Department of Agriculture Rural Development Rural Housing Service v. Kirtz, 601 U.S. ___ (2024)

NOTICE: This opinion is subject to formal revision before publication in the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, pio@supremecourt.gov, of any typographical or other formal errors.

SUPREME COURT OF THE UNITED STATES

_________________

No. 22–846

_________________

DEPARTMENT OF AGRICULTURE RURAL DEVELOP- MENT RURAL HOUSING SERVICE, PETITIONER v. REGINALD KIRTZ

on writ of certiorari to the united states court of appeals for the third circuit

[February 8, 2024]

Justice Gorsuch delivered the opinion of the Court.

A credit report can determine everything from whether a person can secure a credit card, purchase a home, win a new job, or start a small business. Recognizing the importance of accuracy in credit reporting, Congress adopted the Fair Credit Reporting Act in 1970 (FCRA). In its present form, the Act allows consumers to sue private lenders who willfully or negligently supply false information about them to agencies that generate credit reports. The question we face is whether one of the Nation’s largest lenders—the federal government—is also susceptible to suit when it supplies false information, or whether it may invoke sovereign immunity to avoid liability.

I

This case arises from a loan Reginald Kirtz secured from the Rural Housing Service. The Service, a division of the United States Department of Agriculture (USDA), “issues loans to promote the development of safe and affordable housing in rural communities.” Kirtz v. Trans Union LLC, 46 F. 4th 159, 163 (CA3 2022). According to Mr. Kirtz, he repaid his loan in full by mid-2018. See Amended Complaint in No. 2:20–cv–05231 (ED Pa.), ECF Doc. 20, p. 3, ¶11. Despite this, the USDA repeatedly told TransUnion, a company engaged in the business of preparing consumer credit reports, that his account was past due. Ibid., ¶12. These misrepresentations damaged his credit score and threatened his ability to secure future loans at affordable rates. See id., at 3–4, ¶¶12–14. In an effort to resolve the problem, Mr. Kirtz alerted TransUnion to the error, and the company, in turn, notified the USDA. Id., at 5, ¶¶16, 20. But, Mr. Kirtz says, the USDA failed to take “any action to investigate or correct” its records. 46 F. 4th, at 163. So he eventually decided to sue the agency under the FCRA. Ibid.

As originally enacted in 1970, the FCRA focused largely on two groups. First, it addressed “consumer reporting agenc[ies]” like TransUnion, charging them with various new duties designed to ensure the accuracy and confidentiality of their work. See, e.g., 84Stat. 1129, 1132; 15 U. S. C. §§1681b, 1681i. Second, it imposed new regulations on “person[s]” who procure credit information from consumer reporting agencies. So, for example, the Act provided that a “person” who requests “an investigative consumer report on any consumer” must inform the consumer in writing “not later than three days after the date on which the report was first requested.” 84Stat. 1130; see §1681d(a). The FCRA proceeded to define the term “person” broadly to “mea[n] any individual, partnership, corporation, trust, estate, cooperative, association, government or governmental subdivision or agency, or other entity.” 84Stat. 1128; see §1681a(b). The Act further authorized consumers to seek damages for violations of its terms, but only against consumer reporting agencies and those who use the information they produce. 84Stat. 1134; see §§1681n, 1681o (1970 ed.).

In the Consumer Credit Reporting Reform Act of 1996, Congress amended the FCRA to broaden its reach. As relevant here, Congress added provisions addressing those who furnish information to consumer reporting agencies. Referencing the definition of “person” it had adopted in 1970, Congress instructed that, if a consumer disputes “the completeness or accuracy” of his credit information, the “person” who furnished it must investigate the matter and take steps to correct any mistake. 110Stat. 3009–448; see §1681s–2(b). To enforce these new duties, Congress revised the 1970 Act’s remedial provisions. Where it had once authorized consumer suits against only consumer reporting agencies and users of their information, Congress now authorized consumer suits against “[a]ny person” who willfully or negligently fails to comply with “any” of the law’s “requirement[s].” 110Stat. 3009–446; see §§1681n(a), 1681o(a).

