No. 18-1023 No. 18-1028 No. 18-1038
(Slip Opinion) Syllabus NOTE: Where it is feasible, a syllabus (headnote) will be released, as is being done in connection with this case, at the time the opinion is issued. The syllabus constitutes no part of the opinion of the Court but has been prepared by the Reporter of Decisions for the convenience of the reader. See United States v. Detroit Timber & Lumber Co., 200 U. S. 321, 337. CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT The Patient Protection and Affordable Care Act established online exchanges where insurers could sell their healthcare plans. The now-expired "Risk Corridors" program aimed to limit the plans' profits and losses during the exchanges' first three years (2014 through 2016). See §1342, 124 Stat. 211. Section 1342 set out a formula for computing a plan's gains or losses at the end of each year, providing that eligible profitable plans "shall pay" the Secretary of the Department of Health and Human Services (HHS), while the Secretary "shall pay" eligible unprofitable plans. The Act neither appropriated funds for these yearly payments nor limited the amounts that the Government might pay. Nor was the program required to be budget neutral. Each year, the Government owed more money to unprofitable insurers than profitable insurers owed to the Government, resulting in a total deficit of more than $12 billion. And at the end of each year, the appropriations bills for the Centers for Medicare and Medicaid Services (CMS) included a rider preventing CMS from using the funds for Risk Corridors payments. Petitioners—four health-insurance companies that claim losses under the program—sued the Federal Government for damages in the Court of Federal Claims. Invoking the Tucker Act, they alleged that §1342 obligated the Government to pay the full amount of their losses as calculated by the statutory formula and sought a money judgment for the unpaid sums owed. Only one petitioner prevailed in the trial courts, and the Federal Circuit ruled for the Government in each appeal, holding that §1342 had initially created a Government obligation to pay the full amounts, but that the subsequent appropriations riders impliedly "repealed or suspended" that obligation. Held:
1. The Risk Corridors statute created a Government obligation to pay insurers the full amount set out in §1342's formula. Pp. 9-16.
(a) The Government may incur an obligation directly through statutory language, without also providing details about how the obligation must be satisfied. See United States v. Langston, 118 U. S. 389. Pp. 9-11.
(b) Section 1342 imposed a legal duty of the United States that could mature into a legal liability through the insurers' participation in the exchanges. This conclusion flows from the express terms and context of §1342, which imposed an obligation by using the mandatory term "shall." The section's mandatory nature is underscored by the adjacent provisions, which differentiate between when the HHS Secretary "shall" take certain actions and when she "may" exercise discretion. See §§1341(b)(2), 1343(b). Section 1342 neither requires the Risk Corridors program to be budget-neutral nor suggests that the Secretary's payments to unprofitable plans pivoted on profitable plans' payments to the Secretary or that a partial payment would satisfy the Government's whole obligation. It thus must be given its plain meaning: The Government "shall pay" the sum prescribed by §1342. Pp. 11-13.
(c) Contrary to the Government's contention, neither the Appropriations Clause nor the Anti-Deficiency Act addresses whether Congress itself can create or incur an obligation directly by statute. Nor does §1342's obligation-creating language turn on whether Congress expressly provided budget authority before appropriating funds. The Government's arguments also conflict with well-settled principles of statutory interpretation. That §1342 contains no language limiting the obligation to the availability of appropriations, while Congress expressly used such limiting language in other Affordable Care Act provisions, indicates that Congress intended a different meaning in §1342. Pp. 13-16.
2. Congress did not impliedly repeal the obligation through its appropriations riders. Pp. 16-23.
(a) Because "'repeals by implication are not favored,'" Morton v. Mancari, 417 U. S. 535, 549, this Court will regard each of two statutes effective unless Congress' intention to repeal is "'clear and manifest,'" or the laws are "irreconcilable," id., at 550-551. In the appropriations
context, this requires the Government to show "something more than the mere omission to appropriate a sufficient sum." United States v. Vulte, 233 U. S. 509, 515. As Langston and Vulte confirm, the appropriations riders here did not manifestly repeal or discharge the Government's uncapped obligation, see Langston, 118 U. S., at 394, and do not indicate "any other purpose than the disbursement of a sum of money for the particular fiscal years," Vulte, 233 U. S., at 514. Nor is there any indication that HHS and CMS thought that the riders clearly expressed an intent to repeal. Pp. 16-19.
(b) Appropriations measures have been found irreconcilable with statutory obligations to pay, but the riders here did not use the kind of "shall not take effect" language decisive in United States v. Will, 449 U. S. 200, 222-223, or purport to "suspen[d]" §1342 prospectively or to foreclose funds from "any other Act" "notwithstanding" §1342's money-mandating text, United States v. Dickerson, 310 U. S. 554, 556-557. They also did not reference §1342's payment formula, let alone "irreconcilabl[y]" change it, United States v. Mitchell, 109 U. S. 146, 150, or provide that payments from profitable plans would be "'in full compensation'" of the Government's obligation to unprofitable plans, United States v. Fisher, 109 U. S. 143, 150. Pp. 19-21.
(c) The legislative history cited by the Federal Circuit is also unpersuasive. Pp. 22-23.
3. Petitioners properly relied on the Tucker Act to sue for damages in the Court of Federal Claims. Pp. 23-30.
(a) The United States has waived its immunity for certain damages suits in the Court of Federal Claims through the Tucker Act. Because that Act does not create "substantive rights," United States v. Navajo Nation, 556 U. S. 287, 290, a plaintiff must premise her damages action on "other sources of law," like "statutes or contracts," ibid., provided those statutes "'can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained,'" United States v. White Mountain Apache Tribe, 537 U. S. 465, 472. The Act does, however, yield when the obligation-creating statute provides its own detailed remedies or when the Administrative Procedure Act provides an avenue for relief. Pp. 23-26.
(b) Petitioners clear each hurdle: The Risk Corridors statute is fairly interpreted as mandating compensation for damages, and neither exception to the Tucker Act applies. Section 1342's mandatory "'shall pay' language" falls comfortably within the class of statutes that permit recovery of money damages in the Court of Federal Claims. This finding is bolstered by §1342's focus on compensating insurers for past conduct. And there is no separate remedial scheme supplanting the Court of Federal Claims' power to adjudicate petitioners' claims.
See United States v. Bormes, 568 U. S. 6, 12. Nor does the Administrative Procedure Act bar petitioners' Tucker Act suit. In contrast to Bowen v. Massachusetts, 487 U. S. 879, a Medicaid case where the State sued the HHS Secretary under the Administrative Procedure Act in district court, petitioners here seek not prospective, nonmonetary relief to clarify future obligations but specific sums already calculated, past due, and designed to compensate for completed labors. The Risk Corridors statute and Tucker Act allow them that remedy. And because the Risk Corridors program expired years ago, this litigation presents no special concern, as Bowen did, about managing a complex ongoing relationship or tracking ever-changing accounting sheets. Pp. 26-30.No. 18-1023 and No. 18-1028 (second judgment), 729 Fed. Appx. 939; No. 18-1028 (first judgment), 892 F. 3d 1311; No. 18-1038, 892 F. 3d 1184, reversed and remanded.
SOTOMAYOR, J., delivered the opinion of the Court, in which ROBERTS, C. J., and GINSBURG, BREYER, KAGAN, and KAVANAUGH, JJ., joined, and in which THOMAS and GORSUCH, JJ., joined as to all but Part III-C. ALITO, J., filed a dissenting opinion. Opinion of the Court NOTICE: This opinion is subject to formal revision before publication in the preliminary print of the United States Reports. Readers are requested to notify the Reporter of Decisions, Supreme Court of the United States, Washington, D. C. 20543, of any typographical or other formal errors, in order that corrections may be made before the preliminary print goes to press. ON WRITS OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT JUSTICE SOTOMAYOR delivered the opinion of the Court.
JUSTICE THOMAS and JUSTICE GORSUCH join all but Part III-C of this opinion.
The Patient Protection and Affordable Care Act expanded healthcare coverage to many who did not have or could not afford it. The Affordable Care Act did this by, among other things, providing tax credits to help people buy insurance and establishing online marketplaces where insurers could sell plans. To encourage insurers to enter those marketplaces, the Act created several programs to defray the carriers' costs and cabin their risks.
Among these initiatives was the "Risk Corridors" program, a temporary framework meant to compensate insurers for unexpectedly unprofitable plans during the marketplaces' first three years. The since-expired Risk Corridors statute, §1342, set a formula for calculating payments under the program: If an insurance plan loses a certain amount of money, the Federal Government "shall pay" the plan; if the plan makes a certain amount of money, the plan "shall pay" the Government. See §1342, 124 Stat. 211-212 (codified at 42 U. S. C. §18062). Some plans made money and paid the Government. Many suffered losses and sought reimbursement. The Government, however, did not pay.
These cases are about whether petitioners—insurers who claim losses under the Risk Corridors program—have a right to payment under §1342 and a damages remedy for the unpaid amounts. We hold that they do. We conclude that §1342 of the Affordable Care Act established a money-mandating obligation, that Congress did not repeal this obligation, and that petitioners may sue the Government for damages in the Court of Federal Claims.
