United States Trustee v. John Q. Hammons Fall 2006, LLC, 602 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. 22–1238

_________________

OFFICE OF THE UNITED STATES TRUSTEE, PETITIONER v. JOHN Q. HAMMONS FALL 2006, LLC, et al.

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

[June 14, 2024]

Justice Jackson delivered the opinion of the Court.

Two Terms ago, in Siegel v. Fitzgerald, 596 U. S. 464 (2022), we held that a statute violated the Bankruptcy Clause’s uniformity requirement because it permitted different fees for Chapter 11 debtors depending on the district where their case was filed. See id., at 479–480, and n. 2. Today, we are asked to determine the remedy for that constitutional violation. We agree with the Government that the appropriate remedy is prospective parity. Requiring equal fees for otherwise identical Chapter 11 debtors going forward comports with congressional intent, corrects the constitutional wrong, and complies with due process.

Resisting this conclusion, respondents, a group of Chapter 11 debtors, argue that they are entitled to a refund. But, as respondents forthrightly concede, adopting their preferred remedy would require us to undercut congressional intent and transform, by judicial fiat, a program that Congress designed to be self-funding into an estimated $326 million bill for taxpayers. Neither remedial principles nor due process requires that incongruous result. We reverse.

I

A

The federal bankruptcy system is administered by two programs. See id., at 468–470. The U. S. Trustee Program, housed within the Department of Justice, administers 88 of the 94 bankruptcy districts. The six remaining districts, all in Alabama and North Carolina, are administered by the Bankruptcy Administrator Program, which the Administrative Office of the U. S. Courts runs under the supervision of the Judicial Conference.

For our purposes, the most salient difference between these two programs is their funding. Congress designed the U. S. Trustee Program to be entirely self-funded by user fees paid by debtors. See 28 U. S. C. §589a(b). By contrast, Congress supports the Bankruptcy Administrator Program through its general appropriation for the Judiciary, with fees used only to offset that funding. See §1930(a)(7).

Despite these different funding schemes, the fees charged to debtors in U. S. Trustee and Bankruptcy Administrator districts were identical between 2001 and 2018. See Siegel, 596 U. S., at 470. During that almost two-decade period, Congress would set the filing and quarterly fees for U. S. Trustee districts, and then the Judicial Conference, pursuant to a standing order, would require Bankruptcy Administrator districts to match them. See ibid. (citing Report of the Proceedings of the Judicial Conference of the United States 46 (Sept./Oct. 2001) (Report Proceedings)).

In 2017, facing a funding shortfall for the U. S. Trustee Program, Congress amended the fee statute to raise fees in U. S. Trustee districts. See 596 U. S., at 470–471. Specifically, Congress increased quarterly fees for new and pending Chapter 11 cases in which debtors disbursed $1 million or more in that quarter. See Div. B, 131Stat. 1232 (2017 Act). Consistent with its goal of maintaining a self-funding U. S. Trustee Program, Congress made the fee increase for large debtors conditional on the operating fund for the program falling below $200 million in the prior fiscal year. See Siegel, 596 U. S., at 470–471. That threshold was met in 2017, so, starting in January 2018, fees increased for large Chapter 11 debtors in U. S. Trustee districts. See ibid.

Despite the Judicial Conference’s standing order, though, fees did not immediately increase in Bankruptcy Administrator districts. See ibid. For reasons that remain obscure, it was not until October 2018 that the Judicial Conference increased fees for newly filed cases in Bankruptcy Administrator districts. See Report Proceedings 11–12 (Sept. 13, 2018). And fees for already pending large Chapter 11 cases in Bankruptcy Administrator districts remained at their 2017 level until Congress mandated equal fees in 2021. See Pub. L. 116–325, §3(d)(2), 134Stat. 5088 (2021 Act). In the interim, a disparity emerged between the fees paid by large Chapter 11 debtors in U. S. Trustee districts and those paid by large Chapter 11 debtors in Bankruptcy Administrator districts. See Siegel, 596 U. S., at 478–479.