Mr. Kirtz sought relief under these new provisions. According to his complaint, the USDA furnished information to TransUnion. The agency had notice that the information it supplied was false. That false information impaired Mr. Kirtz’s ability to access affordable credit. Yet the agency failed to take any steps to correct its mistake—either willfully (in violation of §1681n) or negligently (in violation of §1681o). By way of remedy, Mr. Kirtz sought money damages consistent with what the FCRA allows. See Amended Complaint 12.

In response, the USDA moved to dismiss the complaint. The agency did not dispute that allegations like Mr. Kirtz’s state a viable claim for relief. Instead, it pointed to this Court’s precedents holding that, as sovereign, the federal government enjoys immunity from suits for money damages unless Congress waives that immunity. And, the agency contended, nothing in the FCRA purports to render the federal government amenable to suit. The district court sided with the USDA, but the Third Circuit reversed. Speaking for a unanimous panel, Judge Krause observed that §§1681n and 1681o authorize suits for damages against “any person” who violates the Act, and §1681a expressly defines “person” to include “any” government agency. 46 F. 4th, at 164–166.

The question whether Mr. Kirtz may sue the federal government holds significance for many. A 2021 study cited by the Consumer Financial Protection Bureau “found that over 34% of consumers surveyed were able to identify at least one error in their credit reports.” 87 Fed. Reg. 64689–64690 (2022). Mistakes like these can lead lenders to insist on higher interest rates or other terms that make it “difficult or impossible” for consumers “to obtain a mortgage, auto loan, student loan, or other credit.” Id., at 64689. These days, too, federal agencies are among “ ‘the largest furnishers of credit information’ ” to consumer reporting agencies. Brief for Petitioner 38 (quoting Robinson v. Department of Education, 590 U. S. ___, ___ (2020) (Thomas, J., dissenting from denial of certiorari) (slip op., at 3)). Yet the lower courts have reached different views on the question whether federal agencies are answerable under the FCRA for their mistakes. Like the Third Circuit, the Seventh and D. C. Circuits have held that the FCRA authorizes suits against government agencies no less than it does private lenders. The Fourth and Ninth Circuits, by contrast, have held that sovereign immunity bars consumer suits against federal agencies. We agreed to hear this case to resolve that conflict. 599 U. S. ___ (2023).

II

The parties agree on the principles that guide our analysis even as they disagree on the answer those principles yield. Under this Court’s precedents, both sides acknowledge, the United States, as sovereign, is generally immune from suits seeking money damages. See, e.g., United States v. Testan, 424 U.S. 392, 399 (1976). At the same time, Congress may choose to waive that immunity. Ibid. But because the power to waive the federal government’s immunity is Congress’s prerogative, not ours, this Court applies a “clear statement” rule. Under the rule’s terms, we will permit a suit against the government only when a statute “unmistakabl[y]” allows it. FAA v. Cooper, 566 U.S. 284, 291 (2012). “Congress need not state its intent in any particular way.” Ibid. It need not “use magic words.” Ibid. Nor must it “make its clear statement in a single section or in statutory provisions enacted at the same time.” Kimel v. Florida Bd. of Regents, 528 U.S. 62, 76 (2000). But, one way or another, a waiver of sovereign immunity must be “unmistakably clear in the language of the statute.” Id., at 73 (internal quotation marks omitted).

Necessarily, this inquiry trains on statutory text rather than legislative history. Because “[a]ny ambiguities in the statutory language are to be construed in favor of immunity,” no amount of legislative history can “supply a waiver that is not clearly evident from the language of the statute.” Cooper, 566 U. S., at 290; accord, Lane v. Peña, 518 U.S. 187, 192 (1996). Conversely, when an “ ‘unmistakably clear’ ” waiver of sovereign immunity appears in a statute, no amount of legislative history can dislodge it. Dellmuth v. Muth, 491 U.S. 223, 230 (1989); see Food Marketing Institute v. Argus Leader Media, 588 U.S. 427, 436 (2019) (“Even those of us who sometimes consult legislative history will never allow it to be used to muddy the meaning of clear statutory language” (internal quotation marks omitted)). Either way, then, a court charged with asking whether Congress has spoken clearly has its answer long before it might have reason to consult the Congressional Record.