In 2010, Congress passed the Patient Protection and Affordable Care Act, 124 Stat. 119, seeking to improve national health-insurance markets and extend coverage to millions of people without adequate (or any) health insurance. To that end, the Affordable Care Act called for the creation of virtual health-insurance markets, or "Health Benefit Exchanges," in each State. 42 U. S. C. §18031(b)(1). Individuals may buy health-insurance plans directly on an exchange and, depending on their household income, receive tax credits for doing so. 26 U. S. C. §36B; 42 U. S. C. §§18081, 18082. Once an insurer puts a plan on an exchange, it must "accept every employer and individual in the State that applies for such coverage," 42 U. S. C. §300gg-1(a), and may not tether premiums to a particular applicant's health, §300gg(a). In other words, the Act "ensure[s] that anyone can buy insurance." King v. Burwell, 576 U. S. 473, 493 (2015).
Insurance carriers had many reasons to participate in these new exchanges. Through the Affordable Care Act, they gained access to millions of new customers with tax credits worth "billions of dollars in spending each year." Id., at 485. But the exchanges posed some business risks, too—including a lack of "reliable data to estimate the cost of providing care for the expanded pool of individuals seeking coverage." 892 F. 3d 1311, 1314 (CA Fed. 2018) (case below in No. 18-1028).
The others were the "Reinsurance" and "Risk Adjustment" programs. The former ran from 2014 to 2016 and required insurers to pay premiums into a pool that compensated carriers covering "high risk individuals." §1341, 124 Stat. 208, 42 U. S. C. §18061. The latter is still in effect and annually transfers funds from insurance plans with relatively low-risk enrollees to plans with higher risk enrollees. See §1343, 124 Stat. 212, 42 U. S. C. §18063.
The Risk Corridors program aimed to limit participating plans' profits and losses for the exchanges' first three years (2014, 2015, and 2016). See §1342, 124 Stat. 211, 42 U. S. C. §18062. It did so through a formula that computed a plan's gains or losses at the end of each year. Plans with profits above a certain threshold would pay the Government, while plans with losses below that threshold would receive payments from the Government. §1342(b), 124 Stat. 211. Specifically, §1342 stated that the eligible profitable plans "shall pay" the Secretary of the Department of Health and Human Services (HHS), while the Secretary "shall pay" the eligible unprofitable plans. Ibid.
If a health insurance plan made (or lost) up to 3 percentage points more than expected in a plan year, the plan would keep the gains (or losses). If the plan made (or lost) between 3 and 8 percentage points more than predicted, it would give up half of the earnings (or would be compensated for half of the shortfalls) exceeding the 3 percentage-point threshold. If the gains (or losses) exceeded predictions by eight percentage points, the insurers would pay (or receive) 80 percent of the gains (or losses) exceeding the 8 percentage-point mark. See §1342(b), 124 Stat. 211, 42 U. S. C. §18062(b).
When it enacted the Affordable Care Act in 2010, Congress did not simultaneously appropriate funds for the yearly payments the Secretary could potentially owe under the Risk Corridors program. Neither did Congress limit the amounts that the Government might pay under §1342. Nor did the Congressional Budget Office (CBO) "score"—that is, calculate the budgetary impact of—the Risk Corridors program.
In later years, the CBO noted that the Risk Corridors statute did not require the program to be budget neutral. The CBO reported that, "[i]n contrast" to the Act's other risk-mitigation programs, "risk corridor collections (which will be recorded as revenues) will not necessarily equal risk corridor payments, so that program can have net effects on the budget deficit." CBO, The Budget and Economic Outlook: 2014 to 2024, p. 59 (2014). The CBO thus recognized that "[i]f insurers' costs exceed their expectations, on average, the risk corridor program will impose costs on the federal budget." Id., at 110.
Like the CBO, the federal agencies charged with implementing the program agreed that §1342 did not require budget neutrality. Nine months before the program started, HHS acknowledged that the Risk Corridors program was "not statutorily required to be budget neutral." 78 Fed. Reg. 15473 (2013). HHS assured, however, that "[r]egardless of the balance of payments and receipts, HHS will remit payments as required under Section 1342 of the Affordable Care Act." Ibid.
Similar guidance came from the Centers for Medicare and Medicaid Services (CMS), the agency tasked with helping the HHS Secretary collect and remit program payments. CMS confirmed that a lack of payments from profitable plans would not relieve the Government from making its payments to the unprofitable ones. See 79 Fed. Reg. 30260 (2014). Citing "concerns that risk corridors collections may not be sufficient to fully fund risk corridors payments" to the unprofitable plans, CMS declared that "[i]n the unlikely event of a shortfall . . . HHS recognizes that the Affordable Care Act requires the Secretary to make full payments to issuers." Ibid.
The program's first year, 2014, tallied a deficit of about $2.5 billion. Profitable plans owed the Government $362 million, while the Government owed unprofitable plans $2.87 billion. See CMS, Risk Corridors Payment Proration Rate for 2014 (2015).
At the end of the first year, Congress enacted a bill appropriating a lump sum for CMS' Program Management. See Pub. L. 113-235, Div. G, Tit. II, 128 Stat. 2130-2131 (providing for the fiscal year ending September 30, 2015). The bill included a rider restricting the appropriation's effect on Risk Corridors payments out to issuers:
"None of the funds made available by this Act . . . or transferred from other accounts funded by this Act to the 'Centers for Medicare and Medicaid Services—
Program Management' account, may be used for payments under section 1342(b)(1) of Public Law 111-148 (relating to risk corridors)." §227, id., at 2491.
The program's second year resembled its first. In February 2015, HHS repeated its belief that "risk corridors collections w[ould] be sufficient to pay for all" of the Government's "risk corridors payments." 80 Fed. Reg. 10779 (2015). The agency again "recognize[d] that the Affordable Care Act requires the Secretary to make full payments to issuers." Ibid. "In the unlikely event that risk corridors collections" were "insufficient to make risk corridors payments," HHS reassured, the Government would "use other sources of funding for the risk corridors payments, subject to the availability of appropriations." Ibid.
The 2015 program year also ran a deficit, this time worth about $5.5 billion. See CMS, Risk Corridors Payment and Charge Amounts for the 2015 Benefit Year (2016). Facing a second shortfall, CMS continued to "recogniz[e] that the Affordable Care Act requires the Secretary to make full payments to issuers." CMS, Risk Corridors Payments for 2015, p. 1 (2016). CMS also confirmed that "HHS w[ould] record risk corridors payments due as an obligation of the United States Government for which full payment is required." Ibid. And at the close of the second year, Congress enacted another appropriations bill with the same rider as before. See Pub. L. 114-113, §225, 129 Stat. 2624 (providing for the fiscal year ending September 30, 2016).
The program's final year, 2016, was similar. The Government owed unprofitable insurers about $3.95 billion more than profitable insurers owed the Government. See CMS, Risk Corridors Payment and Charge Amounts for the 2016 Benefit Year (2017). And Congress passed an appropriations bill with the same rider. See Pub. L. 115-31, §223, 131 Stat. 543 (providing for the fiscal year ending September 30, 2017).
All told, the Risk Corridors program's deficit exceeded $12 billion.
The dispute here is whether the Government must pay the remaining deficit. Petitioners in these consolidated cases are four health-insurance companies that participated in the healthcare exchanges: Maine Community Health Options, Blue Cross and Blue Shield of North Carolina, Land of Lincoln Mutual Health Insurance Company, and Moda Health Plan, Inc. They assert that their plans were unprofitable during the Risk Corridors program's 3-year term and that, under §1342, the HHS Secretary still owes them hundreds of millions of dollars.
These insurers sued the Federal Government for damages in the United States Court of Federal Claims, invoking the Tucker Act, 28 U. S. C. §1491. They alleged that §1342 of the Affordable Care Act obligated the Government to pay the full amount of their losses as calculated by the statutory formula and sought a money judgment for the unpaid sums owed—a claim that, if successful, could be satisfied through the Judgment Fund. These lawsuits saw mixed results in the trial courts. Petitioner Moda prevailed; the others did not.
For a meritorious claim brought within the Tucker Act's 6-year statute of limitations, 28 U. S. C. §2501, federal law generally requires that the "final judgment rendered by the United States Court of Federal Claims against the United States . . . be paid out of any general appropriation therefor." §2517(a). The Judgment Fund is a permanent and indefinite appropriation for "[n]ecessary amounts . . . to pay final judgments, awards, compromise settlements, and interest and costs specified in the judgments or otherwise authorized by law when . . . payment is not otherwise provided for." 31 U. S. C. §1304(a)(1).
Compare 130 Fed. Cl. 436 (2017) (granting Moda Health Plan partial summary judgment on its statutory and implied-in-fact-contract claims), with 129 Fed. Cl. 81 (2016) (dismissing Land of Lincoln's statutory, contract, and Takings Clause claims), 131 Fed. Cl. 457 (2017) (dismissing Blue Cross Blue Shield's statutory and contract claims), and 133 Fed. Cl. 1 (2017) (dismissing Maine Community Health's statutory claims).