B

In Siegel, we traced the origin of that disparity to a single statutory word. See id., at 479–480. The fee statute passed by Congress, and in effect at the time of the 2017 increase, read: “[T]he Judicial Conference of the United States may require the debtor in a case under chapter 11 of title 11” in a Bankruptcy Administrator district “to pay fees equal to those imposed” on otherwise identical debtors in U. S. Trustee districts. 28 U. S. C. §1930(a)(7) (emphasis added). That permissive language, we explained, violated the Constitution’s Bankruptcy Clause. Siegel, 596 U. S., at 480, n. 2.

The Bankruptcy Clause empowers Congress “[t]o establish . . . Laws on the subject of Bankruptcies throughout the United States,” but it requires that such laws be “uniform.” Art. I, §8, cl. 4. Though the Clause “confers broad authority on Congress,” including the flexibility to “enact geographically limited bankruptcy laws . . . if it is responding to a geographically limited problem,” we concluded that the Clause’s grant of power did not extend to the disparate fees facilitated by the permissive language in the fee statute. Siegel, 596 U. S., at 476–477. As we explained, Congress could not constitutionally “treat identical debtors differently based on an artificial funding distinction that Congress itself created.” Id., at 479–480.

Having found a constitutional wrong, we then faced the question of how to remedy it. We acknowledged three options: (1) refund fees for those charged more in U. S. Trustee districts, (2) retroactively extract higher fees from those charged less in Bankruptcy Administrator districts, or (3) require only prospective parity. See id., at 480. The final option, we noted, was already in effect: By the time Siegel reached our Court, Congress had replaced the permissive “may” in the fee statute with a mandatory “shall,” resulting in equal fees for U. S. Trustee and Bankruptcy Administrator districts as of April 2021. Id., at 471 (quoting Pub. L. 116–325, §3(d)(2), 134Stat. 5088). But, because the remedial question had not been passed on below, we remanded for the Fourth Circuit to address it in the first instance. See Siegel, 596 U. S., at 481.

C

As in Siegel, this case arises from a Chapter 11 case filed in a U. S. Trustee district. Cf. id., at 471. In 2016, a group of 76 legal entities related to a chain of hotels and resorts filed for bankruptcy in the District of Kansas. Starting in January 2018, the debtors were subjected to increased quarterly fees under the amended fee statute. In March 2020, the debtors challenged the constitutionality of those fees, seeking both a refund of fees already paid and a reversion of future fees to their 2017 level. See Debtors’ Motion To Determine Extent of Liability for Quarterly Fees Payable in No. 16–21142 (Bkrtcy. Ct. Kan., Mar. 3, 2020), ECF Doc. 2823. Finding no constitutional violation, the Bankruptcy Court did not reach the remedial question. See In re John Q. Hammons Fall 2006, LLC, 618 B. R. 519, 525–526 (Kan. 2020).

The Tenth Circuit reversed. See In re John Q. Hammons Fall 2006, LLC, 15 F. 4th 1011, 1016 (2021). It anticipated our holding in Siegel and found that the fee statute permitting nonuniform fees violated the Bankruptcy Clause. See id., at 1025. To remedy that violation, the panel then ordered a refund of the debtors’ quarterly fees so that they equaled the lower fees the debtors would have paid had their case been filed in a Bankruptcy Administrator district. See id., at 1025–1026. The U. S. Trustee sought certiorari. After deciding Siegel, we granted the petition, vacated the Tenth Circuit’s judgment, and remanded for further consideration. 596 U. S. ___ (2022). The Tenth Circuit sought supplemental briefing, but ultimately reinstated its original opinion without alteration. See In re John Q. Hammons Fall 2006, LLC, 2022 WL 3354682, *1 (Aug. 15, 2022). After rehearing was denied, the U. S. Trustee again petitioned for review.

We granted certiorari to answer the remedial question left open in Siegel. 600 U. S. ___ (2023).

II

Across remedial contexts, “the nature of the violation determines the scope of the remedy.” Swann v. Charlotte-Mecklenburg Bd. of Ed., 402 U. S. 1, 16 (1971); see also Ayotte v. Planned Parenthood of Northern New Eng., 546 U. S. 320, 328 (2006) (“Generally speaking, when confronting a constitutional flaw in a statute, we try to limit the solution to the problem”). Accordingly, before we can determine the appropriate remedy for the Bankruptcy Clause violation in this case, we must bear down upon the particulars of the constitutional violation we identified in Siegel. Three aspects of our holding are worth highlighting.