To date, this Court has found a clear waiver of sovereign immunity “in only two situations.” Financial Oversight and Management Bd. for P. R. v. Centro De Periodismo Investigativo, Inc., 598 U.S. 339, 347 (2023). “The first is when a statute says in so many words that it is stripping immunity from a sovereign entity.” Ibid. Congress has employed this approach in the Bankruptcy Code, for example, where it has stated that, “[n]otwithstanding an assertion of sovereign immunity, sovereign immunity is abrogated as to a governmental unit . . . with respect to” enumerated provisions of the Code. 11 U. S. C. §106(a). The second situation “is when a statute creates a cause of action” and explicitly “authorizes suit against a government on that claim.” Financial Oversight and Management Bd., 598 U. S., at 347. Statutes like these may not discuss sovereign immunity in so many words. But dismissing a claim against the government in these circumstances would effectively “negat[e]” a claim Congress has clearly authorized. Id., at 348.

The Court encountered a statute falling into this second category in Kimel, 528 U.S. 62. That case involved the question whether the Age Discrimination in Employment Act of 1967 (ADEA) abrogated state sovereign immunity. As originally enacted, the ADEA authorized employees to bring claims against employers who discriminate based on age. 29 U. S. C. §§623(a), 626(c)(1). A later amendment incorporated into the ADEA a provision of the Fair Labor Standards Act of 1938 (FLSA) that extended the right to bring claims for age-based discrimination “against any employer (including a public agency).” §216(b) (emphasis added); see §626(b). And, elsewhere, the FLSA defined “ ‘[p]ublic agency’ ” to include “the government of a State or political subdivision thereof ” and “any agency of . . . a State, or a political subdivision of a State.” §203(x). In light of these amendments and their cross-references, Kimel held, the statute’s “plain language . . . clearly demonstrate[d] Congress[’s] intent to subject the States to suit for money damages at the hands of individual employees.” 528 U. S., at 74.

Guided by these principles, we think the Third Circuit reached the right decision in this case: The FCRA effects a clear waiver of sovereign immunity. In §1681s–2, the Act requires “person[s]” who furnish information to consumer reporting agencies to investigate consumer complaints and make any necessary corrections. 15 U. S. C. §1681s–2(b). In §§1681n and 1681o, the Act authorizes consumer suits for money damages against “[a]ny person” who willfully or negligently fails to comply with this directive. §§1681n(a), 1681o(a). In §1681a, the Act defines the term “ ‘person’ ” to include “any . . . governmental . . . agency.” §1681a(b). And the same provision instructs us to apply this definition throughout the entire “subchapter” where §§1681n and 1681o appear. §1681a(a). Through this series of statutory directions, no less than those we encountered in Kimel, Congress has explicitly permitted consumer claims for damages against the government. Dismissing suits like Mr. Kirtz’s would effectively “negat[e]” suits Congress has clearly authorized. Financial Oversight and Management Bd., 598 U. S., at 348.

We need look no further to resolve this case. But if we do, other portions of the FCRA point to the same conclusion. Section 1681a(y) excludes from the definition of “consumer report” certain communications that “are not provided to any person except . . . any Federal or State officer, agency, or department, or any officer, agency, or department of a unit of general local government.” (Emphasis added.) Section 1681b requires a “person” who intends to take an “adverse [employment] action” based on a consumer credit report to provide the affected individual with a copy of the report unless “an agency or department of the United States Government” seeks to use the report as part of a national security investigation. §§1681b(b)(3)(A), (4)(A). Both provisions thus exempt government agencies from the Act’s otherwise-broad definition of “person” for particular reasons in particular contexts. All of which tends to confirm what the Act tells us explicitly: Throughout the Act, the term “person” includes the government unless otherwise noted. See Southwest Airlines Co. v. Saxon, 596 U.S. 450, 457–458 (2022) (recounting the traditional canon of construction that material variations in usage can illuminate a statute’s meaning).

To be sure, there are other provisions in the FCRA—just as there are elsewhere in the U. S. Code—that address the question of sovereign immunity in different and arguably even more obvious terms. For example, Congress added §1681u to the FCRA as part of the Intelligence Authorization Act for Fiscal Year 1996. See 109Stat. 974. That provision allows the Federal Bureau of Investigation to access consumer information, subject to a number of constraints. See §§1681u(a)–(d). At the same time, the statute indicates that the failure to respect those constraints can expose “[a]ny agency or department of the United States” to “liab[ility] to the consumer” for money damages. §1681u(j). While nothing in §1681u discusses sovereign immunity as such, everyone agrees its language clearly waives sovereign immunity.