A divided panel of the United States Court of Appeals for the Federal Circuit ruled for the Government in each appeal. See 892 F. 3d 1311; 892 F. 3d 1184 (2018); 729 Fed. Appx. 939 (2018). As relevant here, the Federal Circuit concluded that §1342 had initially created a Government obligation to pay the full amounts that petitioners sought under the statutory formula. See 892 F. 3d, at 1320-1322. The court also recognized that "it has long been the law that the government may incur a debt independent of an appropriation to satisfy that debt, at least in certain circumstances." Id., at 1321.
Even so, the court held that Congress' appropriations riders impliedly "repealed or suspended" the Government's obligation. Id., at 1322. Although the panel acknowledged that "[r]epeals by implication are generally disfavored"—especially when the "alleged repeal occurred in an appropriations bill"—it found that the riders here "adequately expressed Congress's intent to suspend" the Government's payments to unprofitable plans "beyond the sum of payments" it collected from profitable plans. Id., at 1322-1323, 1325.
Judge Newman dissented, observing that the Government had not identified any "statement of abrogation or amendment of the statute," nor any "disclaimer" of the Government's "statutory and contractual commitments." Id., at 1335. The dissent also reasoned that precedent undermined the court's conclusion and that the appropriations riders could not apply retroactively because the Government had used the Risk Corridors program to induce insurers to enter the exchanges. Id., at 1336-1339. Emphasizing the importance of Government credibility in public-private enterprise, the dissent warned that the majority's decision would "undermin[e] the reliability of dealings with the government." Id., at 1340.
A majority of the Federal Circuit declined to revisit the court's decision en banc, 908 F. 3d 738 (2018) (per curiam); see also id., at 740 (Newman, J., dissenting); id., at 741 (Wallach, J., dissenting), and we granted certiorari, 588 U. S. ___ (2019).
These cases present three questions: First, did §1342 of the Affordable Care Act obligate the Government to pay participating insurers the full amount calculated by that statute? Second, did the obligation survive Congress' appropriations riders? And third, may petitioners sue the Government under the Tucker Act to recover on that obligation? Because our answer to each is yes, we reverse.
The Risk Corridors statute created a Government obligation to pay insurers the full amount set out in §1342's formula.
An "obligation" is a "definite commitment that creates a legal liability of the government for the payment of goods and services ordered or received, or a legal duty . . . that could mature into a legal liability by virtue of actions on the part of the other party beyond the control of the United States." GAO, A Glossary of Terms Used in the Federal Budget Process 70 (GAO-05-734SP, 2005). The Government may incur an obligation by contract or by statute. See ibid.
Incurring an obligation, of course, is different from paying one. After all, the Constitution's Appropriations Clause provides that "No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law." Art. I, §9, cl. 7; see also GAO, Principles of Federal Appropriations Law 2-3 (4th ed. 2016) (hereinafter GAO Redbook) ("[T]he authority to incur obligations by itself is not sufficient to authorize payments from the Treasury"). Creating and satisfying a Government obligation, therefore, typically involves four steps: (1) Congress passes an organic statute (like the Affordable Care Act) that creates a program, agency, or function; (2) Congress passes an Act authorizing appropriations; (3) Congress enacts the appropriation, granting "budget authority" to incur obligations and make payments, and designating the funds to be drawn; and (4) the relevant Government entity begins incurring the obligation. See id., at 2-56; see also Op. Comp. Gen., B-193573 (Dec. 19, 1979).
But Congress can deviate from this pattern. It may, for instance, authorize agencies to enter into contracts and "incur obligations in advance of appropriations." GAO Redbook 2-4. In that context, the contracts "constitute obligations binding on the United States," such that a "failure or refusal by Congress to make the necessary appropriation would not defeat the obligation, and the party entitled to payment would most likely be able to recover in a lawsuit." Id., at 2-5; see also, e.g., Cherokee Nation of Okla. v. Leavitt, 543 U. S. 631, 636-638 (2005) (rejecting the Government's argument that it is legally bound by its contractual promise to pay "if, and only if, Congress appropriated sufficient funds"); Salazar v. Ramah Navajo Chapter, 567 U. S. 182, 191 (2012) ("Although the agency itself cannot disburse funds beyond those appropriated to it, the Government's 'valid obligations will remain enforceable in the courts'" (quoting 2 GAO Redbook 6-17 (2d ed. 1992)).
Congress can also create an obligation directly by statute, without also providing details about how it must be satisfied. Consider, for example, United States v. Langston, 118 U. S. 389 (1886). In that case, Congress had enacted a statute fixing an official's annual salary at "$7,500 from the date of the creation of his office." Id., at 394. Years later, however, Congress failed to appropriate enough funds to pay the full amount, prompting the officer to sue for the remainder. Id., at 393. Understanding that Congress had created the obligation by statute, this Court held that a subsequent failure to appropriate enough funds neither "abrogated [n]or suspended" the Government's pre-existing commitment to pay. Id., at 394. The Court thus affirmed judgment for the officer for the balance owed. Ibid.
The Government suggests that Langston is irrelevant because that case predates the Judgment Fund, cf. n. 3, supra, meaning that the Court "had no occasion" to determine whether the statute at issue "authorized a money-damages remedy" against the Government, Brief for United States 30. But by affirming a judgment against the United States, Langston necessarily confirmed the Government's obligation to pay independent of a specific appropriation. What remedies ensure that the Government makes good on its duty to pay is a separate question that we take up below. See Part IV, infra.
The GAO shares this view. As the Redbook explains, if Congress created an obligation by statute without detailing how it will be paid, "an agency could presumably meet a funding shortfall by such measures as making prorated payments." GAO Redbook 2-36, n. 39. But "such actions would be only temporary pending receipt of sufficient funds to honor the underlying obligation" and "[t]he recipient would remain legally entitled to the balance." Ibid. Thus, the GAO warns, although a "failure to appropriate" funds "will prevent administrative agencies from making payment," that failure "is unlikely to prevent recovery by way of a lawsuit." Id., at 2-63 (citing, e.g., Langston, 118 U. S., at 394).
Put succinctly, Congress can create an obligation directly through statutory language.
Section 1342 imposed a legal duty of the United States that could mature into a legal liability through the insurers' actions—namely, their participating in the healthcare exchanges.
This conclusion flows from §1342's express terms and context. See, e.g., Merit Management Group, LP v. FTI Consulting, Inc., 583 U. S. ___, ___ (2018) (slip op., at 11) (statutory interpretation "begins with the text"). The first sign that the statute imposed an obligation is its mandatory language: "shall." "Unlike the word 'may,' which implies discretion, the word 'shall' usually connotes a requirement." Kingdomware Technologies, Inc. v. United States, 579 U. S. ___, ___ (2016) (slip op., at 9); see also Lexecon Inc. v. Milberg Weiss Bershad Hynes & Lerach, 523 U. S. 26, 35 (1998) (observing that "'shall'" typically "creates an obligation impervious to . . . discretion"). Section 1342 uses the command three times: The HHS Secretary "shall establish and administer" the Risk Corridors program from 2014 to 2016, "shall provide" for payments according to a precise statutory formula, and "shall pay" insurers for losses exceeding the statutory threshold. §§1342(a), (b)(1), 114 Stat. 211, 42 U. S. C. §§18062(a), (b)(1).
Section 1342's adjacent provisions also underscore its mandatory nature. In §1341 (a reinsurance program) and §1343 (a risk-adjustment program), the Affordable Care Act differentiates between when the HHS Secretary "shall" take certain actions and when she "may" exercise discretion. See §1341(b)(2), 124 Stat. 209, 42 U. S. C. §18061(b)(2) ("[T]he Secretary . . . shall include" a formula that "may be designed" in multiple ways); §1343(b), 124 Stat. 212, 42 U. S. C. §18063(b) ("The Secretary . . . shall establish" and "may utilize" certain criteria). Yet Congress chose mandatory terms for §1342. "When," as is the case here, Congress "distinguishes between 'may' and 'shall,' it is generally clear that 'shall' imposes a mandatory duty." Kingdomware, 579 U. S., at ___ (slip op., at 9).
Nothing in §1342 requires the Risk Corridors program to be budget neutral, either. Nor does the text suggest that the Secretary's payments to unprofitable plans pivoted on profitable plans' payments to the Secretary, or that a par- tial payment would satisfy the Government's whole obligation. Thus, without "any indication" that §1342 allows the Government to lessen its obligation, we must "give effect to [Section 1342's] plain command." Lexecon, 523 U. S., at 35. That is, the statute meant what it said: The Government "shall pay" the sum that §1342 prescribes.
Our conclusion matches the interpretations that HHS and CMS have repeated since before the Risk Corridors program began. In the agencies' view, the Risk Corridors program was "not statutorily required to be budget neutral" and instead required HHS to "remit payments" "[r]egardless of the balance of payments and receipts." 78 Fed. Reg. 15473 (HHS regulation); accord, 79 Fed. Reg. 30260 (CMS regulation noting that even "[i]n the unlikely event of a shortfall for the 2015 program year, . . . the Affordable Care Act requires the Secretary to make full payments to issuers").
