238–239. For simplicity, we here use the latter designation. 9 This “ancient principle” no longer guides the Court’s jurisprudence. See Kirchberg v. Feenstra, 450 U. S. 455, 456 (1981) (invalidating, on equal protection inspection, Louisiana’s former “head and master” rule). 10 See generally C. Bredbenner, A Nationality of Her Own: Women, Marriage, and the Law of Citizenship 58–61 (1998); Sapiro, Women, Citizenship, and Nationality: Immigration and Naturalization Policies in the United States, 13 Politics & Society 1, 4–10 (1984). In 1907, Congress codified several judicial decisions and prevailing State Department views by providing that a female U. S. citizen automatically lost her citizenship upon marriage to an alien. Act of Mar. 2, 1907, ch. 2534, §3, 34Stat. 1228; see L. Gettys, The Law of Citizenship in the United States 119 (1934). This Court upheld the statute. Mackenzie v. Hare, 239 U. S. 299, 311 (1915) . 11 Act of Mar. 26, 1790, ch. 3, 1Stat. 104; Act of Jan. 29, 1795, §3, 1Stat. 415; Act of Apr. 14, 1802, §4, 2Stat. 155; Act of Feb. 10, 1855, ch. 71, §2, 10Stat. 604; see 2 J. Kent, Commentaries on American Law *52–*53 (8th ed. 1854) (explaining that the 1802 Act, by adding “fathers,” “seem[ed] to remove the doubt” about “whether the act intended by the words, ‘children of persons,’ both the father and mother, . . . or the father only”); Kerber, No Constitutional Right To Be Ladies: Women and the Obligations of Citizenship 36 (1998); Brief for Professors of History et al. as Amici Curiae 5–6. In 1934, Congress moved in a new direction by allowing a married mother to transmit her citizenship to her child. Act of May 24, 1934, ch. 344, §1, 48Stat. 797. 12 Lehr v. Robertson, 463 U. S. 248 (1983) , on which the Court relied in Tuan Anh Nguyen v. INS, 533 U. S. 53 –64 (2001), recognized that laws treating fathers and mothers differently “may not constitutionally be applied . . . where the mother and father are in fact similarly situated with regard to their relationship with the child,” Lehr, 463 U. S., at 267. The “similarly situated” condition was not satisfied in Lehr, however, for the father in that case had “never established any custodial, personal, or financial relationship” with the child. Ibid. 13 Even if stereotypes frozen into legislation have “statistical support,” our decisions reject measures that classify unnecessarily and overbroadly by gender when more accurate and impartial lines can be drawn. J. E. B. v. Alabama ex rel. T. B., 511 U. S. 127, 139, n. 11 (1994) ; see, e.g., Craig v. Boren, 429 U. S. 190 –199 (1976); Weinberger v. Wiesenfeld, 420 U. S. 636, 645 (1975) . In fact, unwed fathers assume responsibility for their children in numbers already large and notably increasing. See Brief for Population and Family Scholars as Amici Curiae 3, 5–13 (documenting that nonmarital fathers “are [often] in a parental role at the time of their child’s birth,” and “most . . . formally acknowledge their paternity either at the hospital or in the birthing center just after the child is born”); Brief for American Civil Liberties Union et al. as Amici Curiae 22 (observing, inter alia, that “[i]n 2015, fathers made up 16 percent of single parents with minor children in the United States”). 14 In 1986, nine years after the decision in Fiallo, Congress amended the governing law. The definition of “child” that included offspring of natural mothers but not fathers was altered to include children born out of wedlock who established a bona fide parent-child relationship with their natural fathers. See Immigration Reform and Control Act of 1986, §315(a), 100Stat. 3439, as amended, 8 U. S. C. §1101(b)(1)(D) (1982 ed., Supp. IV); Miller v. Albright, 523 U. S. 420 , n. 4 (1998) (opinion of Stevens, J.). 15 Section 1409(a), following amendments in 1986 and 1988, see §13, 100Stat. 3657; §8(k), 102Stat. 2618, now states: 16 In Flores-Villar, the Government asserted only the risk-of-statelessness rationale, which it repeats here. See Brief for United States, O. T. 2010, No. 09–5801, at 22–39; infra, at 19–23. 17 But see §1409(a) (unmarried U. S.-citizen father who satisfies the physical-presence requirements and, after his child is born, accepts parental responsibility transmits his citizenship to the child “as of the date of birth”). 18 When a child is born abroad to married parents, both U. S. citizens, the child ranks as a U. S. citizen at birth if either parent “has had a residence in the United States or one of its outlying possessions, prior to the birth of [the child].” §1401(a)(3) (1958 ed.); §1401(c) (2012 ed.) (same). 19 Brief for Respondent 26, n. 9, presents this example: “Child A is born in Germany and raised there by his U. S.-citizen mother who spent only a year of her life in the United States during infancy; Child B is born in Germany and is legitimated and raised in Germany by a U. S.-citizen father who spent his entire life in the United States before leaving for Germany one week before his nineteenth birthday. Notwithstanding the fact that Child A’s ‘legal relationship’ with his U. S.-citizen mother may have been established ‘at the moment of birth,’ and Child B’s ‘legal relationship’ with his U. S.-citizen father may have been established a few hours later, Child B is more likely than Child A to learn English and assimilate U. S. values. Nevertheless, under the discriminatory scheme, only Child A obtains U. S. citizenship at birth.” For another telling example, see Brief for Equality Now et al. as Amici Curiae 19–20. 20 A Senate Report dated January 29, 1952, is the sole exception. That Report relates that a particular problem of statelessness accounts for the 1952 Act’s elimination of a 1940 Act provision the State Department had read to condition a citizen mother’s ability to transmit nationality to her child on the father’s failure to legitimate the child prior to the child’s 18th birthday. See 1940 Act, §205, 54Stat. 1140 (“In the absence of . . . legitimation or adjudication [during the child’s minority], . . . the child” born abroad to an unmarried citizen mother “shall be held to have acquired at birth [the mother’s] nationality status.” (emphasis added)). The 1952 Act eliminated this provision, allowing the mother to transmit citizenship independent of the father’s actions. S. Rep. No. 1137, 82d Cong., 2d Sess., 39 (1952) (“This provision establish[es] the child’s nationality as that of the [citizen] mother regardless of legitimation or establishment of paternity . . . .” (emphasis added)). This sole reference to a statelessness problem does not touch or concern the different physical-presence requirements carried over from the 1940 Act into the 1952 Act. 21 Justice Thomas, joined by Justice Alito, sees our equal protection ruling as “unnecessary,” post, at 1, given our remedial holding. But, “as we have repeatedly emphasized, discrimination itself . . . perpetuat[es] ‘archaic and stereotypic notions’ ” incompatible with the equal treatment guaranteed by the Constitution. Heckler v. Mathews, 465 U. S. 728, 739 (1984) (quoting Mississippi Univ. for Women v. Hogan, 458 U. S. 718, 725 (1982) ). 22 After silently following the path Justice Harlan charted in Welsh v. United States, 398 U. S. 333 (1970) , in several cases involving gender-based discrimination, see, e.g., Wiesenfeld, 420 U. S., at 642, 653 (extending benefits); Frontiero v. Richardson, 411 U. S. 677 –691, and n. 25 (1973) ( plurality opinion) (same), the Court unanimously adopted his formulation in Califano v. Westcott, 443 U. S. 76 (1979) . See id., at 89–90 (opinion for the Court); id., at 94–95 (Powell, J., concurring in part and dissenting in part). The appropriate remedy, the Westcott majority held, was extension to unemployed mothers of federal family-aid unemployment benefits provided by statute only for families of unemployed fathers. Id., at 90–93. In the dissent’s view, nullification was the proper course. Id., at 94–96. 23 Because the manner in which a State eliminates discrimination “is an issue of state law,” Stanton v. Stanton, 421 U. S. 7, 18 (1975) , upon finding state statutes constitutionally infirm, we have generally remanded to permit state courts to choose between extension and invalidation. See Levin v. Commerce Energy, Inc., 560 U. S. 413, 427 (2010) . In doing so, we have been explicit in leaving open on remand the option of removal of a benefit, as opposed to extension. See, e.g., Orr v. Orr, 440 U. S. 268 –284 (1979) (leaving to state courts remedy for unconstitutional imposition of alimony obligations on husbands but not wives); Stanton, 421 U. S., at 17–18 (how to eliminate unconstitutional age differential, for child-support purposes, between male and female children, is “an issue of state law to be resolved by the Utah courts”). 24 We note, however, that a defendant convicted under a law classifying on an impermissible basis may assail his conviction without regard to the manner in which the legislature might subsequently cure the infirmity. In Grayned v. City of Rockford, 408 U. S. 104 (1972) , for example, the defendant participated in a civil rights demonstration in front of a school. Convicted of violating a local “antipicketing” ordinance that exempted “peaceful picketing of any school involved in a labor dispute,” he successfully challenged his conviction on equal protection grounds. Id., at 107 (internal quotation marks omitted). It was irrelevant to the Court’s decision whether the legislature likely would have cured the constitutional infirmity by excising the labor-dispute exemption. In fact, the legislature had done just that subsequent to the defendant’s conviction. Ibid., and n. 2. “Necessarily,” the Court observed, “we must consider the facial constitutionality of the ordinance in effect when [the defendant] was arrested and convicted.” Id., at 107, n. 2. See also Welsh, 398 U. S., at 361–364 (Harlan, J., concurring in result) (reversal required even if, going forward, Congress would cure the unequal treatment by extending rather than invalidating the criminal proscription). 25 Distinctions based on parents’ marital status, we have said, are subject to the same heightened scrutiny as distinctions based on gender. Clark v. Jeter, 486 U. S. 456, 461 (1988) . 26 In crafting the INA in 1952, Congress considered, but did not adopt, an amendment that would have applied the shorter one-year continuous physical-presence requirement now contained in §1409(c) to all foreign-born children of parents with different nationalities. See S. 2842, 82d Cong., 2d Sess., §301(a)(5) (1952). 27 Compare with the remedial issue presented here suits under Title VII of the Civil Rights Act of 1964 challenging laws prescribing terms and conditions of employment applicable to women only, e.g., minimum wage, premium pay, rest breaks, or lunch breaks. Most courts, perhaps mindful of the mixed motives implicated in passage of such legislation (some conceiving the laws as protecting women, others, as discouraging employers from hiring women), and, taking into account the economic burdens extension would impose on employers, have invalidated the provisions. See, e.g., Homemakers, Inc., of Los Angeles v. Division of Industrial Welfare, 509 F. 2d 20, 22–23 (CA9 1974), aff’g 356 F. Supp. 1111 (1973) (ND Cal. 1973); Burns v. Rohr Corp., 346 F. Supp. 994, 997–998 (SD Cal. 1972); RCA del Caribe, Inc. v. Silva Recio, 429 F. Supp. 651, 655–658 (PR 1976); Doctors Hospital, Inc. v. Recio, 383 F. Supp. 409, 417–418 (PR 1974); State v. Fairfield Communities Land Co., 260 Ark. 277, 279–281, 538 S. W. 2d 698, 699–700 (1976); Jones Metal Products Co. v. Walker, 29 Ohio St. 2d 173, 178–183, and n. 6, 281 N. E. 2d 1, 6–9, and n. 6 (1972); Vick v. Pioneer Oil Co., 569 S. W. 2d 631, 633–635 (Tex. Civ. App. 1978). 28 The Court of Appeals found the remedial issue “the most vexing problem in this case.” 804 F. 3d 520, 535 (2015). 29 That Morales-Santana did not seek this outcome does not restrain the Court’s judgment. The issue turns on what the legislature would have willed. “The relief the complaining party requests does not circumscribe this inquiry.” Levin, 560 U. S., at 427.

580.US.2016_15-5991

Petitioner Shaw used identifying numbers of a bank account belonging to bank customer Hsu in a scheme to transfer funds from that account to accounts at other institutions from which Shaw was able to obtain Hsu’s funds. Shaw was convicted of violating 18 U. S. C. §1344(1), which makes it a crime to “knowingly execut[e] a scheme . . . to defraud a financial institution.” The Ninth Circuit affirmed. Held: 1. Subsection (1) of the bank fraud statute covers schemes to deprive a bank of money in a customer’s deposit account. Shaw’s arguments in favor of his claim that subsection (1) does not apply to him because he intended to cheat only a bank depositor, not a bank, are unpersuasive. First, the bank did have property rights in Hsu’s bank deposits: When a customer deposits funds, the bank ordinarily becomes the owner of the funds, which the bank has a right to use as a source of loans that help the bank earn profits. Sometimes, the contract between the customer and the bank provides that the customer retains ownership of the funds and the bank only assumes possession; even then, the bank has a property interest in the funds because its role is akin to that of a bailee. Hence, for purposes of the bank fraud statute, a scheme fraudulently to obtain funds from a bank depositor’s account normally is also a scheme fraudulently to obtain property from a “financial institution,” at least where, as here, the defendant knew that the bank held the deposits, the funds obtained came from the deposit account, and the defendant misled the bank in order to obtain those funds. Second, Shaw may not have intended to cause the bank financial harm, but the statute, while insisting upon “a scheme to defraud,” demands neither a showing that the bank suffered ultimate financial loss nor a showing that the defendant intended to cause such loss. This Court has found no case that interprets the statute as Shaw does. Cf. Carpenter v. United States, 484 U. S. 19 . Third, that Shaw may have been ignorant of relevant bank-related property law is no defense to criminal prosecution for bank fraud. Shaw knew that the bank possessed Hsu’s account, Shaw made false statements to the bank, Shaw believed that those false statements would lead the bank to release from that account funds that ultimately and wrongfully ended up with Shaw, and the bank in fact possessed a property interest in the account. These facts are sufficient to show that Shaw knew that he was entering into a scheme to defraud the bank even if he was not aware of the niceties of bank-related property law. Cf. Pasquantino v. United States, 544 U. S. 349 –356. Fourth, Shaw mistakenly contends that the statute requires the Government to prove not just that he acted with the knowledge that he would likely harm the bank’s property interest but also that such was his purpose. This Court has found no relevant authority supporting the view that a statute making criminal the “knowin[g] execut[ion of] a scheme . . . to defraud” requires something more than knowledge. Allison Engine Co. v. United States ex rel. Sanders, 553 U. S. 662 –668; Tanner v. United States, 483 U. S. 107 –112; United States v. Cohn, 270 U. S. 339 ; and Bridges v. United States, 346 U. S. 209 –222, distinguished. Fifth, subsection (2) of the bank fraud statute, which makes criminal the use of “false or fraudulent pretenses” to obtain “property . . . under the custody or control of” a bank, may overlap with subsection (1), but it does not do so completely. Thus, it should not be read as excluding from subsection (1) applications that would otherwise fall within the scope of subsection (1), such as the conduct at issue in this case. See Loughrin v. United States, 573 U. S. ___, ___, n. 4. Finally, because the bank fraud statute is clear enough, the rule of lenity is not implicated. Pp. 2–8. 2. With regard to the parties’ dispute over whether the District Court improperly instructed the jury that a scheme to defraud a bank must be one to deceive the bank or deprive it of something of value, instead of one to deceive and deprive, the Ninth Circuit is left to determine whether that question was properly presented and if so, whether the instruction given is lawful, and, if not, whether any error was harmless in this case. Pp. 8–9. 781 F. 3d 1130, vacated and remanded. Breyer, J., delivered the opinion for a unanimous Court.