None of that, however, makes the waiver of sovereign immunity reflected in the provisions now before us any less clear. “If no magic words are required” to waive sovereign immunity, then the clarity of “each statute must be evaluated on its own terms.” Pennsylvania v. Union Gas Co., 491 U.S. 1, 13, n. 4 (1989). And the fact that Congress chose to use certain language to waive sovereign immunity in one amendment to the FCRA hardly means it was “foreclose[d] . . . from using different language to accomplish th[e] same goal” in a different set of amendments to the same law. Lac du Flambeau Band of Lake Superior Chippewa Indians v. Coughlin, 599 U.S. 382, 395 (2023). The question we must answer is not whether §§1681a, 1681n, and 1681o speak in the same terms as §1681u. The only question we face is whether those provisions speak clearly to the government’s

liability. Because they do, that is the end of the matter.[1]

III

A

While the government largely accepts our understanding of this Court’s sovereign-immunity jurisprudence, it disputes some of the finer points. As an initial matter, the government asserts that, “to impose liability on a sovereign, a plaintiff must identify both a ‘source of substantive law’ that ‘provides an avenue for relief’ and ‘a waiver of sovereign immunity.’ ” Brief for Petitioner 14 (quoting FDIC v. Meyer, 510 U.S. 471, 484 (1994)). The implication is that a cause of action explicitly against the government is insufficient unless accompanied by a separate provision addressing sovereign immunity. See Brief for Petitioner 14.

That implication is incorrect. At the risk of repeating ourselves, a cause of action authorizing suit against the government may waive sovereign immunity even without a separate waiver provision. Financial Oversight and Management Bd., 598 U. S., at 347; see, e.g., Seminole Tribe of Fla. v. Florida, 517 U.S. 44, 56–57 (1996) (Congress abrogated state sovereign immunity when it explicitly authorized a cause of action against “Stat[es],” despite the absence of a separate waiver provision). Nor does FDIC v. Meyer, where this Court refused to recognize an implied cause of action, say anything to the contrary. See 510 U. S., at 484. The government must know as much. Why else would it hold out §1681u—a section that contains an express cause of action against the government but no separate waiver provision—as “a model for authorizing suits against the United States”? Brief for Petitioner 34.

Changing tack but pursuing the same end, the government points to the canon against superfluity. Proper respect for Congress cautions courts against lightly assuming that any of the statutory terms it has chosen to employ are “superfluous” or “void” of significance. TRW Inc. v. Andrews, 534 U.S. 19, 31 (2001) (internal quotation marks omitted). From this familiar teaching, the government seeks to extrapolate a new rule: A provision can waive sovereign immunity only if that provision would have no other role to play in the statutory scheme. Brief for Petitioner 20–21. That rule should foreclose suit here, the government submits, because allowing federal agencies a sovereign-immunity defense would not foreclose every suit under §§1681n and 1681o. See id., at 24. After all, even if consumers injured by government agencies could not seek relief under these provisions, other consumers harmed by private creditors still could. See ibid.

We cannot agree with this suggestion any more than the last. The canon against rendering statutory terms a nullity has a long lineage. But this Court has never endorsed the notion that a statute may effect a waiver of sovereign immunity only if that is the sole work it performs. Doing so would (again) effectively force Congress to address sovereign immunity in so many words in a discrete statutory provision. It would come perilously close, as well, to imposing a “magic-words” requirement. For good reason, then, the government’s supposed rule appears in none of the decisions it directs us to—not in Seminole Tribe of Fla., 517 U.S. 44, not in Kimel, 528 U.S. 62, and not in Nevada Dept. of Human Resources v. Hibbs, 538 U.S. 721 (2003). See Brief for Petitioner 18–19.