The Government does not contest that §1342's plain terms appeared to create an obligation to pay whatever amount the statutory formula provides. It insists instead that the Appropriations Clause, Art. I, §9, cl. 7, and the Anti-Deficiency Act, 31 U. S. C. §1341, "qualified" that obligation by making "HHS's payments contingent on appropriations by Congress." Brief for United States 20. "Because Congress did not appropriate funds beyond the amounts collected" from profitable plans, this argument goes, "HHS's statutory duty [to pay unprofitable plans] extended only to disbursing those collected amounts." Id., at 24-25.
That does not follow. Neither the Appropriations Clause nor the Anti-Deficiency Act addresses whether Congress itself can create or incur an obligation directly by statute. Rather, both provisions constrain how federal employees and officers may make or authorize payments without appropriations. See U. S. Const., Art. I, §9, cl. 7 (requiring an "Appropriatio[n] made by Law" before money may "be drawn" to satisfy a payment obligation); 31 U. S. C. §1341(a)(1)(A) ("[A]n officer or employee of the United States Government . . . may not . . . make or authorize an expenditure or obligation exceeding an amount available in an appropriation or fund for the expenditure or obligation"). As we have explained, "'[a]n appropriation per se merely imposes limitations upon the Government's own agents,'" but "'its insufficiency does not pay the Government's debts, nor cancel its obligations.'" Ramah, 567 U. S., at 197 (quoting Ferris v. United States, 27 Ct. Cl. 542, 546 (1892)). If anything, the Anti-Deficiency Act confirms that Congress can create obligations without contemporaneous funding sources: That Act's prohibitions give way "as specified" or "authorized" by "any other provision of law." 31 U. S. C. §1341(a)(1). Here, the Government's obligation was authorized by the Risk Corridors statute.
And contrary to the Government's view, §1342's obligation-creating language does not turn on whether Congress expressly provided "budget authority" before appropriating funds. Budget authority is an agency's power "provided by Federal law to incur financial obligations," 88 Stat. 297, 2 U. S. C. §622(2)(A), "that will result in immediate or future outlays of government funds," GAO Redbook 2-1; see also id., at 2-55 ("Agencies may incur obligations only after Congress grants budget authority"); GAO, A Glossary of Terms Used in the Federal Budget Process, at 20-21. As explained above, Congress usually gives budget authority through an appropriations Act or by expressly granting an agency authority to contract for the Government. See GAO Redbook 2-1 to 2-5. But budget authority is not necessary for Congress itself to create an obligation by statute. See Langston, 118 U. S., at 394; cf. Raines v. Byrd, 521 U. S. 811, 815 (1997) (treating legal obligations of the Government as distinct from budget authority).
The Government's arguments also conflict with well-settled principles of statutory interpretation. At bottom, the Government contends that the existence and extent of its obligation here is "subject to the availability of appropriations." Brief for United States 41. But that language appears nowhere in §1342, even though Congress could have expressly limited an obligation to available appropriations or specific dollar amounts. Indeed, Congress did so explicitly in other provisions of the Affordable Care Act.
See, e.g., 42 U. S. C. §280k(a) ("The Secretary . . . shall, subject to the availability of appropriations, establish a 5-year national, public education campaign"); §293k(c) ("Fifteen percent of the amount appropriated . . . in each . . . fiscal year shall be allocated to [certain] physician assistant training programs"); §293k-1(e) ("There is authorized to be appropriated to carry out this section, $10,000,000"); §293k-2(e) (payments "made to an entity from an award of a grant or contract under [§293k-2(a)] shall be . . . subject to the availability of appropriations for the fiscal year involved to make the payments"); §300hh-31(a) ("Subject to the availability of appropriations, the Secretary . . . shall establish [an epidemiology-laboratory program] to award grants"); note following §1396a ("In no case may . . . the aggregate amount of payments made by the Secretary to eligible States under this section exceed $75,000,000"); §1397m-1(b)(2)(A) ("Subject to the availability of appropriations . . . the amount paid to a State for a fiscal year under [an adult protective services program] shall equal . . . ").
This kind of limiting language is not unique to the Affordable Care Act. When Congress has restricted "shall pay" language to an appropriation or available funds, it has done so expressly. See, e.g., 2 U. S. C. §2064; 5 U. S. C. §8334; 7 U. S. C. §§2013, 2031, 3243, 6523, 7717; 10 U. S. C. §§1175, 1413a, 1598, 2031, 2410j, 2774, 9780; 12 U. S. C. §3337; 15 U. S. C. §4723; 16 U. S. C. §§45f, 410aa1, 426n, 459e1, 460m16, 698f, 1852; 20 U. S. C. §§80q5, 1070a, 1134b, 1161g; 22 U. S. C. §2906; 25 U. S. C. §1912; 30 U. S. C. §1314; 32 U. S. C. §716; 34 U. S. C. §12573; 38 U. S. C. §5317A; 42 U. S. C. §§303, 624, 655, 677, 1203, 1353, 1396b, 8623, 12622, 16014, 16512; 46 U. S. C. §§51504, 53106, 53206; 47 U. S. C. §395; 49 U. S. C. §5312; 50 U. S. C. §§4236, 4237; 52 U. S. C. §21061.
Congress has also been explicit when it has capped payments, often setting a dollar amount or designating a specific fund from which the Government shall pay. See, e.g., 5 U. S. C. §§8102a, 8134, 8461; 7 U. S. C. §§26, 6523; 10 U. S. C. §1413a; 16 U. S. C. §§450e1, 460kk; 19 U. S. C. §2296; 20 U. S. C. §§1070g1, 1078, 3988, 5607; 22 U. S. C. §3681; 30 U. S. C. §1240a; 31 U. S. C. §3343; 38 U. S. C. §1542; 42 U. S. C. §§290bb38, 295h, 618, 5318a, 15093; 43 U. S. C. §§1356a, 1619; 46 U. S. C. §53106; 50 U. S. C. §4114.
These common limitations—and our discussion below, see Part IV, infra—diminish the dissent's concern that other statutes may support a damages action in the Court of Federal Claims. Post, at 3 (opinion of ALITO, J.).
This Court generally presumes that "'when Congress includes particular language in one section of a statute but omits it in another,'" Congress "'intended a difference in meaning.'" Digital Realty Trust, Inc. v. Somers, 583 U. S. ___, ___ (2018) (slip op., at 10) (quoting Loughrin v. United States, 573 U. S. 351, 358 (2014) (alterations omitted)). The Court likewise hesitates "'to adopt an interpretation of a congressional enactment which renders superfluous another portion of that same law.'" Republic of Sudan v. Harrison, 587 U. S. ___, ___ (2019) (slip op., at 10) (quoting Mackey v. Lanier Collection Agency & Service, Inc., 486 U. S. 825, 837 (1988)). The "subject to appropriations" and payment-capping language in other sections of the Affordable Care Act would be meaningless had §1342 simultaneously achieved the same end with silence.
In sum, the plain terms of the Risk Corridors provision created an obligation neither contingent on nor limited by the availability of appropriations or other funds.
The next question is whether Congress impliedly repealed the obligation through its appropriations riders. It did not.
Because Congress did not expressly repeal §1342, the Government seeks to show that Congress impliedly did so. But "repeals by implication are not favored," Morton v. Mancari, 417 U. S. 535, 549 (1974) (internal quotation marks omitted), and are a "rarity," J. E. M. Ag Supply, Inc. v. Pioneer Hi-Bred Int'l, Inc., 534 U. S. 124, 142 (2001) (in- ternal quotation marks omitted). Presented with two statutes, the Court will "regard each as effective"—unless Congress' intention to repeal is "'"clear and manifest,"'" or the two laws are "irreconcilable." Morton, 417 U. S., at 550-551 (quoting United States v. Borden Co., 308 U. S. 188, 198 (1939)); see also FCC v. NextWave Personal Communications Inc., 537 U. S. 293, 304 (2003) ("[W]hen two statutes are capable of co-existence, it is the duty of the courts, absent a clearly expressed congressional intention to the contrary, to regard each as effective" (internal quotation marks omitted)).
This Court's aversion to implied repeals is "especially" strong "in the appropriations context." Robertson v. Seattle Audubon Soc., 503 U. S. 429, 440 (1992); see also New York Airways, Inc. v. United States, 177 Ct. Cl. 800, 810, 369 F. 2d 743, 748 (1966). The Government must point to "something more than the mere omission to appropriate a sufficient sum." United States v. Vulte, 233 U. S. 509, 515 (1914); accord, GAO Redbook 2-63 ("The mere failure to appropriate sufficient funds is not enough"). The question, then, is whether the appropriations riders manifestly repealed or discharged the Government's uncapped obligation.