A federal statute makes it a crime “knowingly [to] execut[e] a scheme . . . to defraud a financial institution,” 18 U. S. C. §1344(1), for example, a federally insured bank, 18 U. S. C. §20. The petitioner, Lawrence Shaw, was convicted of violating this provision. He argues here that the provision does not apply to him because he intended to cheat only a bank depositor, not a bank. We do not accept his arguments. I The relevant criminal statute makes it a crime: “knowingly [to] execut[e] a scheme . . . “(1) to defraud a financial institution; or “(2) to obtain any of the moneys, funds, credits, assets, securities, or other property owned by, or under the custody or control of, a financial institution, by means of false or fraudulent pretenses, representations, or promises.” §1344. Shaw obtained the identifying numbers of a Bank of America account belonging to a bank customer, Stanley Hsu. Shaw used those numbers (and other related information) to transfer funds from Hsu’s account to other accounts at other institutions from which Shaw could obtain (and eventually did obtain) Hsu’s funds. Shaw was convicted of violating the first clause of the statute, namely,the prohibition against “defraud[ing] a financial insti-tution.” The Ninth Circuit affirmed his conviction. 781F. 3d 1130 (2015). Shaw then filed a petition for certiorari arguing that the words “scheme to defraud a financial institution” require the Government to prove that the defendant had “a specific intent not only to deceive, but also to cheat, a bank,” rather than “a non-bank third party.” Pet. for Cert. i. We granted review. II Shaw makes several related arguments in favor of his basic claim, namely, that the statute does not cover schemes to deprive a bank of customer deposits. First, he says that subsection (1) requires “an intent to wrong a victim bank [a ‘financial institution’] in its property rights . . . .” Brief for Petitioner 23. He adds that the property he took, money in Hsu’s bank account, belonged to Hsu, the bank’s customer, and that Hsu is not a “financial institution.” Id., at 25, 45. Hence Shaw’s was a scheme “designed” to obtain only “a bank customer’s property,” not “a bank’s own property.” Id., at 24–25. The basic flaw in this argument lies in the fact that the bank, too, had property rights in Hsu’s bank account. When a customer deposits funds, the bank ordinarily becomes the owner of the funds and consequently has the right to use the funds as a source of loans that help the bank earn profits (though the customer retains the right, for example, to withdraw funds). 5A Michie, Banks and Banking, ch. 9, §1, pp. 1–7 (2014) (Michie); id., §4b, at 54–58; id., §38, at 162; Phoenix Bank v. Risley, 111 U. S. 125, 127 (1884) . Sometimes, the contract between the customerand the bank provides that the customer retains ownership of the funds and the bank merely assumes possession. Michie, ch. 9, §38, at 162; Phoenix Bank, supra, at 127. But even then the bank is like a bailee, say, a garage that stores a customer’s car. Michie, ch. 9, §38, at 162. And as bailee, the bank can assert the right to possess the deposited funds against all the world but for the bailor (or, say, the bailor’s authorized agent). 8A Am. Jur. 2d, Bailment §166, pp. 685–686 (2009). This right, too, is a property right. 2 W. Blackstone, Commentaries on the Laws of England 452–454 (1766) (referring to a bailee’s right in a bailment as a “special qualified property”). Thus, Shaw’s scheme to cheat Hsu was also a scheme to deprive the bank of certain bank property rights. Hence, for purposes of the bank fraud statute, a scheme fraudulently to obtain funds from a bank depositor’s account normally is also a scheme fraudulently to obtain property from a “financial institution,” at least where, as here, the defendant knew that the bank held the deposits, the funds obtained came from the deposit account, and the defendant misled the bank in order to obtain those funds. Second, Shaw says he did not intend to cause the bank financial harm. Indeed, the parties appear to agree that, due to standard banking practices in place at the time of the fraud, no bank involved in the scheme ultimately suffered any monetary loss. Brief for Petitioner 4; Brief for United States 4, 27–28. But the statute, while insisting upon “a scheme to defraud,” demands neither a showing of ultimate financial loss nor a showing of intent to cause financial loss. Many years ago Judge Learned Hand pointed out that “[a] man is none the less cheated out of his property, when he is induced to part with it by fraud,” even if “he gets a quid pro quo of equal value.” United States v. Rowe, 56 F. 2d 747, 749 (CA2 1932). That is because “[i]t may be impossible to measure his loss by the gross scales available to a court, but he has suffered a wrong; he has lost,” for example, “his chance to bargain with the facts before him.” Ibid. Cf. O. Holmes, The Common Law 132 (1881) (“[A] man is liable to an action for deceit if he makes a false representation to another, knowing it to be false, but intending that the other should believe and act upon it”); Neder v. United States, 527 U. S. 1 –25 (1999) (bank fraud statute’s definition of fraud reflects the common law). It is consequently not surprising that, when interpreting the analogous mail fraud statute, we have held it “sufficient” that the victim (here, the bank) be “deprived of its right” to use of the property, even if it ultimately did not suffer unreimbursed loss. Carpenter v. United States, 484 U. S. 19 –27 (1987). Lower courts have explained that, where cash is taken from a bank “but the bank [is] fully insured[,] [t]he theft [is] complete when the cash [i]s taken; the fact that the bank ha[s] a contract with an insur-ance company enabling it to shift the loss to that company [is] immaterial.” United States v. Kucik, 844 F. 2d 493, 495 (CA7 1988). And commentators have made clear that “on the criminal side, it is generally held that the lack of financial loss is no defense to false pretenses.” 2 W. LaFave & A. Scott, Substantive Criminal Law §8.7(i)(3), p. 404 (1986). We have found no case from this Court interpreting the bank fraud statute as requiring that the victim bank ultimately suffer financial harm, or that the defendant intend that the victim bank suffer such harm. Third, Shaw appears to argue that, whatever the true state of property law, he did not know that the bank had a property interest in Hsu’s account; hence he could not have intended to cheat the bank of its property. Shaw did know, however, that the bank possessed Hsu’s account. He did make false statements to the bank. He did correctly believe that those false statements would lead the bankto release from that account funds that ultimately and wrongfully ended up in Shaw’s pocket. And the bank did in fact possess a property interest in the account. These facts are sufficient to show that Shaw knew he was entering into a scheme to defraud the bank even if he was not aware of the niceties of bank-related property law. To require more, i.e., to require actual knowledge of those bank-related property-law niceties, would free (or convict) equally culpable defendants depending upon their property-law expertise—an arbitrary result. We have found nocase from this Court requiring legal knowledge of the kind Shaw suggests he lacked. But we have found cases in roughly similar fraud-related contexts where this Court has asked only whether the targeted property was in fact property in the hands of the victim, not whether the defendant knew that the law would characterize the items at issue as “property.” See Pasquantino v. United States, 544 U. S. 349 –356 (2005) (Canada’s right to uncollected excise taxes on imported liquor counted as “property” for purposes of the wire fraud statute); Carpenter, supra, at 25–26 (a newspaper’s interest in the confidentiality of the contents and timing of a news column counted as property for the purposes of the mail and wire fraud statutes). We conclude that the legal ignorance that Shaw claims here is no defense to criminal prosecution for bank fraud. Fourth, Shaw argues that the bank fraud statute requires the Government to prove more than his simple knowledge that he would likely harm the bank’s property interest; in his view, the Government must prove that such was his purpose. See Voisine v. United States, 579 U. S. ___, ___ (2016) (slip op., at 4) (“knowingly” committing an assault requires an awareness “ ‘that [harm] is practically certain,’ ” whereas “purposefully” committing an assault is “to have that result as a ‘conscious object’ ” (quoting ALI, Model Penal Code §§2.02(2)(a)–(b) (1962))). Shaw adds that his purpose was to take money from Hsu; taking property from the bank was not his purpose. But the statute itself makes criminal the “knowin[g] execut[ion of] a scheme . . . to defraud.” To hold that something other than knowledge is required would assume that Congress intended to distinguish, in respect to states of mind, between (1) the fraudulent scheme, and (2) its fraudulent elements. Why would Congress wish to do so? Shaw refers us to a number of cases involving fraud against the Government and points to language in those cases suggesting that the relevant statutes required that the defendant’s purpose be to harm the statutorily protected target and not a third party. Brief for Petitioner 25–29. But in two of those cases, the fraudulent statement was made not to the Government but to the third party—a circumstance not present here. See Allison Engine Co. v. United States ex rel. Sanders, 553 U. S. 662 –668 (2008); Tanner v. United States, 483 U. S. 107 –112 (1987). In the third, the relevant portion of the statute expressly required a false statement “ ‘for the purpose . . . of . . . defrauding the Government of the United States.’ ” United States v. Cohn, 270 U. S. 339 (1926) (emphasis added). As for the fourth case, the language Shaw cites states the uncontroversial proposition that “defrauding or attempting to defraud the United States” means “fraud against the Government.” Bridges v. United States, 346 U. S. 209 –222 (1953). In any event, these cases all involved crimes of fraud targeting the Government—an area of the law with its own special rules and protections. We have found no relevant authority in the area of mail fraud, wire fraud, financial frauds, or the like supporting Shaw’s view. Fifth, Shaw, reading the bank fraud statute as a whole, urges us to compare subsection (1) with subsection (2). Supra, at 1. Subsection (2), he points out, makes criminal the use of “false or fraudulent pretenses” to obtain “property . . . under the custody or control of” a bank. And in his view that fact means that we should read subsection (1) not to apply to those circumstances. That is to say, given the language of subsection (2), efforts such as his effort fraudulently to obtain money deposited in a bank account should not fall within the scope of the subsection (1) phrase “scheme . . . to defraud a financial institution.” Brief for Petitioner 30–33. As we read the two subsections, however, they do not demand that interpretation. The two subsections overlap substantially but not completely. Subsection (2) makes criminal the use of a scheme “to obtain any of the moneys, funds, credits, assets, securities, or other property owned by, or under the custody or control of, a financial institution, by means of false or fraudulent pretenses, representations, or promises.” This language covers much that subsection (1) also covers, for example, making a false representation to a bank in order to obtain property belonging to that bank. See Loughrin v. United States, 573 U. S. ___, ___–___, n. 4 (2014) (slip op., at 6–7, n. 4) (recognizing the “substantial” overlap between the two subsections and noting that such overlap “is not uncommon in criminal statutes”). At the same time, it applies to a circumstance in which a shopper makes a false statement to a department store cashier in order to pay for goods with money “under the custody or control of a financial institution,” say, Bank A. The shopper’s false statement, though designed to obtain Bank A’s property, might well not amount to an effort (under subsection (1)) to defraud Bank A (since the statement was made not to Bank A but to an agent of the department store). Given, on the one hand, the overlap and, on the other hand, a plausible reading of the language that applies it to circumstances significantly different from those at issue here, we have no good reason to read subsection (2) as excluding from subsection (1) applications that would otherwise fall within the scope of subsection (1), such as conduct of the kind before us. Finally, Shaw asks us to apply the rule of lenity. Brief for Petitioner 40–41. We have said that the rule applies if “at the end of the process of construing what Congress has expressed,” Callanan v. United States, 364 U. S. 587, 596 (1961) , there is “ ‘a grievous ambiguity or uncertainty in the statute,’ ” Muscarello v. United States, 524 U. S. 125 –139 (1998) (quoting Staples v. United States, 511 U. S. 600, 619, n. 17 (1994) ). The statute is clear enough that we need not rely on the rule of lenity. As we have said, a deposit account at a bank counts as bank property for purposes of subsection (1). Supra, at 2–3. The defendant, in circumstances such as those present here, need not know that the deposit account is, as a legal matter, characterized as bank property. Supra, at 4–5. Moreover, in those circumstances, the Government need not prove that the defendant intended that the bank ultimately suffer monetary loss. Supra, at 3–4. Finally, the statute as applied here requires a state of mind equivalent to knowledge, not purpose. Supra, at 5–6. III Shaw further argues that the instructions the District Court gave the jury were erroneous. He points out that the District Court told the jury that the “phrase ‘scheme to defraud’ means any deliberate plan of action or course of conduct by which someoneintends to deceive, cheat, or deprive a financial institution of something of value.” App. 18 (emphasisadded). This instruction, Shaw says, could be understood as permitting the jury to find him guilty if it found no more than that his scheme was one to deceive the bank but notto “deprive” the bank of anything of value. Brief for Petitioner 22–23. The parties agree, as do we, that the scheme must be one to deceive the bank and deprive it of something of value. For reasons previously pointed out, we have held that a plan to deprive a bank of money in a customer’s deposit account is a plan to deprive the bank of “something of value” within the meaning of the bank fraud statute. The parties dispute whether the jury instruction is nonetheless ambiguous or otherwise improper. We leave to the Ninth Circuit to determine whether that question was fairly presented to that court and, if so, whether the instruction is lawful, and, if not, whether any error was harmless in this case. For these reasons, the judgment of the Ninth Circuit is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.

580.US.2016_15-866

The Copyright Act of 1976 makes “pictorial, graphic, or sculptural features” of the “design of a useful article” eligible for copyright protection as artistic works if those features “can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.” 17 U. S. C. §101. Respondents have more than 200 copyright registrations for two-dimensional designs—consisting of various lines, chevrons, and colorful shapes—appearing on the surface of the cheerleading uniforms that they design, make, and sell. They sued petitioner, who also markets cheerleading uniforms, for copyright infringement. The District Court granted petitioner summary judgment, holding that the designs could not be conceptually or physically separated from the uniforms and were therefore ineligible for copyright protection. In reversing, the Sixth Circuit concluded that the graphics could be “identified separately” and were “capable of existing independently” of the uniforms under §101. Held: A feature incorporated into the design of a useful article is eligible for copyright protection only if the feature (1) can be perceived as a two- or three-dimensional work of art separate from the useful article, and (2) would qualify as a protectable pictorial, graphic, or sculptural work—either on its own or fixed in some other tangible medium of expression—if it were imagined separately from the useful article into which it is incorporated. That test is satisfied here. Pp. 3–17. (a) Separability analysis is necessary in this case. Respondents claim that two-dimensional surface decorations are always separable, even without resorting to a §101 analysis, because they are “on a useful article” rather than “designs of a useful article.” But this argument is inconsistent with §101’s text. ”[P]ictorial” and “graphic” denote two-dimensional features such as pictures, paintings, or drawings. Thus, by providing protection for “pictorial, graphical, and sculptural works” incorporated into the “design of a useful article,” §101 necessarily contemplates that such a design can include two-dimensional features. This Court will not adjudicate in the first instance the Government’s distinct argument against applying separability analysis, which was neither raised below nor advanced here by any party. Pp. 4–6. (b) Whether a feature incorporated into a useful article “can be identified separately from,” and is “capable of existing independently of,” the article’s “utilitarian aspects” is a matter of “statutory interpretation.” Mazer v. Stein, 347 U. S. 201 . Pp. 6–10. (1) Section 101’s separate-identification requirement is met if the decisionmaker is able to look at the useful article and spot some two- or three-dimensional element that appears to have pictorial, graphic, or sculptural qualities. To satisfy the independent-existence requirement, the feature must be able to exist as its own pictorial, graphic, or sculptural work once it is imagined apart from the useful article. If the feature could not exist as a pictorial, graphic, or sculptural work on its own, it is simply one of the article’s utilitarian aspects. And to qualify as a pictorial, graphic, or sculptural work on its own, the feature cannot be a useful article or “[a]n article that is normally a part of a useful article,” §101. Neither could one claim a copyright in a useful article by creating a replica of it in another medium. Pp. 7–8. (2) The statute as a whole confirms this interpretation. Section 101, which protects art first fixed in the medium of a useful article, is essentially the mirror image of §113(a), which protects art first fixed in a medium other than a useful article and subsequently applied to a useful article. Together, these provisions make clear that copyright protection extends to pictorial, graphic, and sculptural works regardless of whether they were created as freestanding art or as features of useful articles. P. 8. (3) This interpretation is also consistent with the Copyright Act’s history. In Mazer, a case decided under the 1909 Copyright Act, the Court held that respondents owned a copyright in a statuette created for use as a lamp base. In so holding, the Court approved a Copyright Office regulation extending protection to works of art that might also serve a useful purpose and held that it was irrelevant to the copyright inquiry whether the statuette was initially created as a freestanding sculpture or as a lamp base. Soon after, the Copyright Office enacted a regulation implementing Mazer’s holding that anticipated the language of §101, thereby introducing the modern separability test to copyright law. Congress essentially lifted the language from those post-Mazer regulations and placed it in §101 of the 1976 Act. Pp. 8–10. (c) Applying the proper test here, the surface decorations on the cheerleading uniforms are separable and therefore eligible for copyright protection. First, the decorations can be identified as features having pictorial, graphic, or sculptural qualities. Second, if those decorations were separated from the uniforms and applied in another medium, they would qualify as two-dimensional works of art under §101. Imaginatively removing the decorations from the uniforms and applying them in another medium also would not replicate the uniform itself. The dissent argues that the decorations are ineligible for copyright protection because, when imaginatively extracted, they form a picture of a cheerleading uniform. Petitioner similarly claims that the decorations cannot be copyrighted because, even when extracted from the useful article, they retain the outline of a cheerleading uniform. But this is not a bar to copyright. Just as two-dimensional fine art correlates to the shape of the canvas on which it is painted, two-dimensional applied art correlates to the contours of the article on which it is applied. The only feature of respondents’ cheerleading uniform eligible for a copyright is the two-dimensional applied art on the surface of the uniforms. Respondents may prohibit the reproduction only of the surface designs on a uniform or in any other medium of expression. Respondents have no right to prevent anyone from manufacturing a cheerleading uniform that is identical in shape, cut, or dimensions to the uniforms at issue here. Pp. 10–12. (d) None of the objections raised by petitioner or the Government is meritorious. Pp. 12–17. (1) Petitioner and the Government focus on the relative utility of the plain white uniform that would remain if the designs were physically removed from the uniform. But the separability inquiry focuses on the extracted feature and not on any aspects of the useful article remaining after the imaginary extraction. The statute does not require the imagined remainder to be a fully functioning useful article at all. Nor can an artistic feature that would be eligible for copyright protection on its own lose that protection simply because it was first created as a feature of the design of a useful article, even if it makes that article more useful. This has been the rule since Mazer, and it is consistent with the statute’s explicit protection of “applied art.” In rejecting petitioner’s view, the Court necessarily abandons the distinction between “physical” and “conceptual” separability adopted by some courts and commentators. Pp. 12–15. (2) Petitioner also suggests incorporating two “objective” components into the test—one requiring consideration of evidence of the creator’s design methods, purposes, and reasons, and one looking to the feature’s marketability. The Court declines to incorporate these components because neither is grounded in the statute’s text. Pp. 15–16. (3) Finally, petitioner claims that protecting surface decorations is inconsistent with Congress’ intent to entirely exclude industrial design from copyright. But Congress has given limited copyright protection to certain features of industrial design. Approaching the statute with presumptive hostility toward protection for industrial design would undermine that choice. In any event, the test adopted here does not render the underlying uniform eligible for copyright protection. Pp. 16–17. 799 F. 3d 468, affirmed. Thomas, J., delivered the opinion of the Court, in which Roberts, C. J., and Alito, Sotomayor, and Kagan, JJ., joined. Ginsburg, J., filed an opinion concurring in the judgment. Breyer, J., filed a dissenting opinion, in which Kennedy, J., joined.