The government has another theory to offer. We may not find a waiver of sovereign immunity, it suggests, “when a cause of action merely cross-references a general definition that includes sovereigns along with non-sovereigns.” Id., at 22. Running with this idea, the government concedes that Congress would have clearly waived sovereign immunity if it had “plug[ged]” the full definition of “persons” from §1681a directly into §§1681n and 1681o. Tr. of Oral Arg. 7; accord, Brief for Petitioner 22. But, the government argues, a waiver of sovereign immunity cannot be effected by reading these provisions in combination.

This theory encounters its own difficulties. Under this Court’s precedents, Congress need not “make its clear statement in a single section” adopted at a single moment in time. Kimel, 528 U. S., at 76. Instead, what matters is whether Congress has authorized a waiver of sovereign immunity that is “clearly discernible” from the sum total of its work. Lac du Flambeau, 599 U. S., at 388 (internal quotation marks omitted). Were the rule otherwise, large swathes of our modern sovereign-immunity case law would be cast into doubt. After all, in Kimel this Court relied on the ADEA’s incorporation of the FLSA’s enforcement provision, and the latter provision’s incorporation, in turn, of a separate definitional provision. See 528 U. S., at 73–75. In Union Gas, the Court relied on the definition of “person” in the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 along with other clues from surrounding statutory provisions. 491 U. S., at 7–10. And in Seminole Tribe we “confirm[ed]” the “clear statement in one statutory subsection by looking to provisions in [an]other subsection.” Kimel, 528 U. S., at 104–105 (Thomas, J., concurring in part and dissenting in part) (citing Seminole Tribe, 517 U. S., at 56–57).

Alternatively still, the government points to Atascadero State Hospital v. Scanlon, 473 U.S. 234 (1985), and Employees of Dept. of Public Health and Welfare of Mo. v. Department of Public Health and Welfare of Mo., 411 U.S. 279 (1973). These cases, the government insists, impose still other and more demanding rules a court must follow before finding a waiver of sovereign immunity. See Brief for Petitioner 21, 25.

To appreciate the problem with this line of thinking, some background helps. For a period in the mid-20th century, this Court’s approach to sovereign immunity looked considerably different than it does today (or did before). Back then, in cases like Parden v. Terminal R. Co. of Ala. Docks Dept., 377 U.S. 184 (1964), this Court was content to do away with state sovereign immunity without clear authorization from Congress. Instead, the Court would infer a congressional intention to abrogate immunity from statutory text that made no mention of the government, id., at 187, 199 (White, J., dissenting), sometimes resting on clues found in legislative history, see, e.g., Hutto v. Finney, 437 U.S. 678, 694 (1978). In time, the Court began to break from this approach. See College Savings Bank v. Florida Prepaid Postsecondary Ed. Expense Bd., 527 U.S. 666, 677–678 (1999). But decades passed before the Court definitively repudiated Parden. See, e.g., Welch v. Texas Dept. of Highways and Public Transp., 483 U.S. 468, 478 (1987) (indicating that later decisions had implicitly overruled Parden, but explicitly overruling Parden for good measure); Dellmuth, 491 U. S., at 230 (rejecting the use of legislative history when assessing whether Congress abrogated sovereign immunity); College Savings Bank, 527 U. S., at 680 (overruling “[w]hatever may remain of our decision in Parden”).

Atascadero was one of the decisions issued during the course of this journey—and it does nothing to help the government’s cause. In Parden, the Court had held that a private individual could sue “a railroad owned and operated by Alabama . . . under the Federal Employers’ Liability Act.” College Savings Bank, 527 U. S., at 676. This was so “[d]espite the absence of any provision in the statute specifically referring to the States.” Ibid. Why? Because the Act applied “to ‘every’ common carrier by railroad in interstate commerce,” and Alabama’s railroad met that description. Parden, 377 U. S., at 187. When later faced with a similar statute—one that permitted suit against “ ‘any recipient of Federal assistance’ ”—the Atascadero Court rejected Parden’s reasoning, holding that this sort of “general authorization for suit in federal court is not the kind of unequivocal statutory language sufficient to abrogate” state sovereign immunity. 473 U. S., at 245–246. “When Congress chooses to subject the States to federal jurisdiction,” the Court continued, “it must do so specifically.” Id., at 246. Understood in context, then, Atascadero stands only for the now-familiar proposition that Congress must, at a minimum, mention the government when it wishes to scrap sovereign immunity and permit claims for damages. The decision does not—contrary to the government’s submission—counsel against recognizing a waiver of sovereign immunity when Congress authorizes suit against “any person” and takes the further step of expressly defining that term to include “any . . . government . . . agency.”