Langston confirms that the appropriations riders did neither. Recall that in Langston, Congress had established a statutory obligation to pay a salary of $7,500, yet later appropriated a lesser amount. 118 U. S., at 393-394. This Court held that Congress did not "abrogat[e] or suspen[d]" the salary-fixing statute by "subsequent enactments [that] merely appropriated a less amount" than necessary to pay, because the appropriations bill lacked "words that expressly or by clear implication modified or repealed the previous law." Id., at 394.
Vulte reaffirmed that a mere failure to appropriate does not repeal or discharge an obligation to pay. At issue there was whether certain appropriations Acts had repealed a Government obligation to pay bonuses to military servicemen. 233 U. S., at 511-512. A 1902 statute had provided a 10 percent bonus to officers serving outside the contiguous United States, but in 1906 and 1907, Congress enacted appropriations funding the bonuses for officers "except [those in] P[ue]rto Rico and Hawaii." Id., at 512. Then, in 1908, Congress enacted a statute stating "'[t]hat the increase of pay . . . shall be as now provided by law.'" Id., at 513. When Lieutenant Nelson Vulte sought a bonus for his service in Puerto Rico from 1908 to 1909, the Government refused, contending that the appropriations Acts had impliedly repealed its obligation altogether.
Relying on Langston, Vulte rejected that argument. "[I]t is to be remembered," the Court wrote, that the alleged repeals "were in appropriation acts and no words were used to indicate any other purpose than the disbursement of a sum of money for the particular fiscal years." 233 U. S., at 514. At most, the appropriations had "temporarily suspend[ed]" payments, but they did not use "'the most clear and positive terms'" required to "modif[y] or repea[l]" the Government's obligation itself. Id., at 514-515 (quoting Minis v. United States, 15 Pet. 423, 445 (1841)). Because the Government had failed to show that repeal was the only '"reasonable interpretation'" of the appropriation Acts, the obligation persisted. 233 U. S., at 515 (quoting Minis, 15 Pet., at 445).
The parallels among Langston, Vulte, and these cases are clear. Here, like in Langston and Vulte, Congress "merely appropriated a less amount" than that required to satisfy the Government's obligation, without "expressly or by clear implication modif[ying]" it. Langston, 118 U. S., at 394; see also Vulte, 233 U. S., at 515. The riders stated that "[n]one of the funds made available by this Act," as opposed to any other sources of funds, "may be used for payments under" the Risk Corridors statute. §227, 128 Stat. 2491; accord, §225, 129 Stat. 2624; §223, 131 Stat. 543. But "no words were used to indicate any other purpose than the disbursement of a sum of money for the particular fiscal years." Vulte, 233 U. S., at 514. And especially because the Government had already begun incurring the prior year's obligation each time Congress enacted a rider, reasonable (and nonrepealing) interpretations exist. Indeed, finding a repeal in these circumstances would raise serious questions whether the appropriations riders retroactively impaired insurers' rights to payment. See Landgraf v. USI Film Products, 511 U. S. 244, 265-266, 280 (1994); see also GAO Redbook 1-61 to 1-62.
The relevant agencies' responses to the riders also undermine the case for an implied repeal here. Had Congress "clearly expressed" its intent to repeal, one might have expected HHS or CMS to signal the sea change. Morton, 417 U. S., at 551. But even after Congress enacted the first rider, the agencies reiterated that "the Affordable Care Act requires the Secretary to make full payments to issuers," 80 Fed. Reg. 10779, and that "HHS w[ould] record risk corridors payments due as an obligation of the United States Government for which full payment is required," CMS, Risk Corridors Payments for 2015, at 1. They understood that profitable insurers' payments to the Government would not dispel the Secretary's obligation to pay unprofitable insurers, even "in the event of a shortfall." Ibid.
Given the Court's potent presumption in the appropriations context, an implied-repeal-by-rider must be made of sterner stuff.
To be sure, this Court's implied-repeal precedents reveal two situations where the Court has deemed appropriations measures irreconcilable with statutory obligations to pay. But neither one applies here.
The first line of cases involved appropriations bills that, without expressly invoking words of "repeal," reached that outcome by completely revoking or suspending the underlying obligation before the Government began incurring it. See United States v. Will, 449 U. S. 200 (1980); United States v. Dickerson, 310 U. S. 554 (1940). Will concluded that Congress had canceled an obligation to pay cost-of-living raises through appropriations bills that bluntly stated that future raises "'shall not take effect'" or that restricted funds from "'this Act or any other Act.'" 449 U. S., at 206-207, 223. Likewise, Dickerson held that a series of appropriations bills repealed an obligation to pay military-reenlistment bonuses due in particular fiscal years. See 310 U. S., at 561. One enactment "'hereby suspended'" the bonuses before they took effect, and another "continued" this suspension for additional years, providing that "'no part of any appropriation in this or any other Act for the [next] fiscal year . . . shall be available for the payment [of the bonuses] notwithstanding'" the statute creating the Government's obligation to pay. Id., at 555-557.
Still, Will held unconstitutional the changes that purported to reduce the Government's payment obligations after the obligation-creating statutes had already taken effect. See 449 U. S., at 224-226, 230.
Here, by contrast, the appropriations riders did not use the kind of "shall not take effect" language decisive in Will. See 449 U. S., at 222-223. Nor did the riders purport to "suspen[d]" §1342 prospectively or to foreclose funds from "any other Act" "notwithstanding" §1342's money-mandating text. Dickerson, 310 U. S., at 556-557; see also Will, 449 U. S., at 206-207. Neither Will nor Dickerson supports the Federal Circuit's implied-repeal holding.
The second strand of precedent turned on provisions that reformed statutory payment formulas in ways "irreconcilable" with the original methods. See United States v. Mitchell, 109 U. S. 146, 150 (1883); see also United States v. Fisher, 109 U. S. 143, 145-146 (1883). In Mitchell, an ap- propriations bill decreased the salaries for federal interpreters (from $400 to $300) and changed how the agency would distribute any "'additional pay'" (from "'all emoluments and allowances whatsoever'" to payments at the agency head's discretion). 109 U. S., at 147, 149. And in Fisher, Congress altered an obligation to pay judges $3,000 per year by providing that a lesser appropriation would be "'in full compensation'" for services rendered in the next fiscal year. 109 U. S., at 144.
The Federal Circuit has also recognized that Congress may override a statutory payment formula through an appropriation that expressly earmarks a lesser amount for that payment obligation in the upcoming fiscal year. See Highland Falls-Fort Montgomery Central School Dist. v. United States, 48 F. 3d 1166, 1169-1171 (1995); see also GAO Redbook 2-62 (discussing Highland Falls and noting that earmarking a lesser amount can create an "irreconcilable conflict" with a statutory payment formula). Perhaps because these cases do not involve an earmark to satisfy an incompatible payment formula, the Federal Circuit did not rely on Highland Falls below.
The appropriations bills here created no such conflict as in Mitchell and Fisher. The riders did not reference §1342's payment formula at all, let alone "irreconcilabl[y]" change it. Mitchell, 109 U. S., at 150. Nor did they provide that Risk Corridors payments from profitable plans would be "'in full compensation'" of the Government's obligation to unprofitable plans. Fisher, 109 U. S., at 146. Instead, the riders here must be taken at face value: as a "mere omission to appropriate a sufficient sum." Vulte, 233 U. S., at 515. Congress could have used the kind of language we have held to effect a repeal or suspension—indeed, it did so in other provisions of the relevant appropriations bills. See, e.g., §716, 128 Stat. 2163 ("None of the funds appropriated or otherwise made available by this or any other Act shall be used . . . "); §714, 129 Stat. 2275 (same); §714, 131 Stat. 168 (same). But for the Risk Corridors program, it did not.
We also find unpersuasive the only pieces of legislative history that the Federal Circuit cited. According to the Court of Appeals, a floor statement and an unpublished GAO letter provided "clear intent" to cancel or "suspend" the Government's Risk Corridors obligation. See 892 F. 3d, at 1318-1319, 1325-1326. We doubt that either source could ever evince the kind of clear congressional intent required to repeal a statutory obligation through an appropriations rider. See United States v. Kwai Fun Wong, 575 U. S. 402, 412 (2015). But even if they could, they did not do so here.
The floor statement (which Congress adopted as an "explanatory statement") does not cross the clear-expression threshold. See 160 Cong. Rec. 17805, 18307 (2014); see also §4, 128 Stat. 2132. That statement interpreted an HHS regulation as saying that "the risk corridor program will be budget neutral, meaning the federal government will never pay out more than it collects." 160 Cong. Rec., at 18307. But that misunderstands the referenced regulation, which provided only that HHS "project[ed]" that the program would be budget neutral and that the agency "intend[ed]" to treat it that way, while making clear that "it [was] difficult to estimate" the "aggregate risk corridors payments and charges at [the] time." 79 Fed. Reg. 13829. HHS' goals did not alter its prior interpretation that the Risk Corridors program was "not statutorily required to be budget neutral." 78 Fed. Reg. 15473. And neither the floor statement nor the appropriations rider said anything requiring budget neutrality or redefining §1342's formula.