Congress has provided copyright protection for original works of art, but not for industrial designs. The line between art and industrial design, however, is often difficult to draw. This is particularly true when an industrial design incorporates artistic elements. Congress has afforded limited protection for these artistic elements by providing that “pictorial, graphic, or sculptural features” of the “design of a useful article” are eligible for copyright protection as artistic works if those features “can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.” 17 U. S. C. §101. We granted certiorari to resolve widespread disagreement over the proper test for implementing §101’s separate-identification and independent-existence requirements. 578 U. S. ___ (2016). We hold that a feature incor-porated into the design of a useful article is eligible for copyright protection only if the feature (1) can be perceived as a two- or three-dimensional work of art separate from the useful article and (2) would qualify as a protectable pictorial, graphic, or sculptural work—either on its own or fixed in some other tangible medium of expression—if it were imagined separately from the useful article into which it is incorporated. Because that test is satisfied in this case, we affirm. I Respondents Varsity Brands, Inc., Varsity Spirit Corporation, and Varsity Spirit Fashions & Supplies, Inc., design, make, and sell cheerleading uniforms. Respondents have obtained or acquired more than 200 U. S. copyright registrations for two-dimensional designs appearing on the surface of their uniforms and other garments. These designs are primarily “combinations, positionings, and arrangements of elements” that include “chevrons . . . , lines, curves, stripes, angles, diagonals, inverted [chevrons], coloring, and shapes.” App. 237. At issue in this case are Designs 299A, 299B, 074, 078, and 0815. See Appendix, infra. Petitioner Star Athletica, L. L. C., also markets and sells cheerleading uniforms. Respondents sued petitioner for infringing their copyrights in the five designs. The District Court entered summary judgment for petitioner on respondents’ copyright claims on the ground that the designs did not qualify as protectable pictorial, graphic, or sculptural works. It reasoned that the designs served the useful, or “utilitarian,” function of identifying the garments as “cheerleading uniforms” and therefore could not be “physically or conceptually” separated under §101 “from the utilitarian function” of the uniform. 2014 WL 819422, *8–*9 (WD Tenn., Mar. 1, 2014). The Court of Appeals for the Sixth Circuit reversed. 799 F. 3d 468, 471 (2015). In its view, the “graphic designs” were “separately identifiable” because the designs “and a blank cheerleading uniform can appear ‘side by side’—one as a graphic design, and one as a cheerleading uniform.” Id., at 491 (quoting Compendium of U. S. Copyright Office Practices §924.2(B) (3d ed. 2014) (Compendium)). And it determined that the designs were “ ‘capable of existing independently’ ” because they could be incorporated onto the surface of different types of garments, or hung on the wall and framed as art. 799 F. 3d, at 491, 492. Judge McKeague dissented. He would have held that, because “identifying the wearer as a cheerleader” is a utilitarian function of a cheerleading uniform and the surface designs were “integral to” achieving that function, the designs were inseparable from the uniforms. Id., at 495–496. II The first element of a copyright-infringement claim is “ownership of a valid copyright.” Feist Publications, Inc. v. Rural Telephone Service Co., 499 U. S. 340, 361 (1991) . A valid copyright extends only to copyrightable subject matter. See 4 M. Nimmer & D. Nimmer, Copyright §13.01[A] (2010) (Nimmer). The Copyright Act of 1976 defines copyrightable subject matter as “original works of authorship fixed in any tangible medium of expression.” 17 U. S. C. §102(a). “Works of authorship” include “pictorial, graphic, and sculptural works,” §102(a)(5), which the statute defines to include “two-dimensional and three-dimensional works of fine, graphic, and applied art, photographs, prints and art reproductions, maps, globes, charts, diagrams, models, and technical drawings, including architectural plans,” §101. And a work of authorship is “ ‘fixed’ in a tangible medium of expression when it[ is] embodi[ed] in a” “material objec[t] . . . from which the work can be perceived, reproduced, or otherwise communicated.” Ibid. (definitions of “fixed” and “copies”). The Copyright Act also establishes a special rule for copyrighting a pictorial, graphic, or sculptural work incorporated into a “useful article,” which is defined as “an article having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information.” Ibid. The statute does not protect useful articles as such. Rather, “the design of a useful article” is “considered a pictorial, graphical, or sculptural work only if, and only to the extent that, such design incorporates pictorial, graphic, or sculptural features that can be identified separately from, and are capable of existing independently of, the utilitarian aspects of the article.” Ibid. Courts, the Copyright Office, and commentators have described the analysis undertaken to determine whether a feature can be separately identified from, and exist independently of, a useful article as “separability.” In this case, our task is to determine whether the arrangements of lines, chevrons, and colorful shapes appearing on the surface of respondents’ cheerleading uniforms are eligible for copyright protection as separable features of the design of those cheerleading uniforms. A As an initial matter, we must address whether separability analysis is necessary in this case. 1 Respondents argue that “[s]eparability is only implicated when a [pictorial, graphic, or sculptural] work is the ‘design of a useful article.’ ” Brief for Respondents 25. They contend that the surface decorations in this case are “two-dimensional graphic designs that appear on useful articles,” but are not themselves designs of useful articles. Id., at 52. Consequently, the surface decorations are protected two-dimensional works of graphic art without regard to any separability analysis under §101. Ibid.; see 2 W. Patry, Copyright §3:151, p. 3–485 (2016) (Patry) (“Courts looking at two-dimensional design claims should not apply the separability analysis regardless of the three-dimensional form that design is embodied in”). Under this theory, two-dimensional artistic features on the surface of useful articles are “inherently separable.” Brief for Respondents 26. This argument is inconsistent with the text of §101. The statute requires separability analysis for any “pictorial, graphic, or sculptural features” incorporated into the “design of a useful article.” “Design” refers here to “the combination” of “details” or “features” that “go to make up” the useful article. 3 Oxford English Dictionary 244 (def. 7, first listing) (1933) (OED). Furthermore, the words “pictorial” and “graphic” include, in this context, two-dimensional features such as pictures, paintings, or drawings. See 4 id., at 359 (defining “[g]raphic” to mean “[o]f or pertaining to drawing or painting”); 7 id., at 830 (defining “[p]ictorial” to mean “of or pertaining to painting or drawing”). And the statute expressly defines “[p]ictorial, graphical, and sculptural works” to include “two-dimensional . . . works of . . . art.” §101. The statute thus provides that the “design of a useful article” can include two-dimensional “pictorial” and “graphic” features, and separability analysis applies to those features just as it does to three-dimensional “sculptural” features. 2 The United States makes a related but distinct argument against applying separability analysis in this case, which respondents do not and have not advanced. As part of their copyright registrations for the designs in this case, respondents deposited with the Copyright Office drawings and photographs depicting the designs incorporated onto cheerleading uniforms. App. 213–219; Appendix, infra. The Government argues that, assuming the other statutory requirements were met, respondents obtained a copyright in the deposited drawings and photographs and have simply reproduced those copyrighted works on the surface of a useful article, as they would have the exclusive right to do under the Copyright Act. See Brief for United States as Amicus Curiae 14–15, 17–22. Accordingly, the Government urges, separability analysis is unnecessary on the record in this case. We generally do not entertain arguments that were not raised below and that are not advanced in this Court by any party, Burwell v. Hobby Lobby Stores, Inc., 573 U. S. ___, ___ (2014), because “[i]t is not the Court’s usual practice to adjudicate either legal or predicate factual questions in the first instance,” CRST Van Expedited, Inc. v. EEOC, 578 U. S. ___, ___ (2016) (slip op., at 16). We decline to depart from our usual practice here. B We must now decide when a feature incorporated into a useful article “can be identified separately from” and is “capable of existing independently of” “the utilitarian aspects” of the article. This is not a free-ranging search for the best copyright policy, but rather “depends solely on statutory interpretation.” Mazer v. Stein, 347 U. S. 201, 214 (1954) . “The controlling principle in this case is the basic and unexceptional rule that courts must give effect to the clear meaning of statutes as written.” Estate of Cowart v. Nicklos Drilling Co., 505 U. S. 469, 476 (1992) . We thus begin and end our inquiry with the text, giving each word its “ordinary, contemporary, common meaning.” Walters v. Metropolitan Ed. Enterprises, Inc., 519 U. S. 202, 207 (1997) (internal quotation marks omitted). We do not, however, limit this inquiry to the text of §101 in isolation. “[I]nterpretation of a phrase of uncertain reach is not confined to a single sentence when the text of the whole statute gives instruction as to its meaning.” Maracich v. Spears, 570 U. S. ___, ___ (2013) (slip op., at 15). We thus “look to the provisions of the whole law” to determine §101’s meaning. United States v. Heirs of Boisdoré, 8 How. 113, 122 (1849). 1 The statute provides that a “pictorial, graphic, or sculptural featur[e]” incorporated into the “design of a useful article” is eligible for copyright protection if it (1) “can be identified separately from,” and (2) is “capable of existing independently of, the utilitarian aspects of the article.” §101. The first requirement—separate identification—is not onerous. The decisionmaker need only be able to look at the useful article and spot some two- or three-dimensional element that appears to have pictorial, graphic, or sculptural qualities. See 2 Patry §3:146, at 3–474 to3–475. The independent-existence requirement is ordinarily more difficult to satisfy. The decisionmaker must determine that the separately identified feature has the capacity to exist apart from the utilitarian aspects of the article. See 2 OED 88 (def. 5) (defining “[c]apable” of as “[h]aving the needful capacity, power, or fitness for”). In other words, the feature must be able to exist as its own pictorial, graphic, or sculptural work as defined in §101 once it is imagined apart from the useful article. If the feature is not capable of existing as a pictorial, graphic, or sculptural work once separated from the useful article, then it was not a pictorial, graphic, or sculptural feature of that article, but rather one of its utilitarian aspects. Of course, to qualify as a pictorial, graphic, or sculptural work on its own, the feature cannot itself be a useful article or “[a]n article that is normally a part of a useful article” (which is itself considered a useful article). §101. Nor could someone claim a copyright in a useful article merely by creating a replica of that article in some other medium—for example, a cardboard model of a car. Al-though the replica could itself be copyrightable, it would not give rise to any rights in the useful article that inspired it. 2 The statute as a whole confirms our interpretation. The Copyright Act provides “the owner of [a] copyright” with the “exclusive righ[t] . . . to reproduce the copyrighted work in copies.” §106(1). The statute clarifies that this right “includes the right to reproduce the [copyrighted] work in or on any kind of article, whether useful or otherwise.” §113(a). Section 101 is, in essence, the mirror image of §113(a). Whereas §113(a) protects a work of authorship first fixed in some tangible medium other than a useful article and subsequently applied to a useful article, §101 protects art first fixed in the medium of a useful article. The two provisions make clear that copyright protection extends to pictorial, graphic, and sculptural works regardless of whether they were created as freestanding art or as features of useful articles. The ultimate separability question, then, is whether the feature for which copyright protection is claimed would have been eligible for copyright protection as a pictorial, graphic, or sculptural work had it originally been fixed in some tangible medium other than a useful article before being applied to a useful article. 3 This interpretation is also consistent with the history of the Copyright Act. In Mazer, a case decided under the 1909 Copyright Act, the respondents copyrighted a statuette depicting a dancer. The statuette was intended for use as a lamp base, “with electric wiring, sockets and lamp shades attached.” 347 U. S., at 202. Copies of the statuette were sold both as lamp bases and separately as statuettes. Id., at 203. The petitioners copied the statuette and sold lamps with the statuette as the base. They defended against the respondents’ infringement suit by arguing that the respondents did not have a copyright in a statuette intended for use as a lamp base. Id., at 204–205. Two of Mazer’s holdings are relevant here. First, the Court held that the respondents owned a copyright in the statuette even though it was intended for use as a lamp base. See id., at 214. In doing so, the Court approved the Copyright Office’s regulation extending copyright protection to works of art that might also serve a useful purpose. See ibid. (approving 37 CFR §202.8(a) (1949) (protect-ing “works of artistic craftsmanship, in so far as theirform but not their mechanical or utilitarian aspects are concerned”)). Second, the Court held that it was irrelevant to the copyright inquiry whether the statuette was initially created as a freestanding sculpture or as a lamp base. 347 U. S., at 218–219 (“Nor do we think the subsequent registration of a work of art published as an element in a manufactured article, is a misuse of copyright. This is not different from the registration of a statuette and its later embodiment in an industrial article”). Mazer thus interpreted the 1909 Act consistently with the rule discussed above: If a design would have been copyrightable as a standalone pictorial, graphic, or sculptural work, it is copyrightable if created first as part of a useful article. Shortly thereafter, the Copyright Office enacted a regulation implementing the holdings of Mazer. See 1 Nimmer §2A.08[B][1][b] (2016). As amended, the regulation introduced the modern separability test to copyright law: “If the sole intrinsic function of an article is its utility, the fact that the article is unique and attractively shaped will not qualify it as a work of art. However, if the shape of a utilitarian article incorporates features, such as artistic sculpture, carving, or pictorial representation, which can be identified separately and are capable of existing independently as a work of art, such features will be eligible for registration.” 37 CFR §202.10(c) (1960) (punctuation altered). Congress essentially lifted the language governing protection for the design of a useful article directly from the post-Mazer regulations and placed it into §101 of the 1976 Act. Consistent with Mazer, the approach we outline today interprets §§101 and 113 in a way that would afford copyright protection to the statuette in Mazer regardless of whether it was first created as a standalone sculptural work or as the base of the lamp. See 347 U. S., at 218–219. C In sum, a feature of the design of a useful article is eligible for copyright if, when identified and imagined apart from the useful article, it would qualify as a pictorial, graphic, or sculptural work either on its own or when fixed in some other tangible medium. Applying this test to the surface decorations on the cheerleading uniforms is straightforward. First, one can identify the decorations as features having pictorial, graphic, or sculptural qualities. Second, if the arrangement of colors, shapes, stripes, and chevrons on the surface of the cheerleading uniforms were separated from the uniform and applied in another medium—for example, on a painter’s canvas—they would qualify as “two-dimensional . . . works of . . . art,” §101. And imaginatively removing the surface decorations from the uniforms and applying them in another medium would not replicate the uniform itself. Indeed, respondents have applied the designs in this case to other media of expression—different types of clothing—without replicating the uniform. See App. 273–279. The decorations are therefore separable from the uniforms and eligible for copyright protection.[1] The dissent argues that the designs are not separable because imaginatively removing them from the uniforms and placing them in some other medium of expression—a canvas, for example—would create “pictures of cheerleader uniforms.” Post, at 10 (opinion of Breyer, J.). Petitioner similarly argues that the decorations cannot be copyrighted because, even when extracted from the useful article,they retain the outline of a cheerleading uniform. Brief for Petitioner 48–49. This is not a bar to copyright. Just as two-dimensional fine art corresponds to the shape of the canvas on which it is painted, two-dimensional applied art correlates to the contours of the article on which it is applied. A fresco painted on a wall, ceiling panel, or dome would not lose copyright protection, for example, simply because it was designed to track the dimensions of the surface on which it was painted. Or consider, for example, a design etched or painted on the surface of a guitar. If that entire design is imaginatively removed from the guitar’s surface and placed on an album cover, it would still resemble the shape of a guitar. But the image on the cover does not “replicate” the guitar as a useful article. Rather, the design is a two-dimensional work of art that corresponds to the shape of the useful article to which it was applied. The statute protects that work of art whether it is first drawn on the album cover and then applied to the guitar’s surface, or vice versa. Failing to protect that art would create an anomaly: It would extend protection to two-dimensional designs that cover a part of a useful article but would not protect the same design if it covered the entire article. The statute does not support that distinction, nor can it be reconciled with the dissent’s recognition that “artwork printed on a t-shirt” could be protected. Post, at 4 (internal quotation marks omitted). To be clear, the only feature of the cheerleading uniform eligible for a copyright in this case is the two-dimensional work of art fixed in the tangible medium of the uniform fabric. Even if respondents ultimately succeed in establishing a valid copyright in the surface decorations at issue here, respondents have no right to prohibit any person from manufacturing a cheerleading uniform of identical shape, cut, and dimensions to the ones on which the decorations in this case appear. They may prohibit only the reproduction of the surface designs in any tangible medium of expression—a uniform or otherwise.[2] D Petitioner and the Government raise several objections to the approach we announce today. None is meritorious. 1 Petitioner first argues that our reading of the statute is missing an important step. It contends that a feature may exist independently only if it can stand alone as a copyrightable work and if the useful article from which it was extracted would remain equally useful. In other words, copyright extends only to “solely artistic” features of useful articles. Brief for Petitioner 33. According to petitioner, if a feature of a useful article “advance[s] the utility of the article,” id., at 38, then it is categorically beyond the scope of copyright, id., at 33. The designs here are not protected, it argues, because they are necessary to two of the uniforms’ “inherent, essential, or natural functions”—identifying the wearer as a cheerleader and enhancing the wearer’s physical appearance. Id., at 38, 48; Reply Brief 2, 16. Because the uniforms would not be equally useful without the designs, petitioner contends that the designs are inseparable from the “utilitarian aspects” of the uniform. Brief for Petitioner 50. The Government raises a similar argument, although it reaches a different result. It suggests that the appropriate test is whether the useful article with the artistic feature removed would “remai[n] similarly useful.” Brief for United States as Amicus Curiae 29 (emphasis added). In the view of the United States, however, a plain white cheerleading uniform is “similarly useful” to uniforms with respondents’ designs. Id., at 27–28. The debate over the relative utility of a plain white cheerleading uniform is unnecessary. The focus of the separability inquiry is on the extracted feature and not on any aspects of the useful article that remain after the imaginary extraction. The statute does not require the decisionmaker to imagine a fully functioning useful article without the artistic feature. Instead, it requires that the separated feature qualify as a nonuseful pictorial, graphic, or sculptural work on its own. Of course, because the removed feature may not be a useful article—as it would then not qualify as a pictorial, graphic, or sculptural work—there necessarily would be some aspects of the original useful article “left behind” if the feature were conceptually removed. But the statute does not require the imagined remainder to be a fully functioning useful article at all, much less an equally useful one. Indeed, such a requirement would deprive the Mazer statuette of protection had it been created first as a lamp base rather than as a statuette. Without the base, the “lamp” would be just a shade, bulb, and wires. The statute does not require that we imagine a nonartistic replacement for the removed feature to determine whether that feature is capable of an independent existence. Petitioner’s argument follows from its flawed view that the statute protects only “solely artistic” features that have no effect whatsoever on a useful article’s utilitarian function. This view is inconsistent with the statutory text. The statute expressly protects two- and three-dimensional “applied art.” §101. “Applied art” is art “employed in the decoration, design, or execution of useful objects,” Webster’s Third New International Dictionary 105 (1976) (emphasis added), or “those arts or crafts that have a primarily utilitarian function, or . . . the designs and decorations used in these arts,” Random House Dictionary 73 (1966) (emphasis added); see also 1 OED 576 (2d ed. 1989) (defining “applied” as “[p]ut to practical use”). An artistic feature that would be eligible for copyright protection on its own cannot lose that protection simply because it was first created as a feature of the design of a useful article, even if it makes that article more useful. Indeed, this has been the rule since Mazer. In holding that the statuette was protected, the Court emphasized that the 1909 Act abandoned any “distinctions between purely aesthetic articles and useful works of art.” 347 U. S., at 211. Congress did not enact such a distinction in the 1976 Act. Were we to accept petitioner’s argument that the only protectable features are those that play absolutely no role in an article’s function, we would effectively abrogate the rule of Mazer and read “applied art” out of the statute. Because we reject the view that a useful article must remain after the artistic feature has been imaginatively separated from the article, we necessarily abandon the distinction between “physical” and “conceptual” separability, which some courts and commentators have adopted based on the Copyright Act’s legislative history. See H. R. Rep. No. 94–1476, p. 55 (1976). According to this view, a feature is physically separable from the underlying useful article if it can “be physically separated from the article by ordinary means while leaving the utilitarian aspects of the article completely intact.” Compendium §924.2(A); see also Chosun Int’l, Inc. v. Chrisha Creations, Ltd., 413 F. 3d 324, 329 (CA2 2005). Conceptual separability applies if the feature physically could not be removed from the useful article by ordinary means. See Compendium §924.2(B); but see 1 P. Goldstein, Copyright §2.5.3, p. 2:77 (3d ed. 2016) (explaining that the lower courts have been unable to agree on a single conceptual separability test); 2 Patry §§3:140–3:144.40 (surveying the various approaches in the lower courts). The statutory text indicates that separability is a conceptual undertaking. Because separability does not require the underlying useful article to remain, the physical-conceptual distinction is unnecessary. 2 Petitioner next argues that we should incorporate two “objective” components, Reply Brief 9, into our test to provide guidance to the lower courts: (1) “whether the design elements can be identified as reflecting the designer’s artistic judgment exercised independently of functional influence,” Brief for Petitioner 34 (emphasis deleted and internal quotation marks omitted), and (2) whether “there is [a] substantial likelihood that the pictorial, graphic, or sculptural feature would still be marketable to some significant segment of the community without its utilitarian function,” id., at 35 (emphasis deleted and internal quotation marks omitted). We reject this argument because neither consideration is grounded in the text of the statute. The first would require the decisionmaker to consider evidence of the creator’s design methods, purposes, and reasons. Id., at 48. The statute’s text makes clear, however, that our inquiry is limited to how the article and feature are perceived, not how or why they were designed. See Brandir Int’l, Inc. v. Cascade Pacific Lumber Co., 834 F. 2d 1142, 1152 (CA2 1987) (Winter, J., concurring in part and dissenting in part) (The statute “expressly states that the legal test is how the final article is perceived, not how it was developed through various stages”). The same is true of marketability. Nothing in the statute suggests that copyrightability depends on market surveys. Moreover, asking whether some segment of the market would be interested in a given work threatens to prize popular art over other forms, or to substitute judicial aesthetic preferences for the policy choices embodied in the Copyright Act. See Bleistein v. Donaldson Lithographing Co., 188 U. S. 239, 251 (1903) (“It would be a dangerous undertaking for persons trained only to the law to constitute themselves final judges of the worth of pictorial illustrations, outside of the narrowest and most obvious limits”). 3 Finally, petitioner argues that allowing the surface decorations to qualify as a “work of authorship” is inconsistent with Congress’ intent to entirely exclude industrial design from copyright. Petitioner notes that Congress refused to pass a provision that would have provided limited copyright protection for industrial designs, including clothing, when it enacted the 1976 Act, see id., at 9–11 (citing S. 22, Tit. II, 94th Cong., 2d Sess., 122 Cong. Rec. 3856–3859 (1976)), and that it has enacted laws protecting designs for specific useful articles—semiconductor chips and boat hulls, see 17 U. S. C. §§901–914, 1301–1332—while declining to enact other industrial design statutes, Brief for Petitioner 29, 43. From this history of failed legislation petitioner reasons that Congress intends to channel intellectual property claims for industrial design into design patents. It therefore urges us to approach this question with a presumption against copyrightability. Id., at 27. We do not share petitioner’s concern. As an initial matter, “[c]ongressional inaction lacks persuasive significance” in most circumstances. Pension Benefit Guaranty Corporation v. LTV Corp., 496 U. S. 633, 650 (1990) (internal quotation marks omitted). Moreover, we have long held that design patent and copyright are not mutually exclusive. See Mazer, 347 U. S., at 217. Congress has provided for limited copyright protection for certain features of industrial design, and approaching the statute with presumptive hostility toward protection for industrial design would undermine Congress’ choice. In any event, as explained above, our test does not render the shape, cut, and physical dimensions of the cheerleading uniforms eligible for copyright protection. III We hold that an artistic feature of the design of a useful article is eligible for copyright protection if the feature (1) can be perceived as a two- or three-dimensional work of art separate from the useful article and (2) would qualify as a protectable pictorial, graphic, or sculptural work either on its own or in some other medium if imagined separately from the useful article. Because the designs on the surface of respondents’ cheerleading uniforms in this case satisfy these requirements, the judgment of the Court of Appeals is affirmed. It is so ordered. Notes 1 We do not today hold that the surface decorations are copyrightable. We express no opinion on whether these works are sufficiently original to qualify for copyright protection, see Feist Publications, Inc. v. Rural Telephone Service Co., 499 U. S. 340 –359 (1991), or on whether any other prerequisite of a valid copyright has been satisfied. 2 The dissent suggests that our test would lead to the copyrighting of shovels. Post, at 7; Appendix to opinion of Breyer, J., fig. 4, post. But a shovel, like a cheerleading uniform, even if displayed in an art gallery, is “an article having an intrinsic utilitarian function that is not merely to portray the appearance of the article or to convey information.” 17 U. S. C. §101. It therefore cannot be copyrighted. A drawing of a shovel could, of course, be copyrighted. And, if the shovel included any artistic features that could be perceived as art apart from the shovel, and which would qualify as protectable pictorial, graphic, or sculptural works on their own or in another medium, they too could be copyrighted. But a shovel as a shovel cannot.