Employees was another case decided during the long retreat from Parden. And, on first encounter, it might seem more promising for the government. That case concerned the FLSA, which authorizes actions against “employer[s]” for unpaid overtime, and the question whether that law clearly permitted suit against state agencies. See Employees, 411 U. S., at 282. As originally drafted, the Act defined the term “employer” to exclude state agencies. Ibid. But a later amendment to the statute’s definitional section brought some state agencies within its reach. Id., at 282–283. Recognizing that “the literal language” of the Act as amended covered some state agencies, the Court nevertheless concluded that Congress had not spoken clearly enough to abrogate state sovereign immunity. See id., at 283, 285. As the government sees it, the same logic applies with equal force here. Brief for Petitioner 25–28.

By its own terms, however, Employees is distinguishable. The Employees Court stressed that, while Congress amended the definitional section of the FLSA to include States, it had not made any changes to the underlying liability provision. And, the Court reasoned, “it would be surprising” to think Congress meant to deprive a State of immunity on the basis of a change to a definitional section alone, without any accompanying change to the pertinent liability provision. 411 U. S., at 285 (internal quotation marks omitted). But what the FLSA lacked, the FCRA supplies. As we have seen, Congress did amend the FCRA’s liability provisions in 1996. In doing so, Congress replaced the narrow class of defendants originally subject to suit for money damages—consumer reporting agencies and users of the information they supply. See Part I, supra; 84Stat. 1134; §§1681n, 1681o (1970 ed.). In its stead, Congress provided that a different and much larger class of defendants—“[a]ny person,” 110Stat. 3009–446—may be sued for violating “any requirement” of the FCRA. §§1681n(a), 1681o(a). And from the statute’s start, Congress has defined the term “person” to include “any” government agency. 84Stat. 1128; see §1681a(b).

There is another problem with the government’s invocation of Employees. Despite recognizing that “the literal language of the” FLSA permitted suits against States, the Employees Court considered it all but dispositive that it could not find “a word” in the Act’s legislative history indicating that Congress wanted “to make it possible for a citizen of that State or another State to sue the State in the federal courts.” 411 U. S., at 283, 285. As should be clear by now, that is not how this Court’s contemporary sovereign- immunity doctrine works. With time, this Court has resolved that our task is to look for “a clear statement in the text of the statute.” Sossamon v. Texas, 563 U.S. 277, 290 (2011). And just as it is error to displace sovereign immunity based on inferences from legislative history without clear statutory direction (Parden), so it is error to grant sovereign immunity based on inferences from legislative history in the face of clear statutory direction waiving that immunity (Employees). The government itself has elsewhere recognized that such notions are “relic[s] from a ‘bygone era of statutory construction.’ ” Food Marketing Institute, 588 U. S., at 437 (quoting Brief for United States as Amicus Curiae in Food Marketing Institute, O. T. 2018, No. 18–481, p. 19).

In saying this much, we do not wash our hands of Employees. No one before us questions that the decision is entitled to stare decisis effect with respect to the portions of the FLSA it addressed. We recognize only that the Court has since repeatedly disavowed the decision’s methodological approach and cautioned against its use when considering claims of sovereign immunity in other contexts.[2]

B

In a final set of arguments, the government pursues a different theme. Now accepting the contemporary sovereign-immunity principles we have outlined, the government contends the provisions of the FCRA before us are still insufficient to abrogate immunity. Here, the government acknowledges that §§1681n and 1681o expressly authorize suits against “any person.” It acknowledges that §1681a expressly defines “person” to include “any” federal agency. But the government asks us to hold that §§1681n and 1681o do not clearly waive sovereign immunity because they do not “unambiguously incorporate” §1681a’s definition. Brief for Petitioner 28.