The statement provides in full:
"In 2014, HHS issued a regulation stating that the risk corridor program will be budget neutral, meaning that the federal government will never pay out more than it collects from issuers over the three year period risk corridors are in effect. The agreement includes new bill language to prevent the CMS Program Management appropriation account from being used to support risk corridors payments." 160 Cong. Rec., at 18307.
In this implied-repeal context, it is also telling that Congress considered—but did not enact—bills containing the type of text that may have satisfied the clear-expression rule. See e.g., Obamacare Taxpayer Bailout Protection Act, S. 2214, 113th Cong., 2d Sess., §2 (2014) ("'[T]he Secretary shall ensure that payments out and payments in . . . are provided for in amounts that the Secretary determines are necessary to reduce to zero the cost . . . to the Federal Government of carrying out the program under this section'"); Taxpayer Bailout Protection Act, S. 359, 114th Cong., 1st Sess., §2 (2015) ("'The Secretary shall ensure that the amount of payments to plans . . . does not exceed the amount of payments to the Secretary'" and "'shall proportionately decrease the amount of payments to plans'"); Taxpayer Bailout Protection Act, H. R. 724, 114th Cong., 1st Sess., §2 (2015) (same).
The GAO letter is even more inapt. In it, the GAO responded to two legislators' inquiry by identifying two sources of available funding for the first year of Risk Corridors payments: CMS' appropriations for the 2014 fiscal year and profitable insurance plans' payments to the Secretary. 892 F. 3d, at 1318; see also App. in No. 17-1994 (CA Fed.), pp. 234-240. Because the rider cut off the first source of funds, the Federal Circuit inferred congressional intent "to temporarily cap" the Government's payments "at the amount of payments" profitable plans made "for each of the applicable years" of the Risk Corridors program. 892 F. 3d, at 1325. That was error. The letter has little value because it appears nowhere in the legislative record. Perhaps for that reason, the Government does not rely on it.
Having found that the Risk Corridors statute established a valid yet unfulfilled Government obligation, this Court must turn to a final question: Where does petitioners' lawsuit belong, and for what relief? We hold that petitioners properly relied on the Tucker Act to sue for damages in the Court of Federal Claims.
The United States is immune from suit unless it unequivocally consents. United States v. Navajo Nation, 556 U. S. 287, 289 (2009). The Government has waived immunity for certain damages suits in the Court of Federal Claims through the Tucker Act, 24 Stat. 505. See United States v. Mitchell, 463 U. S. 206, 212 (1983). That statute permits "claim[s] against the United States founded either upon the Constitution, or any Act of Congress or any regulation of an executive department, or upon any express or implied contract with the United States, or for liquidated or unliquidated damages in cases not sounding in tort." 28 U. S. C. §1491(a)(1).
The Tucker Act, however, does not create "substantive rights." Navajo Nation, 556 U. S., at 290. A plaintiff relying on the Tucker Act must premise her damages action on "other sources of law," like "statutes or contracts." Ibid. For that reason, "[n]ot every claim invoking the Constitution, a federal statute, or a regulation is cognizable under the Tucker Act." Mitchell, 463 U. S., at 216. Nor will every "failure to perform an obligation . . . creat[e] a right to monetary relief" against the Government. United States v. Bormes, 568 U. S. 6, 16 (2012).
To determine whether a statutory claim falls within the Tucker Act's immunity waiver, we typically employ a "fair interpretation" test. A statute creates a "right capable of grounding a claim within the waiver of sovereign immunity if, but only if, it 'can fairly be interpreted as mandating compensation by the Federal Government for the damage sustained.'" United States v. White Mountain Apache Tribe, 537 U. S. 465, 472 (2003) (quoting Mitchell, 463 U. S., at 217)); see also Navajo Nation, 556 U. S., at 290 ("The other source of law need not explicitly provide that the right or duty it creates is enforceable through a suit for damages"). Satisfying this rubric is generally both necessary and sufficient to permit a Tucker Act suit for damages in the Court of Federal Claims. White Mountain Apache, 537 U. S., at 472-473.
Relying on Alexander v. Sandoval, 532 U. S. 275 (2001), the dissent's logic suggests that a federal statute could never provide a cause of action for damages absent magic words explicitly inviting suit. See post, at 2, 4-7. We have repeatedly rejected that notion—including in opinions written by Sandoval's author. See, e.g., United States v. Bormes, 568 U. S. 6, 15-16 (2012); United States v. Navajo Nation, 556 U. S. 287, 290 (2009). Not even Sandoval went as far as the dissent; that decision instead explained that "[t]he judicial task is to interpret the statute Congress has passed to determine whether it displays an intent to create not just a private right but also a private remedy." 532 U. S., at 286. That is precisely what the money-mandating inquiry does: It provides a framework for determining when Congress has authorized a claim against the Government.
This framework also makes good sense. Cf. post, at 4. As the author of Sandoval explained, if a statutory obligation to pay money is mandatory, then the congressionally conferred "right to receive money," post, at 8, n. 5, will typically display an intent to provide a damages remedy for the defaulted amount, Bowen v. Massachusetts, 487 U. S. 879, 923 (1988) (Scalia, J., dissenting) (a "statute commanding the payment of a specified amount of money by the United States impliedly authorizes (absent other indication) a claim for damages in the defaulted amount"). As this Court recently observed, Congress enacted the Tucker Act to "suppl[y] the missing ingredient for an action against the United States for the breach of monetary obligations not otherwise judicially enforceable." Bormes, 568 U. S., at 12.
By the dissent's contrary suggestion, not only is a mandatory statutory obligation to pay meaningless, so too is a constitutional one. After all, the Constitution did not "expressly create . . . a right of action," post, at 3, when it mandated "just compensation" for Government takings of private property for public use, Amdt. 5; see also First English Evangelical Lutheran Church of Glendale v. County of Los Angeles, 482 U. S. 304, 315316 (1987). Although there is no express cause of action under the Takings Clause, aggrieved owners can sue through the Tucker Act under our case law. E.g., Ruckelshaus v. Monsanto Co., 467 U. S. 986, 10161017 (1984) (citing United States v. Causby, 328 U. S. 256, 267 (1946)).
But there are two exceptions. The Tucker Act yields when the obligation-creating statute provides its own detailed remedies, or when the Administrative Procedure Act, 60 Stat. 237, provides an avenue for relief. See Bormes, 568 U. S., at 13, 16; Bowen v. Massachusetts, 487 U. S. 879, 900-908 (1988).
Petitioners clear each hurdle: The Risk Corridors statute is fairly interpreted as mandating compensation for damages, and neither exception to the Tucker Act applies.
Rarely has the Court determined whether a statute can "fairly be interpreted as mandating compensation by the Federal Government." Mitchell, 463 U. S., at 216-217 (internal quotation marks omitted). Likely this is because so-called money-mandating provisions are uncommon, see M. Solomson, Court of Federal Claims: Jurisdiction, Practice, and Procedure 4-18 (2016), and because Congress has at its disposal several blueprints for conditioning and limiting obligations, see n. 7, supra; see also GAO Redbook 2-22 to 2-24, 2-54 to 2-58. But Congress used none of those tools in §1342. The Risk Corridors statute is one of the rare laws permitting a damages suit in the Court of Federal Claims.
Here again §1342's mandatory text is significant. Statutory "'shall pay' language" often reflects congressional intent "to create both a right and a remedy" under the Tucker Act. Bowen, 487 U. S., at 906, n. 42; see also, e.g., id., at 923 (Scalia, J., dissenting) ("[A] statute commanding the payment of a specified amount of money by the United States impliedly authorizes (absent other indication) a claim for damages in the defaulted amount"); United States v. Testan, 424 U. S. 392, 404 (1976) (suggesting that the Back Pay Act, 5 U. S. C. §5596, may permit damages suits under the Tucker Act "in carefully limited circumstances"); Mitchell, 463 U. S., at 217 (similar). Section 1342's triple mandate—that the HHS Secretary "shall establish and administer" the program, "shall provide" for payment according to the statutory formula, and "shall pay" qualifying insurers—falls comfortably within the class of money-mandating statutes that permit recovery of money damages in the Court of Federal Claims.
Bolstering our finding is §1342's focus on compensating insurers for past conduct. In assessing Tucker Act actions, this Court has distinguished statutes that "attempt to compensate a particular class of persons for past injuries or labors" from laws that "subsidize future state expenditures." Bowen, 487 U. S., at 906, n. 42. (The first group permits Tucker Act suits; the second does not.) The Risk Corridors statute sits securely in the first category: It uses a backwards-looking formula to compensate insurers for losses incurred in providing healthcare coverage for the prior year.
Despite agreeing that "[t]he Court is correct" on the case law, the dissent proposes supplemental briefing and re-argument. Post, at 4, 8. We underscore, however, that all Members of this Court agree that today's cases do not break new doctrinal ground.