First, the violation we identified was nonuniformity, not high fees. There was no doubt raised in Siegel about Congress’s power to raise fees for large Chapter 11 debtors. The constitutional issue arose only because the fee statute’s permissive language effectively “exempted debtors in” Bankruptcy Administrator districts from paying the new rates, resulting in a disparity in fees between the two types of bankruptcy districts. Siegel, 596 U. S., at 468. Though respondents understandably complain about their higher payments, our task is not necessarily to reduce them; it is to remedy the disparity.[1]

Second, the fee disparity at issue here was short lived. It began in January 2018. By October 2018, the Judicial Conference required newly filed Chapter 11 cases in Bankruptcy Administrator districts to pay the higher fees. And starting in April 2021, Congress required uniform fees for pending cases too. Due to these policy shifts by the Judicial Conference and Congress, a large Chapter 11 debtor was subject to, at most, three years and three months of nonuniform treatment.

Finally, the disparity was small. The Government estimates (and respondents do not dispute) that, during the relevant period, only about 50 out of the more than 2,000 cases involving large Chapter 11 debtors were filed in Bankruptcy Administrator districts—a mere 2%. See Brief for Petitioner 11; Reply Brief 16–19. Therefore, even when the statute unconstitutionally permitted the complained-of fee disparity, 98% of the relevant class of debtors still paid uniform fees.

In short, the constitutional violation we identified in Siegel created a monetary disparity in bankruptcy fees that was short lived and small. With the limited nature of the constitutional problem in mind, we now turn to the question of how to remedy it.

III

A

“[T]he touchstone for any decision about remedy is legislative intent.” Ayotte, 546 U. S., at 330. Thus, the key question in determining how to remedy a constitutional violation wrought by the legislative process is always “ ‘what the legislature would have willed had it been apprised of the constitutional infirmity.’ ” Sessions v. Morales-Santana, 582 U. S. 47, 73–74 (2017) (quoting Levin v. Commerce Energy, Inc., 560 U. S. 413, 427 (2010)). In cases involving unequal treatment, answering this question generally leads to a focus on two considerations: Congress’s “ ‘ “intensity of commitment” ’ ” to the more broadly applicable rule, and “ ‘ “the degree of potential disruption of the statutory scheme that would occur” ’ ” if we were to extend the exception. Morales-Santana, 582 U. S., at 75 (quoting Heckler v. Mathews, 465 U. S. 728, 739, n. 5 (1984)). In light of our limited institutional competence, we are also cognizant that Congress likely would not have intended relief that is impractical or unworkable. See, e.g., Seila Law LLC v. Consumer Financial Protection Bureau, 591 U. S. 197, 236–237 (2020) (opinion of Roberts, C. J.); Los Angeles Dept. of Water and Power v. Manhart, 435 U. S. 702, 718–723 (1978). And we must keep in mind that our ultimate aim is to remedy the constitutional wrong consistent with congressional intent, not to provide the complaining parties’ preferred form of relief. See, e.g., Barr v. American Assn. of Political Consultants, Inc., 591 U. S. 610, 634–635 (2020) (opinion of Kavanaugh, J.); Morales-Santana, 582 U. S., at 77, n. 29.

As respondents acknowledge, “Congress’s intentions here were unmistakable.” Brief for Respondents 31. Faced with the constitutional violation we identified in Siegel, Congress would have wanted prospective parity, not a refund or retrospective raising of fees. In other words, to remedy the fee disparity, Congress would have wanted to impose equal fees in all districts going forward. This conclusion is clear from the intensity of Congress’s commitment to raising fees in U. S. Trustee districts, the extreme disruption a refund would cause to the bankruptcy system, and Congress’s own decision to remedy the wrong we face by imposing equal fees going forward. We discuss each of these considerations in turn.