580.US.2016_15-513

The False Claims Act (FCA) authorizes private parties (known as relators) to seek recovery from persons who make false or fraudulent payment claims to the Federal Government, 31 U. S. C. §§3729–3730, and permits the Attorney General to intervene in a relator’s action or bring an FCA suit in the first instance, §§3730(a)–(b). This system is designed to benefit both the relator and the Government. A relator who initiates a meritorious qui tam suit receives, inter alia, a percentage of the ultimate damages award, §3730(d), while “ ‘encourag[ing] more private enforcement suits’ ” serves “ ‘to strengthen the Government’s hand in fighting false claims,’ ” Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, 559 U. S. 280 . The FCA establishes specific procedures for relators to follow, including the requirement relevant here: “The complaint shall be filed in camera, shall remain under seal for at least 60 days, and shall not be served on the defendant until the court so orders.” §3730(b)(2). In the years before Hurricane Katrina, petitioner State Farm issued, as pertinent here, both Federal Government-backed flood insurance policies and petitioner’s own general homeowner policies. Respondents Cori and Kerri Rigsby, former claims adjusters for one of petitioner’s contractors, E. A. Renfroe & Co., filed a complaint under seal in April 2006, claiming that petitioner instructed them and other adjusters to misclassify wind damage as flood damage in order to shift petitioner’s insurance liability to the Government. The District Court extended the length of the seal several times at the Government’s request, but lifted the seal in part in January 2007, allowing disclosure of the action to another District Court hearing a suit by E. A. Renfroe against respondents. In August 2007, the District Court lifted the seal in full. The Government subsequently declined to intervene. Petitioner moved to dismiss the suit on the grounds that respondents had violated the seal requirement. Specifically, it alleged, respondents’ former attorney had disclosed the complaint’s existence to several news outlets, which issued stories about the fraud allegations, but did not mention the existence of the FCA complaint; and respondents had met with a Congressman who later spoke out against the purported fraud. The District Court applied the test for dismissal set out in United States ex rel. Lujan v. Hughes Aircraft Co., 67 F. 3d 242, 245–247. Balancing three factors—actual harm to the Government, severity of the violations, and evidence of bad faith—the court decided against dismissal. Petitioner did not request a lesser sanction. The Fifth Circuit affirmed. It first concluded that a seal violation does not require mandatory dismissal of a relator’s complaint. It then considered the same factors weighed by the District Court and reached a similar conclusion. Held: 1. A seal violation does not mandate dismissal of a relator’s complaint. Pp. 6–9. (a) The FCA does not enact so harsh a rule. Section 3730(b)(2)’s requirement that a complaint “shall” be kept under seal is a mandatory rule for relators. But the statute says nothing about the remedy for violating that rule; and absent congressional guidance regarding a remedy, “the sanction for breach [of a mandatory duty] is not loss of all later powers to act.” United States v. Montalvo-Murillo, 495 U. S. 711 . The FCA’s structure supports this result. The FCA has a number of provisions requiring, in express terms, the dismissal of a relator’s action. E.g., §§3730(b)(5), (e)(1)–(2). It is thus proper to infer that Congress did not intend to require dismissal for a violation of the seal requirement. See Marx v. General Revenue Corp., 568 U. S. ___, ___. This result is also consistent with the general purpose of §3730(b)(2), which was enacted as part of a set of reforms meant to “encourage more private enforcement suits,” S. Rep. No. 99–345, pp. 23–24, and which was intended to protect the Government’s interests, allaying its concern that a relator filing a civil complaint would alert defendants to a pending federal criminal investigation. It would thus make little sense to adopt a rigid interpretation that prejudices the Government by depriving it of needed assistance from private parties. Pp. 6–7. (b) Petitioner’s arguments to the contrary are unavailing. There is no textual indication that Congress conditioned the authority to file a private right of action on compliance with the seal requirement or that the relator’s ability to bring suit depends on adherence to the seal requirement. And the Senate Committee Report’s recitation of the FCA’s general purpose is best understood to support respondents rather than a mandatory dismissal rule. Moreover, because the FCA’s text and structure are clear, there is no need to accept petitioner’s invitation to consider a few stray sentences from the legislative history. Pp. 8–9. 2. The District Court did not abuse its discretion by denying petitioner’s motion to dismiss. The question whether dismissal is appropriate should be left to the sound discretion of the district court. While the Hughes Aircraft factors appear to be appropriate, it is unnecessary to explore these and other relevant considerations, which can be discussed in the course of later cases. Pp. 9–10. 3. On this record, where petitioner requested no sanction other than dismissal, the question whether a lesser sanction—such as monetary penalties—is warranted is not preserved. P. 10. 794 F. 3d 457, affirmed. Kennedy, J., delivered the opinion for a unanimous Court.

This case addresses the question of the proper remedy when there is a violation of the False Claims Act (FCA) requirement that certain complaints must be sealed for a limited time period. See 31 U. S. C. §3730(b)(2). There are two questions presented before this Court. First, do any and all violations of the seal requirement mandate dismissal of a private party’s complaint with prejudice? Second, if dismissal is not mandatory, did the District Court here abuse its discretion by declining to dismiss respondents’ complaint? I A The FCA imposes civil liability on an individual who, inter alia, “knowingly presents . . . a false or fraudulent claim for payment or approval” to the Federal Government. §3729(a)(1)(A). Almost unique to the FCA are its qui tam enforcement provisions, which allow a private party known as a “relator” to bring an FCA action on behalf of the Government. §3730(b)(1); Vermont Agency of Natural Resources v. United States ex rel. Stevens, 529 U. S. 765 , n. 1 (2000) (listing three other qui tam statutes). The Attorney General retains the authority to intervene in a relator’s ongoing action or to bring an FCA suit in the first instance. §§3730(a)–(b). This system is designed to benefit both the relator and the Government. A relator who initiates a meritorious qui tam suit receives a percentage of the ultimate dam-ages award, plus attorney’s fees and costs. §3730(d). In turn, “ ‘encourag[ing] more private enforcement suits’ ” serves “ ‘to strengthen the Government’s hand in fighting false claims.’ ” Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, 559 U. S. 280, 298 (2010) . The FCA places a number of restrictions on suits by relators. For example, under the provision known as the “first-to-file bar,” a relator may not “ ‘bring a related action based on the facts underlying [a] pending action.’ ” Kellogg Brown & Root Services, Inc. v. United States ex rel. Carter, 575 U. S. ___, ___ (2015) (slip op., at 11) (quoting §3730(b)(5); emphasis deleted). Other FCA provisions require compliance with statutory requirements as express conditions on the relators’ ability to bring suit. The paragraph known as the “public disclosure bar,” for instance, provided at the time this suit was filed that “ ‘[n]o court shall have jurisdiction over an action under this section based upon the public disclosure of allegations or transactions . . . unless the action is brought by the Attorney General or . . . an original source of the information.’ ” Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, supra, at 283, n. 1, 285–286 (quoting 31 U. S. C. §3730(e)(4)(A) (2006 ed.); footnote omitted). The FCA also establishes specific procedures for the relator to follow when filing the complaint. Among other things, the relator must serve on the Government “[a] copy of the complaint and written disclosure of substantially all material evidence and information the [relator] possesses.” §3730(b)(2). Most relevant here, the FCA provides: “The complaint shall be filed in camera, shall remain under seal for at least 60 days, and shall not be served on the defendant until the court so orders.” Ibid. B Petitioner State Farm is an insurance company. In the years before Hurricane Katrina, petitioner issued two types of homeowner-insurance policies that are relevant in this case: (1) Federal Government-backed flood insurance policies and (2) petitioner’s own general homeowner insurance policies. The practical effect for homeowners who were affected by Hurricane Katrina and who purchased both policies was that petitioner would be responsible for paying for wind damage, while the Government would pay for flood damage. As the Court of Appeals noted, this arrangement created a potential conflict of interest: Petitioner had “an incentive to classify hurricane damage as flood-related to limit its economic exposure.” 794 F. 3d 457, 462 (CA5 2015). Respondents Cori and Kerri Rigsby are former claims adjusters for one of petitioner’s contractors, E. A. Renfroe & Co. Together with other adjusters, they were responsible for visiting the damaged homes of petitioner’s customers to determine the extent to which a homeowner was entitled to an insurance payout. According to respondents, petitioner instructed them and other adjusters to misclassify wind damage as flood damage in order to shift petitioner’s insurance liability to the Government. See id., at 463–464 (summarizing trial evidence). In April 2006, respondents filed their qui tam complaint under seal. At the Government’s request, the District Court extended the length of the seal a number of times. In January 2007, the court lifted the seal in part, allowing disclosure of the qui tam action to another District Court hearing a suit by E. A. Renfroe against respondents for purported misappropriation of documents related to petitioner’s alleged fraud. See E. A. Renfroe & Co. v. Moran, No. 2:06–cv–1752 (ND Ala.). In August 2007, the District Court lifted the seal in full. In January 2008, the Government declined to intervene. In January 2011, petitioner moved to dismiss respondents’ suit on the grounds that they had violated the seal requirement. The parties do not dispute the essential background. In the months before the seal was lifted in part, respondents’ then-attorney, one Dickie Scruggs, e-mailed a sealed evidentiary filing that disclosed the complaint’s existence to journalists at ABC, the Associated Press, and the New York Times. All three outlets issued stories discussing the fraud allegations, but none revealed the existence of the FCA complaint. Respondents themselves met with Mississippi Congressman Gene Taylor, who later spoke out in public against petitioner’s purported fraud, although he did not mention the existence of the FCA suit at that time. After the seal was lifted in part, Scruggs disclosed the existence of the suit to various others, including a public relations firm and CBS News. At the time of the motion to dismiss in 2011, respondents were represented neither by Scruggs nor by any of the attorneys who had worked with him. In March 2008, Scruggs withdrew from respondents’ case after he was indicted for attempting to bribe a state-court judge. Two months later, the District Court removed the remaining Scruggs-affiliated attorneys from the case, based on their alleged involvement in improper payments made from Scruggs to respondents. The District Court did not punish respondents themselves for the payments because they were not made “aware of the ethical implications” and, as laypersons, “are not bound by the rules of professional conduct that apply to” attorneys. App. 21. In deciding petitioner’s motion the District Court considered only the seal violations that occurred before the seal was lifted in part, reasoning the partial lifting in effect had mooted the seal. Applying the test for dismissal set out in United States ex rel. Lujan v. Hughes Aircraft Co., 67 F. 3d 242, 245–247 (CA9 1995), the District Court balanced three factors: (1) the actual harm to the Government, (2) the severity of the violations, and (3) the evidence of bad faith. The court decided against dismissal. Petitioner did not request some lesser sanction. The case went to trial, resulting in a victory for respondents on what the Court of Appeals referred to as a “bellwether” claim regarding a single damaged home. 794 F. 3d, at 462. The Court of Appeals for the Fifth Circuit affirmed the denial of petitioner’s motion to dismiss. The court recognized that the case presented two related issues of the first impression under its case law: (1) whether a seal violation requires mandatory dismissal of a relator’s complaint and, if not, (2) what standard governs a district court’s decision to dismiss. The court noted that the Courts of Appeals for the Second and Ninth Circuits had held that the FCA does not require automatic dismissal for a seal violation, while the Court of Appeals for the Sixth Circuit had held that dismissal is mandatory. See United States ex rel. Pilon v. Martin Marietta Corp., 60 F. 3d 995, 998 (CA2 1995); United States ex rel. Lujan v. Hughes Aircraft Co., supra, at 245; United States ex rel. Summers v. LHC Group Inc., 623 F. 3d 287, 296 (CA6 2010); see also United States ex rel. Smith v. Clark/Smoot/Russell, 796 F. 3d 424, 430 (CA4 2015) (following Pilon). After a careful analysis, the Court of Appeals for the Fifth Circuit held automatic dismissal is not required by the FCA. 794 F. 3d, at 470–471. It then considered the same factors the District Court had weighed and came to a similar conclusion. Id., at 471–472. First, the Court of Appeals held the Government was in all likelihood not harmed by the disclosures because none of them led to the publication of the pendency of the suit before the seal was lifted in part. Second, the Court of Appeals determined the violations were not severe in their repercussions because respondents had complied with the seal requirement when they first filed their suit. Third, the Court of Appeals assumed, without deciding, that the bad behavior of respondents’ then-attorney could be imputed to respondents; but it held that, even presuming the attribution of bad faith, the other factors favored respondents. This Court granted certiorari, 578 U. S. ___ (2016), and now affirms. II A Petitioner’s primary contention is that a violation of the seal provision necessarily requires a relator’s complaint to be dismissed. The FCA does not enact so harsh a rule. Section 3730(b)(2)’s text provides that a complaint “shall” be kept under seal. True, this language creates a mandatory rule the relator must follow. See Rockwell Int’l Corp. v. United States, 549 U. S. 457, 464 (2007) (“As required under the Act, [the relator] filed his complaint under seal . . . ”); see also Kingdomware Technologies, Inc. v. United States, 579 U. S. ___, ___ (2016) (slip op., at 9) (“[T]he word ‘shall’ usually connotes a requirement”). The statute says nothing, however, about the remedy for a violation of that rule. In the absence of congressional guidance regarding a remedy, “[a]lthough the duty is mandatory, the sanction for breach is not loss of all later powers to act.” United States v. Montalvo-Murillo, 495 U. S. 711, 718 (1990) . The FCA’s structure is itself an indication that violating the seal requirement does not mandate dismissal. This Court adheres to the general principle that Congress’ use of “explicit language” in one provision “cautions against inferring” the same limitation in another provision. Marx v. General Revenue Corp., 568 U. S. ___, ___ (2013) (slip op., at 12). And the FCA has a number of provisions that do require, in express terms, the dismissal of a re-lator’s action. Supra, at 2 (citing §3730(b)(5)); see also §§3730(e)(1)–(2) (“[n]o court shall have jurisdiction” over certain FCA claims by relators against a member of the military or of the judicial, legislative, or executive branches). It is proper to infer that, had Congress intended to require dismissal for a violation of the seal requirement, it would have said so. The Court’s conclusion is consistent with the general purpose of §3730(b)(2). The seal provision was enacted in the 1980’s as part of a set of reforms that were meant to “encourage more private enforcement suits.” S. Rep. No. 99–345, pp. 23–24 (1986). At the time, “perhaps the most serious problem plaguing effective enforcement” of the FCA was “a lack of resources on the part of Federal enforcement agencies.” Id., at 7. The Senate Committee Report indicates that the seal provision was meant to allay the Government’s concern that a relator filing a civil complaint would alert defendants to a pending federal criminal investigation. Id., at 24. Because the seal requirement was intended in main to protect the Government’s interests, it would make little sense to adopt a rigid interpretation of the seal provision that prejudices the Government by depriving it of needed assistance from private parties. The Federal Government agrees with this interpretation. It informs the Court that petitioner’s test “would undermine the very governmental interests that the seal provision is meant to protect.” Brief for United States as Amicus Curiae 10. B Petitioner’s arguments to the contrary are unavailing. First, petitioner urges that because the seal provision appears in the subsection of the FCA creating the relator’s private right of action, Congress intended to condition the right to bring suit on compliance with the seal requirement. It is true that, as discussed further below, the Court sometimes has concluded that Congress conditioned the authority to file a private right of action on compliance with a statutory mandate. E.g., Hallstrom v. Tillamook County, 493 U. S. 20 –26 (1989). There is no textual indication, however, that Congress did so here. Section 3730(b)(2) does not tie the seal requirement to the right to bring the qui tam suit in conditional terms. As noted above, the statute just provides: “The complaint shall be filed in camera, shall remain under seal for at least 60 days, and shall not be served on the defendant until the court so orders.” The text at issue in Hallstrom, by contrast, was quite different than the statutory language that controls here. The Hallstrom statute, part of the Resource Conservation and Recovery Act of 1976, provided: “ ‘No action may be commenced . . . prior to sixty days after the plaintiff has given notice of the violation’ ” to the Government. 493 U. S., at 25. Petitioner cites two additional cases to support its argument, but those decisions concerned statutes that used even clearer conditional words, like “if” and “unless.” See United States ex rel. Texas Portland Cement Co. v. McCord, 233 U. S. 157, 161 (1914) (statute allowed creditors of Government contractors to bring suit “ ‘if no suit should be brought by the United States within six months from the completion and final settlement of said contract’ ”); McNeil v. United States, 508 U. S. 106 , n. 1 (1993) (statute provided that “ ‘[a]n action shall not be instituted upon a claim against the United States for money damages . . . unless the claimant shall have first presented the claim to the appropriate Federal agency’ ”). Again, the FCA’s structure shows that Congress knew how to draft the kind of statutory language that petitioner seeks to read into §3730(b)(2). The applicable version of the public disclosure bar, for example, requires a district court to dismiss an action when the underlying information has already been made available to the public, “ ‘unless’ ” the plaintiff is the Attorney General or an original source. Graham County Soil and Water Conservation Dist. v. United States ex rel. Wilson, 559 U. S., at 286. Second, petitioner contends that because this Court has described the FCA’s qui tam provisions as “effecting a partial assignment of the Government’s damages claim,” Vermont Agency of Natural Resources v. United States ex rel. Stevens, 529 U. S., at 773, adherence to all of the FCA’s mandatory requirements—no matter how small—is a condition of the assignment. This argument fails for the same reason as the one discussed above: Petitioner can show no textual indication in the statute suggesting that the relator’s ability to bring suit depends on adherence to the seal requirement. Third, petitioner points to a few stray sentences in the Senate Committee Report that it claims support the mandatory dismissal rule. As explained above, however, the Report’s recitation of the general purpose of the statute is best understood to support respondents. Supra, at 7. And, furthermore, because the meaning of the FCA’s text and structure is “plain and unambiguous, we need not accept petitioner[’s] invitation to consider the legislative history.” Whitfield v. United States, 543 U. S. 209, 215 (2005) . III Petitioner’s secondary argument is that the District Court did not consider the proper factors when declining to dismiss respondents’ complaint or, at a minimum, that it was plain error not to consider respondents’ conduct after the seal was lifted in part. This Court holds the District Court did not abuse its discretion by denying petitioner’s motion, much less commit plain error. In light of the questionable conduct of respondents’ prior attorney, it well may not have been reversible error had the District Court granted the motion; that possibility, however, need not be considered here. In general, the question whether dismissal is appropriate should be left to the sound discretion of the district court. While the factors articulated in United States ex rel. Lujan v. Hughes Aircraft Co. appear to be appropriate, it is unnecessary to explore these and other relevant considerations. These standards can be discussed in the course of later cases. IV Petitioner and its amici place great emphasis on the reputational harm FCA defendants may suffer when the seal requirement is violated. But even if every seal violation does not mandate dismissal, that sanction remains a possible form of relief. District courts have inherent power, moreover, to impose sanctions short of dismissal for violations of court orders. See Chambers v. NASCO, Inc., 501 U. S. 32 –46 (1991). Remedial tools like monetary penalties or attorney discipline remain available to punish and deter seal violations even when dismissal is not appropriate. Of note in this case, petitioner did not request any sanction other than dismissal. Tr. of Oral Arg. 3–4, 17. Had petitioner sought some lesser sanctions, the District Court might have taken a different course. Yet petitioner failed to do so. On this record, the question whether a lesser sanction is warranted is not preserved. The judgment of the Court of Appeals for the Fifth Circuit is Affirmed.