The Federal Circuit, moreover, concurs in our conclusion. 892 F. 3d, 1311, 1320, n. 2 (2018) (holding that §1342 "is moneymandating for [Tucker Act] purposes" (citing Greenlee County v. United States, 487 F. 3d 871, 877 (CA Fed. 2007))). It also agrees with our analysis broadly, having held that "shall pay" language "generally makes a statute moneymandating" under the Tucker Act. Id., at 877 (internal quotation marks omitted). Conversely, the Court of Appeals has concluded that a statute is not money mandating where the Government enjoys "complete discretion" in determining whether (and whom) to pay. See, e.g., Doe v. United States, 463 F. 3d 1314, 1324 (2006) (noting that the statutory term, "may," creates a rebuttable presumption that the "statute creates discretion").
Nor is there a separate remedial scheme supplanting the Court of Federal Claims' power to adjudicate petitioners' claims.
True, the Tucker Act "is displaced" when "a law assertedly imposing monetary liability on the United States contains its own judicial remedies." Bormes, 568 U. S., at 12. A plaintiff in that instance cannot rely on our "fair interpretation" test, and instead must stick to the money-mandating statute's "own text" to "determine whether the damages liability Congress crafted extends to the Federal Government." Id., at 15-16. Examples include the Fair Credit Reporting Act, 84 Stat. 1127, and the Agricultural Marketing Agreement Act of 1937, 50 Stat. 246. The former superseded the Tucker Act by creating a cause of action, imposing a statute of limitations, and providing subject-matter jurisdiction in federal district courts. See 15 U. S. C. §§1681n, 1681o, 1681p; Bormes, 568 U. S., at 15. And the latter did so by allowing aggrieved parties to petition the Secretary of Agriculture and by paving a path for judicial review. See 7 U. S. C. §608c(15); Horne v. Department of Agriculture, 569 U. S. 513, 527 (2013).
Unlike those statutes, however, the Affordable Care Act did not establish a comparable remedial scheme. Nor has the Government identified one. So this exception to the Tucker Act is no barrier here.
Neither does the Administrative Procedure Act bar petitioners' Tucker Act suit. To be sure, in Bowen, this Court held in the Medicaid context that a State properly sued the HHS Secretary under the Administrative Procedure Act (not the Tucker Act) in district court (not the Court of Federal Claims) for failure to make statutorily required payments. See 487 U. S., at 882-887, 901-905.
But Bowen is distinguishable on several scores. First, the relief requested there differed materially from what petitioners pursue here. In Bowen, the State did not seek money damages, but instead sued for prospective declaratory and injunctive relief to clarify the extent of the Government's ongoing obligations under the Medicaid program. Unlike §1342, which "provide[s] compensation for specific instances of past injuries or labors," id., at 901, n. 31, the pertinent Medicaid provision was a "grant-in-aid program," which "direct[ed] the Secretary . . . to subsidize future state expenditures," id., at 906, n. 42. Thus, the suit in Bowen "was not merely for past due sums, but for an injunction to correct the method of calculating payments going forward." Great-West Life & Annuity Ins. Co. v. Knudson, 534 U. S. 204, 212 (2002). And because the Court of Federal Claims "does not have the general equitable powers of a district court to grant prospective relief," 487 U. S., at 905, the Court reasoned that Bowen belonged in district court.
Second, the parties' relationship in Bowen also differs from the one implicated here. The State had employed the Administrative Procedure Act in Bowen because of the litigants' "complex ongoing relationship," which made it important that a district court adjudicate future disputes. Id., at 905; see also id., at 900, n. 31. The Court added that the Administrative Procedure Act "is tailored" to "[m]anaging the relationships between States and the Federal Government that occur over time and that involve constantly shifting balance sheets," while the Tucker Act is suited to "remedy[ing] particular categories of past injuries or labors for which various federal statutes provide compensation." Id., at 904-905, n. 39.
These observations confirm that petitioners properly sued the Government in the Court of Federal Claims. Petitioners' prayer for relief under the Risk Corridors statute looks nothing like the requested redress in Bowen. Petitioners do not ask for prospective, nonmonetary relief to clarify future obligations; they seek specific sums already calculated, past due, and designed to compensate for completed labors. The Risk Corridors statute and Tucker Act allow them that remedy. And because the Risk Corridors program expired years ago, this litigation presents no special concern about managing a complex ongoing relation- ship or tracking ever-changing accounting sheets. Petitioners' suit thus lies in the Tucker Act's heartland.
The dissent concedes that there may "be some sharply defined categories of claims that may be properly asserted" through the Tucker Act "simply as a matter of precedent." Post, at 6, and nn. 3, 4 (citing takings, breach-of-contract, failure-to-pay-compensation, and breach-of-fiduciary-duty claims as examples). Petitioners' claim—breach of an unambiguous statutory promise to pay for services rendered to the Government—fits easily within those precedents. The only differences the dissent seems to assert here are that the dollar figure is higher and that petitioners do not deserve a "bailout" for their "bet" that the Federal Government would comply with federal law. Post, at 2, 3, 7; but cf., e.g., 79 Fed. Reg. 30260 (assuring insurers with "concerns that risk corridors collections may not be sufficient to fully fund risk corridors payments" that the Government would still pay). Our analysis in Tucker Act cases has never revolved on such results-oriented reasoning.
In establishing the temporary Risk Corridors program, Congress created a rare money-mandating obligation requiring the Federal Government to make payments under §1342's formula. And by failing to appropriate enough sums for payments already owed, Congress did simply that and no more: The appropriation bills neither repealed nor discharged §1342's unique obligation. Lacking other statutory paths to relief, and absent a Bowen barrier, petitioners may seek to collect payment through a damages action in the Court of Federal Claims.
Having found that the Risk Corridors statute is a money-mandating provision for which a Tucker Act suit lies, we need not resolve petitioners' alternative arguments for recovery based on an implied-in-fact contract theory or under the Takings Clause.
These holdings reflect a principle as old as the Nation itself: The Government should honor its obligations. Soon after ratification, Alexander Hamilton stressed this insight as a cornerstone of fiscal policy. "States," he wrote, "who observe their engagements . . . are respected and trusted: while the reverse is the fate of those . . . who pursue an opposite conduct." Report Relative to a Provision for the Support of Public Credit (Jan. 9, 1790), in 6 Papers of Alexander Hamilton 68 (H. Syrett & J. Cooke eds. 1962). Centuries later, this Court's case law still concurs.
The judgments of the Court of Appeals are reversed, and the cases are remanded for further proceedings consistent with this opinion.
It is so ordered. ON WRITS OF CERTIORARI TO THE UNITED STATES COURT OF APPEALS FOR THE FEDERAL CIRCUIT
JUSTICE ALITO, dissenting.
Twice this Term, we have made the point that we have basically gotten out of the business of recognizing private rights of action not expressly created by Congress. Just a month ago in Comcast Corp. v. National Assn. of African American-Owned Media, 589 U. S. ___, ___-___ (2020) (slip op., at 5-6), after noting a 1975 decision inferring a private right of action under 42 U. S. C. §1981, we wrote the following about that decision:
Johnson v. Railway Express Agency, Inc., 421 U. S. 454, 459 (1975).
"That was during a period when the Court often 'assumed it to be a proper judicial function to provide such remedies as are necessary to make effective a statute's purpose.' Ziglar v. Abbasi, 582 U. S. ___, ___ (2017) (slip op., at 8) (internal quotation marks omitted). With the passage of time, of course, we have come to appreciate that, '[l]ike substantive federal law itself, private rights of action to enforce federal law must be created by Congress' and '[r]aising up causes of action where a statute has not created them may be a proper function for common-law courts, but not for federal tribunals.' Alexander v. Sandoval, 532 U. S. 275, 286-287 (2001) (internal quotation marks omitted)."
A month before that, in Hernández v. Mesa, 589 U. S. ___ (2020), we made the same point and accordingly refused to infer a cause of action under the Fourth Amendment for an allegedly unjustified cross-border shooting. We reasoned that "a lawmaking body that enacts a provision that creates a right . . . may not wish to pursue the provision's purpose to the extent of authorizing private suits for damages." Id., at ___ (slip op., at 5). Other recent opinions are similar. See, e.g., Ziglar v. Abbasi, 582 U. S. ___, ___-___, ___ (2017) (slip op., at 9-12, 23); Jesner v. Arab Bank, PLC, 584 U. S. ___, ___-___ (2018) (slip op., at 18-19); id., at ___ (THOMAS, J., concurring) (slip op., at 1); id., at ___, ___-___ (ALITO, J., concurring in part and concurring in judgment) (slip op., at 1, 3-4); id., at ___ (GORSUCH, J., concurring in part and concurring in judgment) (slip op., at 1).
Today, however, the Court infers a private right of action that has the effect of providing a massive bailout for insurance companies that took a calculated risk and lost. These companies chose to participate in an Affordable Care Act program that they thought would be profitable. I assume for the sake of argument that the Court is correct in holding that §1342 of the Affordable Care Act created an obligation that was not rescinded by subsequent appropriations riders. Thus, for present purposes, I do not dispute the thrust of the analysis in Parts I-III of the opinion of the Court.