That is no small ask. When Congress takes the trouble to define the terms it uses, a court must respect its definitions as “virtually conclusive.” Sturgeon v. Frost, 587 U.S. 28, 56 (2019) (internal quotation marks omitted). This Court will not deviate from an express statutory definition merely because it “varies from [the] term’s ordinary meaning.” Digital Realty Trust, Inc. v. Somers, 583 U.S. 149, 160 (2018). Nor will we disregard a statutory definition simply because the question before us happens to involve sovereign immunity. See Seminole Tribe, 517 U. S., at 57, n. 9. Rather, this Court has said it will deviate from a statutory definition only when applying the definition would be “incompatible with Congress[’s] regulatory scheme” or would “destro[y] one of the statute’s major purposes.” Digital Realty Trust, 583 U. S., at 163–164 (internal quotation marks and alterations omitted).

The government does not even try to meet that standard in this case. How could it? The government acknowledges that federal agencies are among “ ‘the largest furnishers of credit information in the country.’ ” Brief for Petitioner 38 (quoting Robinson, 590 U. S., at ___ (opinion of Thomas, J.) (slip op., at 3)). So applying the Act’s definitional and civil liability provisions as written and allowing suits against federal agencies to proceed would, if anything, seem consistent with the Act’s goal of “ensur[ing] fair and accurate credit reporting.” Safeco Ins. Co. of America v. Burr, 551 U.S. 47, 52 (2007).

Recognizing this problem, the government suggests a different kind of “[i]ncongruit[y]” would arise if §§1681n and 1681o incorporated §1681a’s definition of “person.” Brief for Petitioner 33. The government focuses on the fact that §1681a’s definition of “person” includes not just federal agencies but state entities as well. So giving that definition effect in §§1681n and 1681o would render “not just the federal government, but also individual States” susceptible to consumer suits for money damages. Ibid. And that result, the government contends, is unthinkable. Unthinkable because Congress enacted the FCRA pursuant to the Constitution’s Commerce Clause—a provision this Court has held does not endow Congress with the power to abrogate state sovereign immunity. Id., at 34; see Florida Prepaid Postsecondary Ed. Expense Bd. v. College Savings Bank, 527 U.S. 627, 636 (1999) (“Congress may not abrogate state sovereign immunity” through “the Commerce Clause”).

While the premise of the government’s argument is correct, its conclusion is not. If the FCRA is a piece of Commerce Clause legislation, the waiver of sovereign immunity effected by §§1681n and 1681o might be constitutionally invalid as applied against individual States. But none of that means we may disregard the statute’s clear terms. See Seminole Tribe, 517 U. S., at 57, n. 9 (“We already have found the clear statement rule satisfied, and that finding renders the preference for avoiding a constitutional question inapplicable”); United States v. Oakland Cannabis Buyers’ Cooperative, 532 U.S. 483, 494 (2001) (“[T]he canon of constitutional avoidance has no application in the absence of statutory ambiguity”). Instead, we ask two distinct questions in cases involving claims of state sovereign immunity: “first, whether Congress unequivocally expressed its intent to abrogate that immunity; and second, if it did, whether Congress acted pursuant to a valid grant of constitutional authority.” Kimel, 528 U. S., at 73. Often, this Court has found, a federal statute does clearly seek to abrogate a State’s immunity but lacks constitutional authority to accomplish that objective. See, e.g., id., at 67; Seminole Tribe, 517 U. S., at 47; Allen v. Cooper, 589 U.S. 248, 255–256 (2020). Analytically, today’s case is no different. “[P]erson” means what the FCRA says it means, even if state defendants might be able to raise a valid constitutional defense to a consumer suit that the federal government cannot.

Perhaps recognizing as much, the government pivots to a discussion of the Act’s other enforcement mechanisms. Most notably, the government points to §1681q, which makes it a crime—punishable by a fine, imprisonment, or both—for “[a]ny person” to “knowingly and willfully obtai[n]” consumer information “under false pretenses.” As the government sees it, the term “person” in this provision cannot possibly bear its statutory definition because it is “absur[d]” to think Congress might have authorized criminal enforcement against federal agencies. Brief for Petitioner 31 (internal quotation marks omitted). What’s more, the government submits, because the term “person” cannot include “government” in §1681q, it cannot include “government” in §§1681n and 1681o either. Id., at 30–31.