581.US.2016_16-341

The patent venue statute, 28 U. S. C. §1400(b), provides that “[a]ny civil action for patent infringement may be brought in the judicial district where the defendant resides, or where the defendant has committed acts of infringement and has a regular and established place of business.” In Fourco Glass Co. v. Transmirra Products Corp., 353 U. S. 222 , this Court concluded that for purposes of §1400(b) a domestic corporation “resides” only in its State of incorporation, rejecting the argument that §1400(b) incorporates the broader definition of corporate “residence” contained in the general venue statute, 28 U. S. C. §1391(c). Congress has not amended §1400(b) since Fourco, but it has twice amended §1391, which now provides that, “[e]xcept as otherwise provided by law” and “[f]or all venue purposes,” a corporation “shall be deemed to reside, if a defendant, in any judicial district in which such defendant is subject to the court’s personal jurisdiction with respect to the civil action in question.” §§1391(a), (c). Respondent filed a patent infringement suit in the District Court for the District of Delaware against petitioner, a competitor that is organized under Indiana law and headquartered in Indiana but ships the allegedly infringing products into Delaware. Petitioner moved to transfer venue to a District Court in Indiana, claiming that venue was improper in Delaware. Citing Fourco, petitioner argued that it did not “resid[e]” in Delaware and had no “regular and established place of business” in Delaware under §1400(b). The District Court rejected these arguments. The Federal Circuit denied a petition for a writ of mandamus, concluding that §1391(c) supplies the definition of “resides” in §1400(b). The Federal Circuit reasoned that because petitioner resided in Delaware under §1391(c), it also resided there under §1400(b). Held: As applied to domestic corporations, “reside[nce]” in §1400(b) refers only to the State of incorporation. The amendments to §1391 did not modify the meaning of §1400(b) as interpreted by Fourco. Pp. 3–10. (a) The venue provision of the Judiciary Act of 1789 covered patent cases as well as other civil suits. Stonite Products Co. v. Melvin Lloyd Co., 315 U. S. 561 . In 1897, Congress enacted a patent specific venue statute. This new statute (§1400(b)’s predecessor) permitted suit in the district of which the defendant was an “inhabitant” or in which the defendant both maintained a “regular and established place of business” and committed an act of infringement. 29Stat. 695. A corporation at that time was understood to “inhabit” only the State of incorporation. This Court addressed the scope of §1400(b)’s predecessor in Stonite, concluding that it constituted “the exclusive provision controlling venue in patent infringement proceedings” and thus was not supplemented or modified by the general venue provisions. 315 U. S., at 563. In 1948, Congress recodified the patent venue statute as §1400(b). That provision, which remains unaltered today, uses “resides” instead of “inhabit[s].” At the same time, Congress also enacted the general venue statute, §1391, which defined “residence” for corporate defendants. In Fourco, this Court reaffirmed Stonite’s holding, observing that Congress enacted §1400(b) as a standalone venue statute and that nothing in the 1948 recodification evidenced an intent to alter that status, even the fact that §1391(c) by “its terms” embraced “all actions,” 353 U. S., at 228. The Court also concluded that “resides” in the recodified version bore the same meaning as “inhabit[s]” in the pre-1948 version. See id., at 226. This landscape remained effectively unchanged until 1988, when Congress amended the general venue statute, §1391(c). The revised provision stated that it applied “[f]or purposes of venue under this chapter.” In VE Holding Corp. v. Johnson Gas Appliance Co., 917 F. 2d 1574, 1578, the Federal Circuit held that, in light of this amendment, §1391(c) established the definition for all other venue statutes under the same “chapter,” including §1400(b). In 2011, Congress adopted the current version of §1391, which provides that its general definition applies “[f]or all venue purposes.” The Federal Circuit reaffirmed VE Holding in the case below. Pp. 3–7. (b) In Fourco, this Court definitively and unambiguously held that the word “reside[nce]” in §1400(b), as applied to domestic corporations, refers only to the State of incorporation. Because Congress has not amended §1400(b) since Fourco, and neither party asks the Court to reconsider that decision, the only question here is whether Congress changed §1400(b)’s meaning when it amended §1391. When Congress intends to effect a change of that kind, it ordinarily provides a relatively clear indication of its intent in the amended provision’s text. No such indication appears in the current version of §1391. Respondent points out that the current §1391(c) provides a default rule that, on its face, applies without exception “[f]or all venue purposes.” But the version at issue in Fourco similarly provided a default rule that applied “ ‘for venue purposes,’ ” 353 U. S., at 223, and those phrasings are not materially different in this context. The addition of the word “all” to the already comprehensive provision does not suggest that Congress intended the Court to reconsider its decision in Fourco. Any argument based on this language is even weaker now than it was when the Court rejected it in Fourco. Fourco held that §1400(b) retained a meaning distinct from the default definition contained in §1391(c), even though the latter, by its terms, included no exceptions. The current version of §1391 includes a saving clause, which expressly states that the provision does not apply when “otherwise provided by law,” thus making explicit the qualification that the Fourco Court found implicit in the statute. Finally, there is no indication that Congress in 2011 ratified the Federal Circuit’s decision in VE Holding. Pp. 7–10. 821 F. 3d 1338, reversed and remanded. Thomas, J., delivered the opinion of the Court, in which all other Members joined, except Gorsuch, J., who took no part in the consideration or decision of the case.