Start with Congress’s commitment to higher fees in U. S. Trustee districts. Congress designed the U. S. Trustee Program to “be funded in its entirety by user fees.” Siegel, 596 U. S., at 469. Chapter 11 cases play a central role in achieving that goal. Congress required 100% of Chapter 11 quarterly fees to be deposited in the U. S. Trustee’s operating fund. §589a(b)(5).[2] By 2017, almost two-thirds of the U. S. Trustee Program’s funding came from Chapter 11 fees alone. See H. R. Rep. No. 115–130, p. 7, n. 26 (2017). It is not surprising, then, that when there was a funding shortfall for the U. S. Trustee Program, Congress chose to address it by raising fees on the largest Chapter 11 debtors. See Siegel, 596 U. S., at 470.

In 2021, when Congress amended the fee statute, it removed any doubts about its commitment to raising fees in order to keep the U. S. Trustee Program self-funded. The statute specifically stated that the purpose of keeping fees at an elevated level was “to further the long-standing goal of Congress of ensuring that the bankruptcy system is self-funded, at no cost to the taxpayer.” 2021 Act §2(b); see also §2(a)(1). Respondents point to nothing—in the history of the bankruptcy system, the design of the U. S. Trustee Program, or the 2017 or 2021 Acts—that cuts against Congress’s stated commitment to having higher fees for large Chapter 11 debtors in U. S. Trustee districts.

Now consider the flipside of this clear congressional commitment: the disruption that would follow from granting respondents’ request to refund their fees. Our imposition of a refund would significantly undermine Congress’s goal of keeping the U. S. Trustee Program self-funded. Respondents do not dispute that refunding all large Chapter 11 debtors in U. S. Trustee districts would be expensive; the Government estimates it would cost approximately $326 million. See Brief for Petitioner 35–36; see also Brief for Respondents 21, and n. 6. If the Government’s estimate is even close to correct, the cost of the refund would greatly exceed the $200 million threshold Congress selected in 2017 to signal fiscal distress in the U. S. Trustee Program and trigger higher fees. See Siegel, 596 U. S., at 470–471. Thus, in mandating such a remedy, we would transform a program Congress designed to be self-funding into an enormous bill for taxpayers. It is hard to imagine a remedy more diametrically opposed to clear congressional intent.

On top of that, respondents’ proposed refund would almost certainly exacerbate the small fee disparity we are attempting to remedy. As already noted, respondents are among the 98% of large Chapter 11 debtors who paid higher fees starting in 2018, just as Congress wanted. By refunding them, we would add to the past nonuniformity by increasing the tiny percentage of debtors—currently 2%—who paid lower fees. As the Government aptly notes, even if 95% of the debtors in U. S. Trustee districts that paid higher fees received a refund, we would still end up creating a greater overall disparity than what resulted from Congress’s requirement of prospective parity. See Brief for Petitioner 40.

Of course, it is true that the disparity could be entirely eliminated if all the debtors who paid higher fees were given a refund. But that theoretical possibility blinks reality. The Government estimates that 85% of the large Chapter 11 cases subject to higher fees between January 2018 and April 2021 have closed, and some of those debtors have been liquidated or otherwise ceased to exist. See Reply Brief 20. Respondents offer no meaningful path to reducing the small existing disparity through refunds. Instead, they encourage us to defy congressional intent, disrupt the U. S. Trustee Program’s self-funding mandate, and divert the attendant costs to taxpayers—all to give them a remedy that will make the disparity caused by the constitutional violation worse.

The only real question, then, is whether Congress would have wanted to retrospectively impose higher fees on debtors in Bankruptcy Administrator districts. The best evidence that Congress did not intend such a remedy is that Congress itself chose not to pursue that course. In the 2021 Act, as respondents acknowledge, “Congress revised the fee scheme to address this very issue, and it did so by mandating equal fees prospectively only.” Brief for Respondents 31 (citing Pub. L. 116–325, §§3(d)(2), 3(e)(2)(B), 134Stat. 5088–5089); see also 28 U. S. C. §1930(a)(6)(B)(ii)(II).