My disagreement concerns the critical question that the Court decides in the remainder of its opinion. In order for petitioners to recover, federal law must provide a right of action for damages. The Tucker Act, 28 U. S. C. §1491, under which petitioners brought suit, provides a waiver of sovereign immunity and a grant of federal-court jurisdiction, but it does not create any right of action. See, e.g., United States v. Navajo Nation, 556 U. S. 287, 290 (2009). Nor does any other federal statute expressly create such a right of action. The Court, however, holds that §1342 of the Affordable Care Act does so by implication. Because §1342 says that the United States "shall pay" for the companies' losses, 42 U. S. C. §18062(b)(1), the Court finds it is proper to infer a private right of action to recover for these losses.
This is an important step. Under the Court's decision, billions of taxpayer dollars will be turned over to insurance companies that bet unsuccessfully on the success of the program in question. This money will have to be paid even though Congress has pointedly declined to appropriate money for that purpose.
Not only will today's decision have a massive immediate impact, its potential consequences go much further. The Court characterizes provisions like §1342 as "rare," ante, at 26, but the phrase the "Secretary shall pay"—the language that the Court construes as creating a cause of action—appears in many other federal statutes.
The Court concludes that it is proper for us to recognize a right of action to collect damages from the United States under any statute that "'can fairly be interpreted as mandating compensation.'" Ante, at 24. The Court is correct that prior cases have set out this test, but as the Court acknowledges, we have "[r]arely" had to determine whether it was met. See ante, at 26. And we have certainly never inferred such a right in a case even remotely like these.
Nor has any prior case provided a reasoned explanation of the basis for the test. In United States v. Testan, 424 U. S. 392 (1976), the Court simply lifted the language in question from an opinion of the old United States Court of Claims before holding that the test was not met in the case at hand. Id., at 400-402 (citing Eastport S. S. Corp. v. United States, 178 Ct. Cl. 599, 607, 372 F. 2d 1002, 1009 (1967)). The Court of Claims opinion, in turn, did not explain the origin or basis for this test. See id., at 607, 372 F. 2d, at 1009. And not only have later cases parroted this language, they have expanded it. In United States v. White Mountain Apache Tribe, 537 U. S. 465, 473 (2003) (emphasis added), the Court wrote that "[i]t is enough . . . that a statute . . . be reasonably amenable to the reading that it mandates a right of recovery in damages."
Despite the uncertain foundation of this test, our post-Testan decisions have simply taken it as a given. I would not continue that practice. Before holding that this test requires the payment of billions of dollars that Congress has pointedly refused to appropriate, we ought to be sure that there is a reasonable basis for this test. And that is questionable.
Moreover, there is at least an argument that the Court's application of the test here is itself in conflict with United States v. Testan, 424 U. S. 392, 400 (1976), which also directed that the "grant of a right of action must be made with specificity."
There is obvious tension between what the Court now calls the "money-mandating" test, ante, at 26-27, and our recent decisions regarding the recognition of private rights of action. Take the statute at issue in our Comcast decision. That provision, 42 U. S. C. §1981(a), states:
"All persons within the jurisdiction of the United States shall have the same right in every State and Territory to make and enforce contracts, to sue, be parties, give evidence, and to the full and equal benefit of all laws and proceedings for the security of persons and property as is enjoyed by white citizens." (Emphasis added.)Our opinion in Comcast suggested that we might not find this "shall have" language sufficient to justify the recognition of a damages claim if the question came before us today as a matter of first impression. See 589 U. S., at ___-___ (slip op., at 5-6). But if that is so, how can we reach a different conclusion with respect to the "shall pay" language in §1342 of the Affordable Care Act? Similarly, the Fourth Amendment provides that "[t]he right of the people to be secure . . . against unreasonable . . . seizures . . . shall not be violated." (Emphasis added.) Can this rights-mandating language be distinguished from what the Court describes as the "money-mandating" language found in §1342? See Hernández, 589 U. S., at ___, ___-___ (slip op., at 8, 19-20) (rejecting extension of Bivens v. Six Unknown Fed. Narcotics Agents, 403 U. S. 388 (1971), to Fourth Amendment claim arising in a "new context").
One might argue that the assumptions underlying the enactment of the Tucker Act justify our exercising more leeway in inferring rights of action that may be asserted under that Act. When the Tucker Act was enacted in 1887, Congress undoubtedly assumed that the federal courts would "'[r]ais[e] up causes of action,'" Alexander v. Sandoval, 532 U. S. 275, 287 (2001), in the manner of a common-law court. At that time, federal courts often applied general common law. But since Erie R. Co. v. Tompkins, 304 U. S. 64 (1938), the federal courts have lacked this power. Yet the "money-mandating" test that the Court applies today, ante, at 26-27, and n. 13, bears a disquieting resemblance to the sort of test that a common-law court might use in deciding whether to create a new cause of action. To be sure, some of the claims asserted under the Tucker Act, most notably contract claims, are governed by the new federal common law that applies in limited areas involving "'uniquely federal interests.'" Boyle v. United Technologies Corp., 487 U. S. 500, 504 (1988); see also Testan, 424 U. S., at 400. And the recognition of an implied right to recover on such claims is thus easy to reconcile with the post-Erie regime. There may also be some sharply defined categories of claims that may be properly asserted simply as a matter of precedent. But the exercise of common-law power in cases like the ones now before us is a different matter.
Takings claims are an example. During the period when federal courts applied general common law, such claims were brought under the Tucker Act, apparently on the theory of implied contract. See, e.g., Hurley v. Kincaid, 285 U. S. 95, 104 (1932); United States v. Lynah, 188 U. S. 445, 458-459 (1903). But the Court rejected the argument that a takings claim could be based "exclusively on the Constitution, without reference to any statute of the United States, or to any contract arising under an act of Congress." Hooe v. United States, 218 U. S. 322, 335 (1910).
Compare Testan, 424 U. S., at 400 (suggesting that private remedies might be available for contract claims); United States v. Mitchell, 463 U. S. 206, 224-228 (1983) (relying on "fiduciary relationship . . . [that] arises when the Government assumes . . . control over forests and property belonging to Indians" to create cause of action); Bell v. United States, 366 U. S. 393 (1961) (adjudicating suit brought by former service members for compensation while they were prisoners of war), with Bowen v. Massachusetts, 487 U. S. 879, 905, n. 42 (1988) (rejecting cause of action cognizable under the Tucker Act based on "shall pay" requirement under the Medicaid Act, 42 U. S. C. §1396b(a)).
An argument based on Congress's assumptions in enacting the Tucker Act would present a question that is similar to one we have confronted under the Alien Tort Statute (ATS), a provision like the Tucker Act that grants federal jurisdiction but does not itself create any right of action. Sosa v. Alvarez-Machain, 542 U. S. 692, 713 (2004). Our cases have assumed that the ATS was enacted on the assumption that it would provide a jurisdictional basis for plaintiffs to assert common-law claims, see id., at 724, but our recent cases have held that even there, we should exercise "great caution" before recognizing any new claims not created by statute, id., at 728. See also Jesner, 584 U. S., at ___-___ (slip op., at 18-19); Kiobel v. Royal Dutch Petroleum Co., 569 U. S. 108, 116-117 (2013). There is every reason to believe that a similar caution should guide cases under the Tucker Act—especially when billions of dollars of federal funds are at stake. The money-mandating test that the Court applies here is in stark tension with this precedent.
Despite its importance, the legitimacy of inferring a right of action under §1342 has not received much attention in these cases. The Federal Circuit addressed the question in passing in a footnote, 892 F. 3d 1311, 1320, n. 2 (2018), and in this Court, the briefing and argument focused primarily on other issues. No attempt was made to reconcile our approach to inferring rights of action in Tucker Act cases with our broader jurisprudence.
I am unwilling to endorse the Court's holding in these cases without understanding how the "money-mandating" test on which the Court relies fits into our general approach to the recognition of implied rights of action. Because the briefing and argument that we have received have not fully addressed this important question, I would request supplemental briefing and set the cases for re-argument next Term.
The Court claims that the logic of this opinion "suggests that a federal statute could never provide a cause of action for damages absent magic words explicitly inviting suit." Ante, at 25, n. 12. But all I suggest is that the Court request briefing on the question of inferring causes of action to recover damages under the Tucker Act. The Court makes no effort to explain how the test it applies here can be reconciled with our general approach to inferring private rights of action but is apparently content to allow that inconsistency to remain.
The Court is flatly wrong in saying that the test in Alexander v. Sandoval, 532 U. S. 275, 286 (2001)—whether a statute "displays an intent to create not just a private right but also a private remedy"—is "precisely" the same as its "moneymandating inquiry." Ante, at 25, n. 12. In fact, the "moneymandating inquiry" is precisely contrary to the statement in Sandoval. Sandoval said unequivocally that it is not enough if a statute merely "displays an intent to create . . . a private right," 532 U. S., at 286, but according to the Court, it is sufficient for a statute to manifest only an intent to create a right to receive money.
The Court asserts that there is no real difference between the billiondollar private right of action that the Court now creates on behalf of sophisticated economic actors and our prior precedents, ante, at 30, n. 14, but the Court does not identify analogous precedents—perhaps because there are none to cite.
For these reasons, I respectfully dissent.