The question presented in this case is where proper venue lies for a patent infringement lawsuit brought against a domestic corporation. The patent venue statute, 28 U. S. C. §1400(b), provides that “[a]ny civil action for patent infringement may be brought in the judicial district where the defendant resides, or where the defendant has committed acts of infringement and has a regular and established place of business.” In Fourco Glass Co. v. Transmirra Products Corp., 353 U. S. 222, 226 (1957) , this Court concluded that for purposes of §1400(b) a domestic corporation “resides” only in its State of incorporation. In reaching that conclusion, the Court rejected the argument that §1400(b) incorporates the broader definition of corporate “residence” contained in the general venue statute, 28 U. S. C. §1391(c). 353 U. S., at 228. Congress has not amended §1400(b) since this Court construed it in Fourco, but it has amended §1391 twice. Section 1391 now provides that, “[e]xcept as otherwise provided by law” and “[f]or all venue purposes,” a corporation “shall be deemed to reside, if a defendant, in any judicial district in which such defendant is subject to the court’s personal jurisdiction with respect to the civil action in question.” §§1391(a), (c). The issue in this case is whether that definition supplants the definition announced in Fourco and allows a plaintiff to bring a patent infringement lawsuit against a corporation in any district in which the corporation is subject to personal jurisdiction. We conclude that the amendments to §1391 did not modify the meaning of §1400(b) as interpreted by Fourco. We therefore hold that a domestic corporation “resides” only in its State of incorporation for purposes of the patent venue statute. I Petitioner, which is organized under Indiana law and headquartered in Indiana, manufactures flavored drink mixes.[1] Respondent, which is organized under Delaware law and has its principal place of business in Illinois, is a competitor in the same market. As relevant here, respondent sued petitioner in the District Court for the District of Delaware, alleging that petitioner’s products infringed one of respondent’s patents. Although petitioner is not registered to conduct business in Delaware and has no meaningful local presence there, it does ship the al-legedly infringing products into the State. Petitioner moved to dismiss the case or transfer venue to the District Court for the Southern District of Indiana, arguing that venue was improper in Delaware. See 28 U. S. C. §1406. Citing Fourco’s holding that a corporation resides only in its State of incorporation for patent infringement suits, petitioner argued that it did not “resid[e]” in Delaware under the first clause of §1400(b). It further argued that it had no “regular and established place of business” in Delaware under the second clause of §1400(b). Relying on Circuit precedent, the District Court rejected these arguments, 2015 WL 5613160 (D Del., Sept. 24, 2015), and the Federal Circuit denied a petition for a writ of mandamus, In re TC Heartland LLC, 821 F. 3d 1338 (2016). The Federal Circuit concluded that subsequent statutory amendments had effectively amended §1400(b) as construed in Fourco, with the result that §1391(c) now supplies the definition of “resides” in §1400(b). 821 F. 3d, at 1341–1343. Under this logic, because the District of Delaware could exercise personal jurisdiction over petitioner, petitioner resided in Delaware under §1391(c) and, therefore, under §1400(b). We granted certiorari, 580 U. S. ___ (2016), and now reverse. II A The history of the relevant statutes provides important context for the issue in this case. The Judiciary Act of 1789 permitted a plaintiff to file suit in a federal district court if the defendant was “an inhabitant” of that district or could be “found” for service of process in that district. Act of Sept. 24, 1789, §11, 1Stat. 79. The Act covered patent cases as well as other civil suits. Stonite Products Co. v. Melvin Lloyd Co., 315 U. S. 561, 563 (1942) . In 1887, Congress amended the statute to permit suit only in the district of which the defendant was an inhabitant or, in diversity cases, of which either the plaintiff or defendant was an inhabitant. See Act of Mar. 3, 1887, §1, 24 Stat. 552; see also Stonite, supra, at 563–564. This Court’s decision in In re Hohorst, 150 U. S. 653 –662 (1893), arguably suggested that the 1887 Act did not apply to patent cases. As a result, while some courts continued to apply the Act to patent cases, others refused to do so and instead permitted plaintiffs to bring suit (in line with the pre-1887 regime) anywhere a defendant could be found for service of process. See Stonite, supra, at 564–565. In 1897, Congress resolved the confusion by enacting a patent specific venue statute. See Act of Mar. 3, 1897, ch. 395, 29Stat. 695. In so doing, it “placed pat-ent infringement cases in a class by themselves, outside the scope of general venue legislation.” Brunette Machine Works, Ltd. v. Kockum Industries, Inc., 406 U. S. 706, 713 (1972) . This new statute (§1400(b)’s predecessor) permitted suit in the district of which the defendant was an “inhabitant,” or a district in which the defendant both maintained a “regular and established place of business” and committed an act of infringement. 29Stat. 695. At the time, a corporation was understood to “inhabit” only the State in which it was incorporated. Shaw v. Quincy Mining Co., 145 U. S. 444 –450 (1892). The Court addressed the scope of §1400(b)’s predecessor in Stonite. In that case, the two defendants inhabited different districts within a single State. The plaintiff sought to sue them both in the same district, invoking a then governing general venue statute that, if applicable, permitted it to do so. 315 U. S., at 562–563. This Court rejected the plaintiff’s venue choice on the ground that the patent venue statute constituted “the exclusive provision controlling venue in patent infringement proceedings” and thus was not supplemented or modified by the general venue provisions. Id., at 563. In the Court’s view, the patent venue statute “was adopted to define the exact jurisdiction of the federal courts in actions to enforce patent rights,” a purpose that would be undermined by interpreting it “to dovetail with the general provisions relating to the venue of civil suits.” Id., at 565–566. The Court thus held that the patent venue statute “alone should control venue in patent infringement proceedings.” Id., at 566. In 1948, Congress recodified the patent venue statute as §1400(b). See Act of June 25, 1948, 62Stat. 936. The recodified provision, which remains unaltered today, states that “[a]ny civil action for patent infringement may be brought in the judicial district where the defendant resides, or where the defendant has committed acts of infringement and has a regular and established place of business.” 28 U. S. C. §1400(b) (1952 ed.). This version differs from the previous one in that it uses “resides” instead of “inhabit[s].” At the same time, Congress also enacted the general venue statute, §1391, which defined “residence” for corporate defendants. That provision stated that “[a] corporation may be sued in any judicial district in which it is incorporated or licensed to do business or is doing business, and such judicial district shall be regarded as the residence of such corporation for venue purposes.” §1391(c) (1952 ed.). Following the 1948 legislation, courts reached differing conclusions regarding whether §1400(b)’s use of the word “resides” incorporated §1391(c)’s definition of “residence.” See Fourco, 353 U. S., at 224, n. 3 (listing cases). In Fourco, this Court reviewed a decision of the Second Circuitholding that §1391(c) defined residence for purposes of §1400(b), “just as that definition is properly . . . incorporated into other sections of the venue chapter.” Trans-mirra Prods. Corp. v. Fourco Glass Co., 233 F. 2d 885, 886 (1956). This Court squarely rejected that interpretation, reaffirming Stonite’s holding that §1400(b) “is the sole and exclusive provision controlling venue in patent infringement actions, and . . . is not to be supplemented by . . . §1391(c).” 353 U. S., at 229. The Court observed that Congress enacted §1400(b) as a standalone venue statute and that nothing in the 1948 recodification evidenced an intent to alter that status. The fact that §1391(c) by “its terms” embraced “all actions” was not enough to overcome the fundamental point that Congress designed §1400(b) to be “complete, independent and alone controlling in its sphere.” Id., at 228. The Court also concluded that “resides” in the recodified version of §1400(b) bore the same meaning as “inhabit[s]” in the pre-1948 version. See id., at 226 (“[T]he [w]ords ‘inhabitant’ and ‘resident,’ as respects venue, are synonymous” (internal quotation marks omitted)). The substitution of “resides” for “inhabit[s]” thus did not suggest any alteration in the venue rules for corporations in patent cases. Accordingly, §1400(b) continued to apply to domestic corporations in the same way it always had: They were subject to venue only in their States of incorporation. See ibid. (The use of “resides” “negat[es] any intention to make corporations suable, in patent infringement cases, where they are merely ‘doing business,’ because those synonymous words [“inhabitant” and “resident”] mean domicile and, in respect of corporations, mean the state of incorporation only”). B This landscape remained effectively unchanged until 1988, when Congress amended the general venue statute, §1391(c), to provide that “[f]or purposes of venue under this chapter, a defendant that is a corporation shall be deemed to reside in any judicial district in which it is subject to personal jurisdiction at the time the action is commenced.” Judicial Improvements and Access to Justice Act, §1013(a), 102Stat. 4669. The Federal Circuit in VE Holding Corp. v. Johnson Gas Appliance Co., 917 F. 2d 1574 (1990), announced its view of the effect of this amendment on the meaning of the patent venue statute. The court reasoned that the phrase “[f]or purposes of venue under this chapter” was “exact and classic language of incorporation,” id., at 1579, and that §1391(c) accord-ingly established the definition for all other venue statutes under the same “chapter.” Id., at 1580. Because §1400(b) fell within the relevant chapter, the Federal Circuit concluded that §1391(c), “on its face,” “clearly applies to §1400(b), and thus redefines the meaning of the term ‘resides’ in that section.” Id., at 1578. Following VE Holding, no new developments occurred until Congress adopted the current version of §1391 in 2011 (again leaving §1400(b) unaltered). See Federal Courts Jurisdiction and Venue Clarification Act of 2011, §202, 125Stat. 763. Section 1391(a) now provides that, “[e]xcept as otherwise provided by law,” “this section shall govern the venue of all civil actions brought in district courts of the United States.” And §1391(c)(2), in turn, provides that, “[f]or all venue purposes,” certain entities, “whether or not incorporated, shall be deemed to reside, if a defendant, in any judicial district in which such defendant is subject to the court’s personal jurisdiction with respect to the civil action in question.” In its decision below, the Federal Circuit reaffirmed VE Holding, reasoning that the 2011 amendments provided no basis to reconsider its prior decision. III We reverse the Federal Circuit. In Fourco, this Court definitively and unambiguously held that the word “reside[nce]” in §1400(b) has a particular meaning as applied to domestic[2] corporations: It refers only to the State of incorporation. Congress has not amended §1400(b) since Fourco, and neither party asks us to reconsider our holding in that case. Accordingly, the only question we must answer is whether Congress changed the meaning of §1400(b) when it amended §1391. When Congress intends to effect a change of that kind, it ordinarily provides a relatively clear indication of its intent in the text of the amended provision. See United States v. Madigan, 300 U. S. 500, 506 (1937) (“[T]he modification by implication of the settled construction of an earlier and different section is not favored”); A. Scalia & B. Garner, Reading Law 331 (2012) (“A clear, authoritative judicial holding on the meaning of a particular provision should not be cast in doubt and subjected to challenge whenever a related though not utterly inconsistent provision is adopted in the same statute or even in an affiliated statute”). The current version of §1391 does not contain any indication that Congress intended to alter the meaning of §1400(b) as interpreted in Fourco. Although the current version of §1391(c) provides a default rule that applies “[f]or all venue purposes,” the version at issue in Fourco similarly provided a default rule that applied “for venue purposes.” 353 U. S., at 223 (internal quotation marks omitted). In this context, we do not see any material difference between the two phrasings. See Pure Oil Co. v. Suarez, 384 U. S. 202 –205 (1966) (construing “ ‘for venue purposes’ ” to cover “all venue statutes”). Respondent argues that “ ‘all venue purposes’ means ‘all venue purposes’—not ‘all venue purposes except for patentvenue.’ ” Brief for Respondent 21. The plaintiffs in Fourco advanced the same argument. See 353 U. S., at 228 (“The main thrust of respondents’ argument is that §1391(c) is clear and unambiguous and that its terms include all actions—including patent infringement actions”). This Court was not persuaded then, and the addition of the word “all” to the already comprehensive provision does not suggest that Congress intended for us to reconsider that conclusion. This particular argument is even weaker under the current version of §1391 than it was under the provision in place at the time of Fourco, because the current provision includes a saving clause expressly stating that it does not apply when “otherwise provided by law.” On its face, the version of §1391(c) at issue in Fourco included no exceptions, yet this Court still held that “resides” in §1400(b) retained its original meaning contrary to §1391(c)’s default definition. Fourco’s holding rests on even firmer footing now that §1391’s saving clause expressly contemplates that certain venue statutes may retain definitions of “resides” that conflict with its default definition. In short, the saving clause makes explicit the qualification that this Court previously found implicit in the statute. See Pure Oil, supra, at 205 (interpreting earlier version of §1391 to apply “to all venue statutes using residence as a criterion, at least in the absence of contrary restrictive indications in any such statute”). Respondent suggests that the saving clause in §1391(a) does not apply to the definitional provisions in §1391(c), Brief for Respondent 31–32, but that interpretation is belied by the text of §1391(a), which makes clear that the saving clause applies to the entire “section.” See §1391(a)(1) (“Except as otherwise provided by law— . . . this section shall govern the venue of all civil actions” (emphasis added)). Finally, there is no indication that Congress in 2011 ratified the Federal Circuit’s decision in VE Holding. If anything, the 2011 amendments undermine that decision’s rationale. As petitioner points out, VE Holding relied heavily—indeed, almost exclusively—on Congress’ decision in 1988 to replace “for venue purposes” with “[f]or purposes of venue under this chapter” (emphasis added) in §1391(c). Congress deleted “under this chapter” in 2011 and worded the current version of §1391(c) almost identically to the original version of the statute. Compare §1391(c) (2012 ed.) (“[f]or all venue purposes”) with §1391(c) (1952 ed.) (“for venue purposes”). In short, nothing in the text suggests congressional approval of VE Holding. * * * As applied to domestic corporations, “reside[nce]” in §1400(b) refers only to the State of incorporation. Accordingly, we reverse the judgment of the Court of Appeals and remand the case for further proceedings consistent with this opinion. It is so ordered. Justice Gorsuch took no part in the consideration or decision of this case.Notes 1 The complaint alleged that petitioner is a corporation, and petitioner admitted this allegation in its answer. See App. 11a, 60a. Similarly, the petition for certiorari sought review on the question of “corporate” residence. See Pet. for Cert. i. In their briefs before this Court, how-ever, the parties suggest that petitioner is, in fact, an unincorporated entity. See Brief for Respondent 9, n. 4 (the complaint’s allegation was “apparently inaccurat[e]”); Reply Brief 4. Because this case comes to us at the pleading stage and has been litigated on the understanding that petitioner is a corporation, we confine our analysis to the proper venue for corporations. We leave further consideration of the issue of petitioner’s legal status to the courts below on remand. 2 The parties dispute the implications of petitioner’s argument for foreign corporations. We do not here address that question, nor do we express any opinion on this Court’s holding in Brunette Machine Works, Ltd. v. Kockum Industries, Inc., 406 U. S. 706 (1972) (determining proper venue for foreign corporation under then existing statutory regime).

581.US.2016_16-605

Land developer Steven Sherman paid $2.7 million to purchase land in the town of Chester (Town) for a housing subdivision. He also sought the Town’s approval of his development plan. About a decade later, he filed this suit in New York state court, claiming that the Town had obstructed his plans for the subdivision, forcing him to spend around $5.5 million to comply with its demands and driving him to the brink of personal bankruptcy. Sherman asserted, among other claims, a regulatory takings claim under the Fifth and Fourteenth Amendments. The Town removed the case to a Federal District Court, which dismissed the takings claim as unripe. The Second Circuit reversed that determination and remanded for the case to go forward. On remand, real estate development company Laroe Estates, Inc. (respondent here), filed a motion to intervene of right under Federal Rule of Civil Procedure 24(a)(2), which requires a court to permit intervention by a litigant that “claims an interest related to the property or transaction that is the subject of the action, and is so situated that disposing of the action may as a practical matter impair or impede the movant’s ability to protect its interest, unless existing parties adequately represent that interest.” Laroe alleged that it had paid Sherman more than $2.5 million in relation to the development project and the subject property, that its resulting equitable interest in the property would be impaired if it could not intervene, and that Sherman would not adequately represent its interest. Laroe filed, inter alia, an intervenor’s complaint asserting a regulatory takings claim that was substantively identical to Sherman’s and seeking a judgment awarding Laroe compensation for the taking of Laroe’s interest in the property at issue. The District Court denied Laroe’s motion to intervene, concluding that its equitable interest did not confer standing. The Second Circuit reversed, holding that an intervenor of right is not required to meet Article III’s standing requirements. Held: 1. A litigant seeking to intervene as of right under Rule 24(a)(2) must meet the requirements of Article III standing if the intervenor wishes to pursue relief not requested by a plaintiff. To establish Article III standing, a plaintiff seeking compensatory relief must have “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision.” Spokeo, Inc. v. Robins, 578 U. S. ___, ___. The “ plaintiff must demonstrate standing for each claim he seeks to press and for each form of relief that is sought.” Davis v. Federal Election Comm’n, 554 U. S. 724 (internal quotation marks omitted). The same principle applies when there are multiple plaintiffs: At least one plaintiff must have standing to seek each form of relief requested in the complaint. That principle also applies to intervenors of right: For all relief sought, there must be a litigant with standing, whether that litigant joins the lawsuit as a plaintiff, a coplaintiff, or an intervenor of right. Thus, at the least, an intervenor of right must demonstrate Article III standing when it seeks additional relief beyond that requested by the plaintiff. That includes cases in which both the plaintiff and the intervenor seek separate money judgments in their own names. Pp. 4–6. 2. The Court of Appeals is to address on remand the question whether Laroe seeks different relief than Sherman. If Laroe wants only a money judgment of its own running directly against the Town, then it seeks damages different from those sought by Sherman and must establish its own Article III standing in order to intervene. The record is unclear on that point, and the Court of Appeals did not resolve that ambiguity. Pp. 6–8. 828 F. 3d 60, vacated and remanded. Alito, J., delivered the opinion for a unanimous Court.