Again, however, that much does not follow. Suppose, as the Third Circuit did when analyzing Mr. Kirtz’s claim, that “[i]t would be absurd . . . to subject the federal government to criminal prosecution.” 46 F. 4th, at 171–172. Suppose, too, that this absurdity supplies the exceptional reason necessary to deviate from §1681a’s definition of “person” in §1681q’s criminal-enforcement provision. Even spotting the government that much for argument’s sake, absurdity is not contagious: The power to correct for an absurdity “in one portion of a statute” does not imply a “license to distort other provisions of the statute.” NLRB v. Health Care & Retirement Corp. of America, 511 U.S. 571, 579 (1994). And the government offers no basis for us to think that applying §1681a’s definition to the Act’s consumer-suit provisions in §§1681n and 1681o—as opposed to its criminal provisions in §1681q—would lead to absurd results. Our obligation therefore remains “to enforce” the statutes presently before us, each “according to its terms.” Hartford Underwriters Ins. Co. v. Union Planters Bank, N. A., 530 U.S. 1, 6 (2000) (internal quotation marks omitted).

Consider the alternative. If we could ignore §1681a’s definition of “person” when it comes to §§1681n and 1681o simply because applying that definition to other statutory provisions could lead to absurd results, where would §1681a’s definition apply? Before the Seventh Circuit, the government proposed this solution: treating federal agencies as “person[s]” subject to all the Act’s “substantive requirements” but exempt from any of its liability provisions. Bormes v. United States, 759 F.3d 793, 795 (2014). That kind of “wholly artificial,” if surely convenient, distinction lacks any grounding in the statutory text, 46 F. 4th, at 166, and has no proper place in our jurisprudence, cf. Niz-Chavez v. Garland, 593 U.S. 155, 172 (2021) (“[W]ords are how the law constrains power”).

Venturing even further from the relevant statutory text, the government offers one last argument. It observes that the Privacy Act of 1974 covers some of the same ground we attribute to the FCRA. Passed “to protect the privacy of individuals identified in [federal] information systems,” 88Stat. 1896, the Privacy Act addresses the government’s retention and disclosure of personal information, see 5 U. S. C. §552a, including the disclosure of that information to consumer reporting agencies, see 31 U. S. C. §3711(e). If a federal agency supplies inaccurate information, the Privacy Act allows individuals to seek a court order requiring it to correct its records. See 5 U. S. C. §§552a(g)(1)–(2). Money damages are also sometimes available. §552a(g)(4). Because these remedies have long been available to address agency misconduct under the Privacy Act, the government reasons, there was no reason for Congress to supplement them with additional remedies under the FCRA.

That’s an unusual argument. Even the government concedes that, the Privacy Act notwithstanding, it is subject to and liable under at least some provisions of the FCRA. E.g., Brief for Petitioner 34–35 (conceding the government may be held liable under §1681u); id., at 28 (“Sometimes,” the Act’s “use of the word ‘person’ . . . refers to the default statutory definition in [§]1681a(b)”). Nor is the need to juggle multiple and sometimes overlapping legal obligations an unusual feature of contemporary American life for the government any more than it is for the governed. Recognizing this fact—and mindful our role is to apply the law, not rewrite it—we approach federal statutes touching on the same topic with a “strong presumption” they can coexist harmoniously. See Epic Systems Corp. v. Lewis, 584 U.S. 497, 510 (2018) (internal quotation marks and alterations omitted). Only by carrying a “heavy burden” can a party convince us that one statute “displaces” a second. Ibid. Where two laws are merely complementary—as is undisputedly the case here—our duty lies not in preferring one over another but in giving effect to both. Gallardo v. Marstiller, 596 U.S. 420, 432 (2022); see 46 F. 4th, at 176 (“USDA has not identified any actual inconsistency between the Privacy Act and the [FCRA]”).

*

The Executive Branch may question the wisdom of holding federal agencies accountable for their violations of the Fair Credit Reporting Act; certainly the many and resourceful arguments it advances today suggest as much. But Congress’s judgment commands our respect and the law it has adopted speaks clearly: A consumer may sue “any” federal agency for defying the law’s terms. Because it faithfully followed this legislative direction, the judgment of the Court of Appeals for the Third Circuit is

Affirmed.