Must a litigant possess Article III standing in order to intervene of right under Federal Rule of Civil Procedure 24(a)(2)? The parties do not dispute—and we hold—that such an intervenor must meet the requirements of Article III if the intervenor wishes to pursue relief not requested by a plaintiff. In the present case, it is unclear whether the intervenor seeks different relief, and the Court of Appeals did not resolve this threshold issue. Accordingly, we vacate the judgment and remand for that court to determine whether the intervenor seeks such additional relief. I In 2001, land developer Steven Sherman paid $2.7 million to purchase nearly 400 acres of land in the town of Chester, New York (Town). Sherman planned to build a housing subdivision called MareBrook, complete with 385 housing units, a golf course, an onsite restaurant, and other amenities. Sherman applied for approval of his plan and thus began a “journey through the Town’s ever-changing labyrinth of red tape.” Sherman v. Chester, 752 F. 3d 554, 557 (CA2 2014). In 2012, Sherman filed this suit against the Town in New York state court. The suit concerned “the decade’s worth of red tape put in place” by the Town and its regulatory bodies. Id., at 558. According to Sherman, the Town obstructed his plans for the subdivision and forced him to spend around $5.5 million to comply with the Town’s demands. Id., at 558, 560. All of this, Sherman claimed, left him financially exhausted and on the brink of personal bankruptcy. Id., at 560. Sherman brought nine federal- and state-law claims against the Town, including a regulatory takings claim under the Fifth and Fourteenth Amendments. See App. 98–122. The Town removed the case to a Federal District Court, which dismissed Sherman’s takings claim as unripe. Opinion and Order in No. 1:12–cv–00647 (SDNY), Dkt. 14, p. 25. The Court of Appeals for the Second Circuit reversed the ripeness determination and remanded for the case to go forward. Chester, supra, at 557.[1] On remand, real estate development company Laroe Estates, Inc. (the respondent here) filed a motion to intervene of right under Federal Rule of Civil Procedure 24(a)(2). This Rule requires a court to permit intervention by a litigant that “claims an interest related to the property or transaction that is the subject of the action, and is so situated that disposing of the action may as a practical matter impair or impede the movant’s ability to protect its interest, unless existing parties adequately represent that interest.” Laroe alleged that in 2003 it had entered into an agreement with Sherman regarding the MareBrook property. Under this agreement, Laroe was to make $6 million in payments to Sherman, secured by a mortgage on all of the development, and Sherman was to sell Laroe parcels of land within the proposed subdivision when the MareBrook plan was approved. However, Laroe reserved the right to terminate the entire agreement if Sherman was unable to obtain Town approval for a sufficient number of lots. While this agreement was in place and Sherman continued his futile quest for regulatory approval, Laroe paid Sherman more than $2.5 million. In 2013, TD Bank commenced a foreclosure proceeding on Sherman’s property. In an effort to save the deal, Laroe and Sherman entered into a new agreement. That agreement provided that the purchase price of the property would be the $2.5 million that Laroe had already advanced Sherman plus any amount Sherman had to pay to settle with TD Bank. Once the Town approved the plan, Laroe was required to transfer a certain number of lots back to Sherman. In addition to imposing this transfer obligation, the agreement deemed Laroe to have paid for the land in full. Laroe was also given the authority to settle the debt Sherman owed TD Bank and to terminate the agreement if the settlement failed. The settlement did fail, and TD Bank took over the property. But Laroe never terminated its agreement with Sherman. In support of its motion to intervene, Laroe argued that, under New York law, it is “the equitable owner of the Real Property” at issue in Sherman’s suit. App. 131, 135–139. Laroe asserted that its status as equitable owner gave it an interest in the MareBrook property; that its interest would be impaired if it could not intervene; and that Sherman “ha[d] his own agenda” and consequently could not adequately represent Laroe’s interest. Id., at 143–145. Along with its other intervention-related pleadings, Laroe filed an intervenor’s complaint asserting a regulatory takings claim that was substantively identical to Sherman’s. Laroe’s complaint sought, among other things, a “judgment against [the Town] awarding [Laroe] damages,” namely, “compensation for the taking of Laroe’s interest in the subject real property.” Id., at 162. The District Court denied Laroe’s motion to intervene on the ground that Laroe lacked standing to bring a takings claim “based on its status as contract vendee to the property.” App. to Pet. for Cert. 57a. The District Court interpreted Second Circuit precedent—specifically, United States Olympic Comm. v. Intelicense Corp., S. A., 737 F. 2d 263, 268 (1984)—to mean that Laroe’s equitable interest did not confer standing. App. to Pet. for Cert. 55a–56a.[2] The Court of Appeals reversed. 828 F. 3d 60, 62 (CA2 2016). Acknowledging a division among the Courts of Appeals on whether an intervenor of right must meet the requirements of Article III, the Second Circuit sided with the courts that have held that Article III standing is not required. Id., at 64–65. We granted certiorari. 580 U. S. ___ (2017). II Article III of the Constitution limits the exercise of the judicial power to “Cases” and “Controversies.” §2, cl. 1. This fundamental limitation preserves the “tripartite structure” of our Federal Government, prevents the Federal Judiciary from “intrud[ing] upon the powers given to the other branches,” and “confines the federal courts to a properly judicial role.” Spokeo, Inc. v. Robins, 578 U. S. ___, ___ (2016) (slip op., at 5–6). “If a dispute is not a proper case or controversy, the courts have no business deciding it, or expounding the law in the course of doing so.” DaimlerChrysler Corp. v. Cuno, 547 U. S. 332, 341 (2006) . “Standing to sue is a doctrine rooted in the traditional understanding of a case or controversy.” Spokeo, supra, at ___ (slip op., at 6). “The law of Article III standing, which is built on separation-of-powers principles, serves to prevent the judicial process from being used to usurp the powers of the political branches.” Clapper v. Amnesty Int’l USA, 568 U. S. 398, 408 (2013) . Our standing doctrine accomplishes this by requiring plaintiffs to “alleg[e] such a personal stake in the outcome of the controversy as to . . . justify [the] exercise of the court’s remedial powers on [their] behalf.” Simon v. Eastern Ky. Welfare Rights Organization, 426 U. S. 26, 38 (1976) (internal quotation marks omitted). To establish Article III standing, the plaintiff seeking compensatory relief must have “(1) suffered an injury in fact, (2) that is fairly traceable to the challenged conduct of the defendant, and (3) that is likely to be redressed by a favorable judicial decision.” Spokeo, supra, at ___ (slip op., at 6). “Absent such a showing, exercise of its power by a federal court would be gratuitous and thus inconsistent with the Art. III limitation.” Simon, supra, at 38. Our standing decisions make clear that “ ‘standing is not dispensed in gross.’ ” Davis v. Federal Election Comm’n, 554 U. S. 724, 734 (2008) (quoting Lewis v. Casey, 518 U. S. 343 , n. 6 (1996); alteration omitted). To the contrary, “a plaintiff must demonstrate standing for each claim he seeks to press and for each form of relief that is sought.” Davis, supra, at 734 (internal quotation marks omitted); see, e.g., DaimlerChrysler, supra, at 352 (“[A] plaintiff must demonstrate standing separately for each form of relief sought”); Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U. S. 167, 185 (2000) (same); Los Angeles v. Lyons, 461 U. S. 95 –106, and n. 7 (1983) (a plaintiff who has standing to seek damages must also demonstrate standing to pursue injunctive relief). The same principle applies when there are multiple plaintiffs. At least one plaintiff must have standing to seek each form of relief requested in the complaint. Both of the parties accept this simple rule.[3] The same principle applies to intervenors of right. Although the context is different, the rule is the same: For all relief sought, there must be a litigant with standing, whether that litigant joins the lawsuit as a plaintiff, a coplaintiff, or an intervenor of right. Thus, at the least, an intervenor of right must demonstrate Article III standing when it seeks additional relief beyond that which the plaintiff requests. This result follows ineluctably from our Article III case law, so it is not surprising that both parties accept it (as does the United States as amicus curiae). See Brief for Petitioner 13 (arguing that an intervenor must always demonstrate standing); Brief for Respondent 28 (“[A]n intervenor who . . . seeks relief beyond that requested by a party with standing must satisfy Arti-cle III”); Brief for United States as Amicus Curiae 16 (An intervenor must demonstrate its own standing if it “seek[s] damages” or “injunctive relief that is broader than or different from the relief sought by the original plaintiff(s)”). In sum, an intervenor of right must have Article III standing in order to pursue relief that is different from that which is sought by a party with standing. That includes cases in which both the plaintiff and the intervenor seek separate money judgments in their own names. Cf. General Building Contractors Assn., Inc. v. Pennsylvania, 458 U. S. 375, 402, n. 22 (1982) (declining to address the State’s standing “until [it] obtains relief different from that sought by plaintiffs whose standing has not been questioned”). That principle dictates the disposition of this case. It is unclear whether Laroe seeks the same relief as Sherman or instead seeks different relief, such as a money judgment against the Town in Laroe’s own name. Laroe’s complaint—the best evidence of the relief Laroe seeks—requests a judgment awarding damages to Laroe. App. 162. Unsurprisingly, Sherman requests something different: specifically, compensation for the taking of his interest in the property. Id., at 122. In other words, as Laroe’s counsel conceded at oral argument, the complaint plainly seeks separate monetary relief for Laroe directly against the Town. Tr. of Oral Arg. 43–44. And, as Laroe’s counsel conceded further, if Laroe is “seeking additional damages in [its] own name,” “at that point, an Article III inquiry would be required.” Id., at 47. To be sure, at some points during argument in the Court of Appeals, Laroe made statements that arguably indicated that Laroe is not seeking damages different from those sought by Sherman. In particular, Laroe’s counsel stated that he was “not saying that Sherman and [Laroe’s] damages are not the same damages,” and insisted that there is “exactly one fund, and the town doesn’t have to do anything except turn over the fund.” Tr. 16, 33; see also Reply Brief in No. 15–1086 (CA2), p. 12 (similar). At other points, however, the same counsel made statements pointing in the opposite direction. When asked directly whether “there would be separate awards to you and to the Sherman estate” if Sherman’s suit was successful, Laroe’s counsel admitted that he “ha[d] never contemplated how [damages] ge[t] allocated at the end of the day” and suggested bifurcated proceedings so that once liability was settled, Laroe and Sherman could “duke it out” over damages if necessary. Tr. 32–35. And in its Court of Appeals briefing, Laroe argued that it—not Sherman—would be entitled to most of the damages from the takings claim, flagging the allocation issue as one that the District Court would have to resolve. Brief for Appellant in No. 15–1086 (CA2), p. 32 (“[T]he trier of fact will have to determine the relative allocation of rights over the fund . . . . Specifi-cally, what is the value of Sherman’s bare legal title as com-pared to Laroe’s equitable title in the subject property”); Reply Brief in No. 15–1086, at 15 (“[M]ost, if not all of the benefits” of this litigation “will accrue [to] Laroe”); see also 828 F. 3d, at 70 (noting that Sherman and Laroe “may disagree about . . . the issue of damages were they to prevail”). Taken together, these representations at best leave it ambiguous whether Laroe is seeking damages for itself or is simply seeking the same damages sought by Sherman.[4] Unfortunately, the Court of Appeals did not resolve this ambiguity. In fact, the section of its opinion concerning standing did not discuss whether Laroe sought different relief than Sherman. Id., at 64–66. Elsewhere, in a different context, the court did acknowledge Laroe’s statement that it sought “essentially the same” damages as Sherman. Id., at 66. But the court also found that “it is unclear from the record whether Laroe believes the Town is directly liable to Sherman or Laroe for the taking.” Ibid. This confusion needs to be dispelled. If Laroe wants only a money judgment of its own running directly against the Town, then it seeks damages different from those sought by Sherman and must establish its own Article III standing in order to intervene. We leave it to the Court of Appeals to address this question on remand. * * * For these reasons, the judgment of the Court of Appeals is vacated, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 Sherman died in 2013, and his estate replaced him as the plaintiff. 2 We assume for the sake of argument only that Laroe does not have Article III standing. If resolution of this question becomes necessary on remand, the Court of Appeals will be required to determine whether the District Court’s decision was correct. 3 See Brief for Petitioner 23 (“If different parties raising a single issue seek different relief, then standing must be shown for each one”); Brief for Respondent 15 (“[A] case or controversy as to one claim does not extend the judicial power to different claims or forms of relief ”). 4 Before this Court, Laroe’s counsel represented that Laroe is not seeking damages of its own and is seeking only to maximize Sherman’s recovery. Tr. of Oral Arg. 43–44. But in light of the ambiguous record and the lack of a reasoned conclusion on this question from the Court of Appeals, we are not inclined to resolve it in the first instance. Cutter v. Wilkinson, 544 U. S. 709 , n. 7 (2005) (“[W]e are a court of review, not first view”).

582.US.2016_15-577

The Trinity Lutheran Church Child Learning Center is a Missouri preschool and daycare center. Originally established as a nonprofit organization, the Center later merged with Trinity Lutheran Church and now operates under its auspices on church property. Among the facilities at the Center is a playground, which has a coarse pea gravel surface beneath much of the play equipment. In 2012, the Center sought to replace a large portion of the pea gravel with a pour-in-place rubber surface by participating in Missouri’s Scrap Tire Program. The program, run by the State’s Department of Natural Resources, offers reimbursement grants to qualifying nonprofit organizations that install playground surfaces made from recycled tires. The Department had a strict and express policy of denying grants to any applicant owned or controlled by a church, sect, or other religious entity. Pursuant to that policy, the Department denied the Center’s application. In a letter rejecting that application, the Department explained that under Article I, Section 7 of the Missouri Constitution, the Department could not provide financial assistance directly to a church. The Department ultimately awarded 14 grants as part of the 2012 program. Although the Center ranked fifth out of the 44 applicants, it did not receive a grant because it is a church. Trinity Lutheran sued in Federal District Court, alleging that the Department’s failure to approve its application violated the Free Exercise Clause of the First Amendment. The District Court dismissed the suit. The Free Exercise Clause, the court stated, prohibits the government from outlawing or restricting the exercise of a religious practice, but it generally does not prohibit withholding an affirmative benefit on account of religion. The District Court likened the case before it to Locke v. Davey, 540 U. S. 712 , where this Court upheld against a free exercise challenge a State’s decision not to fund degrees in devotional theology as part of a scholarship program. The District Court held that the Free Exercise Clause did not require the State to make funds available under the Scrap Tire Program to Trinity Lutheran. A divided panel of the Eighth Circuit affirmed. The fact that the State could award a scrap tire grant to Trinity Lutheran without running afoul of the Establishment Clause of the Federal Constitution, the court ruled, did not mean that the Free Exercise Clause compelled the State to disregard the broader antiestablishment principle reflected in its own Constitution. Held: The Department’s policy violated the rights of Trinity Lutheran under the Free Exercise Clause of the First Amendment by denying the Church an otherwise available public benefit on account of its religious status. Pp. 6–15. (a) This Court has repeatedly confirmed that denying a generally available benefit solely on account of religious identity imposes a penalty on the free exercise of religion. Thus, in McDaniel v. Paty, 435 U. S. 618 , the Court struck down a Tennessee statute disqualifying ministers from serving as delegates to the State’s constitutional convention. A plurality recognized that such a law discriminated against McDaniel by denying him a benefit solely because of his “status as a ‘minister.’ ” Id., at 627. In recent years, when rejecting free exercise challenges to neutral laws of general applicability, the Court has been careful to distinguish such laws from those that single out the religious for disfavored treatment. See, e.g., Lyng v. Northwest Indian Cemetery Protective Assn., 485 U. S. 439 ; Employment Div., Dept. of Human Resources of Ore. v. Smith, 494 U. S. 872 ; and Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U. S. 520 . It has remained a fundamental principle of this Court’s free exercise jurisprudence that laws imposing “special disabilities on the basis of . . . religious status” trigger the strictest scrutiny. Id., at 533. Pp. 6–9. (b) The Department’s policy expressly discriminates against otherwise eligible recipients by disqualifying them from a public benefit solely because of their religious character. Like the disqualification statute in McDaniel, the Department’s policy puts Trinity Lutheran to a choice: It may participate in an otherwise available benefit program or remain a religious institution. When the State conditions a benefit in this way, McDaniel says plainly that the State has imposed a penalty on the free exercise of religion that must withstand the most exacting scrutiny. 435 U. S., at 626, 628. The Department contends that simply declining to allocate to Trinity Lutheran a subsidy the State had no obligation to provide does not meaningfully burden the Church’s free exercise rights. Absent any such burden, the argument continues, the Department is free to follow the State’s antiestablishment objection to providing funds directly to a church. But, as even the Department acknowledges, the Free Exercise Clause protects against “indirect coercion or penalties on the free exercise of religion, not just outright prohibitions.” Lyng, 485 U. S., at 450. Trinity Lutheran is not claiming any entitlement to a subsidy. It is asserting a right to participate in a government benefit program without having to disavow its religious character. The express discrimination against religious exercise here is not the denial of a grant, but rather the refusal to allow the Church—solely because it is a church—to compete with secular organizations for a grant. Pp. 9–11. (c) The Department tries to sidestep this Court’s precedents by arguing that this case is instead controlled by Locke v. Davey. It is not. In Locke, the State of Washington created a scholarship program to assist high-achieving students with the costs of postsecondary education. Scholarship recipients were free to use state funds at accredited religious and non-religious schools alike, but they could not use the funds to pursue a devotional theology degree. At the outset, the Court made clear that Locke was not like the cases in which the Court struck down laws requiring individuals to “choose between their religious beliefs and receiving a government benefit.” 540 U. S., at 720–721. Davey was not denied a scholarship because of who he was; he was denied a scholarship because of what he proposed to do. Here there is no question that Trinity Lutheran was denied a grant simply because of what it is—a church. The Court in Locke also stated that Washington’s restriction on the use of its funds was in keeping with the State’s antiestablishment interest in not using taxpayer funds to pay for the training of clergy, an “essentially religious endeavor,” id., at 721. Here, nothing of the sort can be said about a program to use recycled tires to resurface playgrounds. At any rate, the Court took account of Washington’s antiestablishment interest only after determining that the scholarship program did not “require students to choose between their religious beliefs and receiving a government benefit.” Id., at 720–721. There is no dispute that Trinity Lutheran is put to the choice between being a church and receiving a government benefit. Pp. 11–14. (d) The Department’s discriminatory policy does not survive the “most rigorous” scrutiny that this Court applies to laws imposing special disabilities on account of religious status. Lukumi, 508 U. S., at 546. That standard demands a state interest “of the highest order” to justify the policy at issue. McDaniel, 435 U. S., at 628 (internal quotation marks omitted). Yet the Department offers nothing more than Missouri’s preference for skating as far as possible from religious establishment concerns. In the face of the clear infringement on free exercise before the Court, that interest cannot qualify as compelling. Pp. 14–15. 788 F. 3d 779, reversed and remanded. Roberts, C. J., delivered the opinion of the Court, except as to footnote 3. Kennedy, Alito, and Kagan, JJ., joined that opinion in full, and Thomas and Gorsuch, JJ., joined except as to footnote 3. Thomas, J., filed an opinion concurring in part, in which Gorsuch, J., joined. Gorsuch, J., filed an opinion concurring in part, in which Thomas, J., joined. Breyer, J., filed an opinion concurring in the judgment. Sotomayor, J., filed a dissenting opinion, in which Ginsburg, J., joined.

except as to footnote 3. The Missouri Department of Natural Resources offers state grants to help public and private schools, nonprofit daycare centers, and other nonprofit entities purchase rubber playground surfaces made from recycled tires. Trinity Lutheran Church applied for such a grant for its preschool and daycare center and would have received one, but for the fact that Trinity Lutheran is a church. The Department had a policy of categorically disqualifying churches and other religious organizations from receiving grants under its playground resurfacing program. The question presented is whether the Department’s policy violated the rights of Trinity Lutheran under the Free Exercise Clause of the First Amendment. I A The Trinity Lutheran Church Child Learning Center is a preschool and daycare center open throughout the year to serve working families in Boone County, Missouri, and the surrounding area. Established as a nonprofit organization in 1980, the Center merged with Trinity Lutheran Church in 1985 and operates under its auspices on church property. The Center admits students of any religion, and enrollment stands at about 90 children ranging from age two to five. The Center includes a playground that is equipped with the basic playground essentials: slides, swings, jungle gyms, monkey bars, and sandboxes. Almost the entire surface beneath and surrounding the play equipment is coarse pea gravel. Youngsters, of course, often fall on the playground or tumble from the equipment. And when they do, the gravel can be unforgiving. In 2012, the Center sought to replace a large portion of the pea gravel with a pour-in-place rubber surface by participating in Missouri’s Scrap Tire Program. Run by the State’s Department of Natural Resources to reduce the number of used tires destined for landfills and dump sites, the program offers reimbursement grants to qualifying nonprofit organizations that purchase playground surfaces made from recycled tires. It is funded through a fee imposed on the sale of new tires in the State. Due to limited resources, the Department cannot offer grants to all applicants and so awards them on a competitive basis to those scoring highest based on several criteria, such as the poverty level of the population in the surrounding area and the applicant’s plan to promote recycling. When the Center applied, the Department had a strict and express policy of denying grants to any applicant owned or controlled by a church, sect, or other religious entity. That policy, in the Department’s view, was compelled by Article I, Section 7 of the Missouri Constitution, which provides: “That no money shall ever be taken from the public treasury, directly or indirectly, in aid of any church, sect or denomination of religion, or in aid of any priest, preacher, minister or teacher thereof, as such; and that no preference shall be given to nor any discrimination made against any church, sect or creed of religion, or any form of religious faith or worship.” In its application, the Center disclosed its status as a ministry of Trinity Lutheran Church and specified that the Center’s mission was “to provide a safe, clean, and attractive school facility in conjunction with an educational program structured to allow a child to grow spiritually, physically, socially, and cognitively.” App. to Pet. for Cert. 131a. After describing the playground and the safety hazards posed by its current surface, the Center detailed the anticipated benefits of the proposed project: increasing access to the playground for all children, including those with disabilities, by providing a surface compliant with the Americans with Disabilities Act of 1990; providing a safe, long-lasting, and resilient surface under the play areas; and improving Missouri’s environment by putting recycled tires to positive use. The Center also noted that the benefits of a new surface would extend beyond its students to the local community, whose children often use the playground during non-school hours. The Center ranked fifth among the 44 applicants in the 2012 Scrap Tire Program. But despite its high score, the Center was deemed categorically ineligible to receive a grant. In a letter rejecting the Center’s application, the program director explained that, under Article I, Section 7 of the Missouri Constitution, the Department could not provide financial assistance directly to a church. The Department ultimately awarded 14 grants as part of the 2012 program. Because the Center was operated by Trinity Lutheran Church, it did not receive a grant. B Trinity Lutheran sued the Director of the Department in Federal District Court. The Church alleged that the Department’s failure to approve the Center’s application, pursuant to its policy of denying grants to religiously affiliated applicants, violates the Free Exercise Clause of the First Amendment. Trinity Lutheran sought declara-tory and injunctive relief prohibiting the Department from discriminating against the Church on that basis in future grant applications. The District Court granted the Department’s motion to dismiss. The Free Exercise Clause, the District Court stated, prohibits the government from outlawing or restricting the exercise of a religious practice; it generally does not prohibit withholding an affirmative benefit on account of religion. The District Court likened the Department’s denial of the scrap tire grant to the situation this Court encountered in Locke v. Davey, 540 U. S. 712 (2004) . In that case, we upheld against a free exercise challenge the State of Washington’s decision not to fund degrees in devotional theology as part of a state scholarship program. Finding the present case “nearly indistinguishable from Locke,” the District Court held that the Free Exercise Clause did not require the State to make funds available under the Scrap Tire Program to religious institutions like Trinity Lutheran. Trinity Lutheran Church of Columbia, Inc. v. Pauley, 976 F. Supp. 2d 1137, 1151 (WD Mo. 2013). The Court of Appeals for the Eighth Circuit affirmed. The court recognized that it was “rather clear” that Missouri could award a scrap tire grant to Trinity Lutheran without running afoul of the Establishment Clause of the United States Constitution. Trinity Lutheran Church of Columbia, Inc. v. Pauley, 788 F. 3d 779, 784 (2015). But, the Court of Appeals explained, that did not mean the Free Exercise Clause compelled the State to disregard the antiestablishment principle reflected in its own Constitution. Viewing a monetary grant to a religious institution as a “ ‘hallmark[ ] of an established religion,’ ” the court concluded that the State could rely on an applicant’s religious status to deny its application. Id., at 785 (quoting Locke, 540 U. S., at 722; some internal quotation marks omitted). Judge Gruender dissented. He distinguished Locke on the ground that it concerned the narrow issue of funding for the religious training of clergy, and “did not leave states with unfettered discretion to exclude the religious from generally available public benefits.” 788 F. 3d, at 791 (opinion concurring in part and dissenting in part). Rehearing en banc was denied by an equally divided court. We granted certiorari sub nom. Trinity Lutheran Church of Columbia, Inc. v. Pauley, 577 U. S. ___ (2016), and now reverse.[1] II The First Amendment provides, in part, that “Congress shall make no law respecting an establishment of religion, or prohibiting the free exercise thereof.” The parties agree that the Establishment Clause of that Amendment does not prevent Missouri from including Trinity Lutheran in the Scrap Tire Program. That does not, however, answer the question under the Free Exercise Clause, because we have recognized that there is “play in the joints” between what the Establishment Clause permits and the Free Exercise Clause compels. Locke, 540 U. S., at 718 (internal quotation marks omitted). The Free Exercise Clause “protect[s] religious observers against unequal treatment” and subjects to the strictest scrutiny laws that target the religious for “special disabilities” based on their “religious status.” Church of Lukumi Babalu Aye, Inc. v. Hialeah, 508 U. S. 520, 533, 542 (1993) (internal quotation marks omitted). Applying that basic principle, this Court has repeatedly confirmed that denying a generally available benefit solely on account of religious identity imposes a penalty on the free exercise of religion that can be justified only by a state interest “of the highest order.” McDaniel v. Paty, 435 U. S. 618, 628 (1978) (plurality opinion) (quoting Wisconsin v. Yoder, 406 U. S. 205, 215 (1972) ). In Everson v. Board of Education of Ewing, 330 U. S. 1 (1947) , for example, we upheld against an Establishment Clause challenge a New Jersey law enabling a local school district to reimburse parents for the public transportation costs of sending their children to public and private schools, including parochial schools. In the course of ruling that the Establishment Clause allowed New Jersey to extend that public benefit to all its citizens regardless of their religious belief, we explained that a State “cannot hamper its citizens in the free exercise of their own religion. Consequently, it cannot exclude individual Catholics, Lutherans, Mohammedans, Baptists, Jews, Methodists, Non-believers, Presbyterians, or the members of any other faith, because of their faith, or lack of it, from receiving the benefits of public welfare legislation.” Id., at 16. Three decades later, in McDaniel v. Paty, the Court struck down under the Free Exercise Clause a Tennessee statute disqualifying ministers from serving as delegates to the State’s constitutional convention. Writing for the plurality, Chief Justice Burger acknowledged that Tennessee had disqualified ministers from serving as legislators since the adoption of its first Constitution in 1796, and that a number of early States had also disqualified ministers from legislative office. This historical tradition, however, did not change the fact that the statute discriminated against McDaniel by denying him a benefit solely because of his “status as a ‘minister.’ ” 435 U. S., at 627. McDaniel could not seek to participate in the convention while also maintaining his role as a minister; to pursue the one, he would have to give up the other. In this way, said Chief Justice Burger, the Tennessee law “effectively penalizes the free exercise of [McDaniel’s] constitutional liberties.” Id., at 626 (quoting Sherbert v. Verner, 374 U. S. 398, 406 (1963) ; internal quotation marks omitted). Joined by Justice Marshall in concurrence, Justice Brennan added that “because the challenged provision requires [McDaniel] to purchase his right to engage in the ministry by sacrificing his candidacy it impairs the free exercise of his religion.” McDaniel, 435 U. S., at 634. In recent years, when this Court has rejected free exercise challenges, the laws in question have been neutral and generally applicable without regard to religion. We have been careful to distinguish such laws from those that single out the religious for disfavored treatment. For example, in Lyng v. Northwest Indian Cemetery Protective Association, 485 U. S. 439 (1988) , we held that the Free Exercise Clause did not prohibit the Government from timber harvesting or road construction on a particular tract of federal land, even though the Government’s action would obstruct the religious practice of several Native American Tribes that held certain sites on the tract to be sacred. Accepting that “[t]he building of a road or the harvesting of timber . . . would interfere significantly with private persons’ ability to pursue spiritual fulfillment according to their own religious beliefs,” we nonetheless found no free exercise violation, because the affected individuals were not being “coerced by the Government’s action into violating their religious beliefs.” Id., at 449. The Court specifically noted, however, that the Government action did not “penalize religious activity by denying any person an equal share of the rights, benefits, and privileges enjoyed by other citizens.” Ibid. In Employment Division, Department of Human Resources of Oregon v. Smith, 494 U. S. 872 (1990) , we rejected a free exercise claim brought by two members of a Native American church denied unemployment benefits because they had violated Oregon’s drug laws by ingesting peyote for sacramental purposes. Along the same lines as our decision in Lyng, we held that the Free Exercise Clause did not entitle the church members to a special dispensation from the general criminal laws on account of their religion. At the same time, we again made clear that the Free Exercise Clause did guard against the government’s imposition of “special disabilities on the basis of religious views or religious status.” 494 U. S., at 877 (citing McDaniel, 435 U. S. 618 ).[2] Finally, in Church of Lukumi Babalu Aye, Inc. v. Hia-leah, we struck down three facially neutral city ordinances that outlawed certain forms of animal slaughter. Members of the Santeria religion challenged the ordinances under the Free Exercise Clause, alleging that despite their facial neutrality, the ordinances had a discriminatory purpose easy to ferret out: prohibiting sacrificial rituals integral to Santeria but distasteful to local residents. We agreed. Before explaining why the challenged ordinances were not, in fact, neutral or generally applicable, the Court recounted the fundamentals of our free exercise jurisprudence. A law, we said, may not discriminate against “some or all religious beliefs.” 508 U. S., at 532. Nor may a law regulate or outlaw conduct because it is religiously motivated. And, citing McDaniel and Smith, we restated the now-familiar refrain: The Free Exercise Clause protects against laws that “ ‘impose[ ] special dis-abilities on the basis of . . . religious status.’ ” 508 U. S., at 533 (quoting Smith, 494 U. S., at 877); see also Mitchell v. Helms, 530 U. S. 793, 828 (2000) (plurality opinion) (noting “our decisions that have prohibited governments from discriminating in the distribution of public benefits based upon religious status or sincerity” (citing Rosenberger v. Rector and Visitors of Univ. of Va., 515 U. S. 819 (1995) ; Lamb’s Chapel v. Center Moriches Union Free School Dist., 508 U. S. 384 (1993) ; Widmar v. Vincent, 454 U. S. 263 (1981) )). III A The Department’s policy expressly discriminates against otherwise eligible recipients by disqualifying them from a public benefit solely because of their religious character. If the cases just described make one thing clear, it is that such a policy imposes a penalty on the free exercise of religion that triggers the most exacting scrutiny. Lukumi, 508 U. S., at 546. This conclusion is unremarkable in light of our prior decisions. Like the disqualification statute in McDaniel, the Department’s policy puts Trinity Lutheran to a choice: It may participate in an otherwise available benefit program or remain a religious institution. Of course, Trinity Lu-theran is free to continue operating as a church, just as McDaniel was free to continue being a minister. But that freedom comes at the cost of automatic and absolute exclusion from the benefits of a public program for which the Center is otherwise fully qualified. And when the State conditions a benefit in this way, McDaniel says plainly that the State has punished the free exercise of religion: “To condition the availability of benefits . . . upon [a recipient’s] willingness to . . . surrender[ ] his religiously impelled [status] effectively penalizes the free exercise of his constitutional liberties.” 435 U. S., at 626 (plurality opinion) (alterations omitted). The Department contends that merely declining to extend funds to Trinity Lutheran does not prohibit the Church from engaging in any religious conduct or otherwise exercising its religious rights. In this sense, says the Department, its policy is unlike the ordinances struck down in Lukumi, which outlawed rituals central to Santeria. Here the Department has simply declined to allocate to Trinity Lutheran a subsidy the State had no obligation to provide in the first place. That decision does not meaningfully burden the Church’s free exercise rights. And absent any such burden, the argument continues, the Department is free to heed the State’s antiestablishment objection to providing funds directly to a church. Brief for Respondent 7–12, 14–16. It is true the Department has not criminalized the way Trinity Lutheran worships or told the Church that it cannot subscribe to a certain view of the Gospel. But, as the Department itself acknowledges, the Free Exercise Clause protects against “indirect coercion or penalties on the free exercise of religion, not just outright prohibitions.” Lyng, 485 U. S., at 450. As the Court put it more than 50 years ago, “[i]t is too late in the day to doubt that the liberties of religion and expression may be infringed by the denial of or placing of conditions upon a benefit or privilege.” Sherbert, 374 U. S., at 404; see also McDaniel, 435 U. S., at 633 (Brennan, J., concurring in judgment) (The “proposition—that the law does not interfere with free exercise because it does not directly prohibit religious activity, but merely conditions eligibility for office on its abandonment—is . . . squarely rejected by precedent”). Trinity Lutheran is not claiming any entitlement to a subsidy. It instead asserts a right to participate in a government benefit program without having to disavow its religious character. The “imposition of such a condition upon even a gratuitous benefit inevitably deter[s] or discourage[s] the exercise of First Amendment rights.” Sherbert, 374 U. S., at 405. The express discrimination against religious exercise here is not the denial of a grant, but rather the refusal to allow the Church—solely because it is a church—to compete with secular organizations for a grant. Cf. Northeastern Fla. Chapter, Associated Gen. Contractors of America v. Jacksonville, 508 U. S. 656, 666 (1993) (“[T]he ‘injury in fact’ is the inability to compete on an equal footing in the bidding process, not the loss of a contract”). Trinity Lutheran is a member of the community too, and the State’s decision to exclude it for purposes of this public program must withstand the strictest scrutiny. B The Department attempts to get out from under the weight of our precedents by arguing that the free exercise question in this case is instead controlled by our decision in Locke v. Davey. It is not. In Locke, the State of Washington created a scholarship program to assist high-achieving students with the costs of postsecondary education. The scholarships were paid out of the State’s general fund, and eligibility was based on criteria such as an applicant’s score on college admission tests and family income. While scholarship recipients were free to use the money at accredited religious and non-religious schools alike, they were not permitted to use the funds to pursue a devotional theology degree—one “devotional in nature or designed to induce religious faith.” 540 U. S., at 716 (internal quotation marks omitted). Davey was selected for a scholarship but was denied the funds when he refused to certify that he would not use them toward a devotional degree. He sued, arguing that the State’s refusal to allow its scholarship money to go toward such degrees violated his free exercise rights. This Court disagreed. It began by explaining what was not at issue. Washington’s selective funding program was not comparable to the free exercise violations found in the “Lukumi line of cases,” including those striking down laws requiring individuals to “choose between their religious beliefs and receiving a government benefit.” Id., at 720–721. At the outset, then, the Court made clear that Locke was not like the case now before us. Washington’s restriction on the use of its scholarship funds was different. According to the Court, the State had “merely chosen not to fund a distinct category of instruction.” Id., at 721. Davey was not denied a scholarship because of who he was; he was denied a scholarship because of what he proposed to do—use the funds to prepare for the ministry. Here there is no question that Trinity Lutheran was denied a grant simply because of what it is—a church. The Court in Locke also stated that Washington’s choice was in keeping with the State’s antiestablishment interest in not using taxpayer funds to pay for the training of clergy; in fact, the Court could “think of few areas in which a State’s antiestablishment interests come more into play.” Id., at 722. The claimant in Locke sought funding for an “essentially religious endeavor . . . akin to a religious calling as well as an academic pursuit,” and opposition to such funding “to support church leaders” lay at the historic core of the Religion Clauses. Id., at 721–722. Here nothing of the sort can be said about a program to use recycled tires to resurface playgrounds. Relying on Locke, the Department nonetheless emphasizes Missouri’s similar constitutional tradition of not furnishing taxpayer money directly to churches. Brief for Respondent 15–16. But Locke took account of Washington’s antiestablishment interest only after determining, as noted, that the scholarship program did not “require students to choose between their religious beliefs and receiving a government benefit.” 540 U. S., at 720–721 (citing McDaniel, 435 U. S. 618 ). As the Court put it, Washington’s scholarship program went “a long way toward including religion in its benefits.” Locke, 540 U. S., at 724. Students in the program were free to use their scholarships at “pervasively religious schools.” Ibid. Davey could use his scholarship to pursue a secular degree at one institution while studying devotional theology at another. Id., at 721, n. 4. He could also use his scholarship money to attend a religious college and take devotional theology courses there. Id., at 725. The only thing he could not do was use the scholarship to pursue a degree in that subject. In this case, there is no dispute that Trinity Lutheran is put to the choice between being a church and receiving a government benefit. The rule is simple: No churches need apply.[3] C The State in this case expressly requires Trinity Lutheran to renounce its religious character in order to participate in an otherwise generally available public benefit program, for which it is fully qualified. Our cases make clear that such a condition imposes a penalty on the free exercise of religion that must be subjected to the “most rigorous” scrutiny. Lukumi, 508 U. S., at 546.[4] Under that stringent standard, only a state interest “of the highest order” can justify the Department’s discriminatory policy. McDaniel, 435 U. S., at 628 (internal quotation marks omitted). Yet the Department offers nothing more than Missouri’s policy preference for skating as far as possible from religious establishment concerns. Brief for Respondent 15–16. In the face of the clear infringement on free exercise before us, that interest cannot qual-ify as compelling. As we said when considering Missouri’s same policy preference on a prior occasion, “the state interest asserted here—in achieving greater separation of church and State than is already ensured under the Establishment Clause of the Federal Constitution—is limited by the Free Exercise Clause.” Widmar, 454 U. S., at 276. The State has pursued its preferred policy to the point of expressly denying a qualified religious entity a public benefit solely because of its religious character. Under our precedents, that goes too far. The Department’s policy violates the Free Exercise Clause.[5] * * * Nearly 200 years ago, a legislator urged the Maryland Assembly to adopt a bill that would end the State’s disqualification of Jews from public office: “If, on account of my religious faith, I am subjected to disqualifications, from which others are free, . . . I cannot but consider myself a persecuted man. . . . An odious exclusion from any of the benefits common to the rest of my fellow-citizens, is a persecution, differing only in degree, but of a nature equally unjustifiable with that, whose instruments are chains and torture.” Speech by H. M. Brackenridge, Dec. Sess. 1818, in H. Brackenridge, W. Worthington, & J. Tyson, Speeches in the House of Delegates of Maryland, 64 (1829). The Missouri Department of Natural Resources has not subjected anyone to chains or torture on account of religion. And the result of the State’s policy is nothing so dramatic as the denial of political office. The consequence is, in all likelihood, a few extra scraped knees. But the exclusion of Trinity Lutheran from a public benefit for which it is otherwise qualified, solely because it is a church, is odious to our Constitution all the same, and cannot stand. The judgment of the United States Court of Appeals for the Eighth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion. It is so ordered.Notes 1 In April 2017, the Governor of Missouri announced that he had directed the Department to begin allowing religious organizations to compete for and receive Department grants on the same terms as secular organizations. That announcement does not moot this case. We have said that such voluntary cessation of a challenged practice does not moot a case unless “subsequent events ma[ke] it absolutely clear that the allegedly wrongful behavior could not reasonably be expected to recur.” Friends of the Earth, Inc. v. Laidlaw Environmental Services (TOC), Inc., 528 U. S. 167, 189 (2000) (internal quotation marks omitted). The Department has not carried the “heavy burden” of making “absolutely clear” that it could not revert to its policy of excluding religious organizations. Ibid. The parties agree. See Letter from James R. Layton, Counsel for Respondent, to Scott S. Harris, Clerk of Court (Apr. 18, 2017) (adopting the position of the Missouri Attorney General’s Office that “there is no clearly effective barrier that would prevent the [Department] from reinstating [its] policy in the future”); Letter from David A. Cortman, Counsel for Petitioner, to Scott S. Harris, Clerk of Court (Apr. 18, 2017) (“[T]he policy change does nothing to remedy the source of the [Department’s] original policy—the Missouri Supreme Court’s interpretation of Article 1, §7 of the Missouri Constitution”). 2 This is not to say that any application of a valid and neutral law of general applicability is necessarily constitutional under the Free Exercise Clause. Recently, in Hosanna-Tabor Evangelical Lutheran Church and School v. EEOC, 565 U. S. 171 (2012) , this Court held that the Religion Clauses required a ministerial exception to the neutral prohibition on employment retaliation contained in the Americans with Disabilities Act. Distinguishing Smith, we explained that while that case concerned government regulation of physical acts, “[t]he present case, in contrast, concerns government interference with an internal church decision that affects the faith and mission of the church itself.” 565 U. S., at 190. 3 This case involves express discrimination based on religious identity with respect to playground resurfacing. We do not address religious uses of funding or other forms of discrimination. 4 We have held that “a law targeting religious beliefs as such is never permissible.” Lukumi, 508 U. S., at 533; see also McDaniel v. Paty, 435 U. S. 618, 626 (1978) (plurality opinion). We do not need to decide whether the condition Missouri imposes in this case falls within the scope of that rule, because it cannot survive strict scrutiny in any event. 5 Based on this holding, we need not reach the Church’s claim that the policy also violates the Equal Protection Clause.