Congress’s choice makes sense. Because fees collected in Bankruptcy Administrator districts go toward offsetting the Judiciary’s appropriation, not to supporting the U. S. Trustee Program, retrospectively raising fees in Bankruptcy Administrator districts would do nothing to achieve Congress’s goal of keeping the U. S. Trustee Program self-funding. See §1930(a)(7). Thus, with the 2021 Act, Congress evinced a clear desire to comply with the constitutional mandate of uniformity by requiring prospective parity, but it reasonably chose not to impose higher fees retrospectively in Bankruptcy Administrator districts.

What is more, there are serious practical challenges to a retrospective imposition of higher fees. As in U. S. Trustee districts, many of the debtors who paid lower fees in Bankruptcy Administrator districts have exited bankruptcy or ceased to exist. See Brief for Respondents 38–39. Indeed, the Government estimates that only 10 of the roughly 50 cases involving debtors who paid lower fees are still open. See Reply Brief 17–18. Moreover, locating all the former debtors or their successors would not end the practical problems. The Government would be forced to extract fees from funds that might already be disbursed, inevitably prompting additional litigation and even the unwinding of closed cases. See ibid. And all that effort would be directed against parties who followed the law and complied with the fee schedule imposed by the Judicial Conference under the 2017 Act.

Our remedial principles do not require us to follow that unintended, impractical course. Faced with far more serious dignitary harms than those implicated by a small and short-lived disparity in bankruptcy fees for large debtors, we have deemed prospective parity sufficient to remedy unconstitutional differences in treatment. See Heckler, 465 U. S., at 740, n. 8 (“[W]e have often recognized that the victims of a discriminatory government program may be remedied by an end to the preferential treatment for others”); see also, e.g., Morales-Santana, 582 U. S., at 77 (sex discrimination); Manhart, 435 U. S., at 721 (same). Here, Congress would have wanted prospective parity, and that remedy is sufficient to address the small, short-lived disparity caused by the constitutional violation we identified in Siegel.

B

The dissent offers three primary responses to our analysis thus far. First, the dissent argues that congressional intent is irrelevant, and we should simply defer to the plaintiffs’ request for damages. See post, at 9, 11 (opinion of Gorsuch, J.). For their part, respondents do not claim that this is how our remedial precedent works; as already noted, they agree that “courts crafting constitutional remedies consult ‘the legislature’s intent.’ ” Brief for Respondents 31 (quoting Morales-Santana, 582 U. S., at 73). That’s for good reason: The dissent’s argument turns on a misapprehension of the constitutional wrong at issue here. The remedial question before us is not whether to pay damages or not; it is how to address a short-lived and small disparity. “How equality is accomplished—by extension or invalidation of the unequally distributed benefit or burden, or some other measure—is a matter on which the Constitution is silent.” Levin, 560 U. S., at 426–427. So, when seeking to remedy an unconstitutional disparity, rather than divining the right answer ourselves or picking a party’s preferred form of relief (which may, as in this case, make the disparity worse), we generally look to the intent of the Legislature. See id., at 427–428.

Second, the dissent argues that, if we are to rely on congressional intent, it actually points to a refund. See post, at 13–14. For support, the dissent cites only to the fiscal year 2020 appropriations law. See post, at 13 (citing 133Stat. 2398). But, again, there is a reason that respondents do not advance this argument; in fact, they concede that “Congress . . . address[ed] this very issue” and mandated prospective equalization of fees. Brief for Respondents 31. The dissent cites boilerplate language that simply allows the U. S. Trustee to respond effectively to commonplace overpayments by debtors. See Pub. L. 116–93, 133Stat. 2398 (“[D]eposits . . . and amounts herein appropriated shall be available in such amounts as may be necessary to pay refunds due depositors”). Such statements have been part of every appropriations law for years, including before the disparity at issue here came into existence. See, e.g., 131Stat. 195 (2017 appropriations law); 129Stat. 2298 (2016 appropriations law). Far from confirming a congressional intent to authorize an estimated $326 million refund here, the broader provision the dissent invokes underscores that a refund would send the U. S. Trustee Program into fiscal freefall, contradicting Congress’s intent to have the program be self-funding. See 133Stat. 2398 (estimating fee revenue and structuring appropriations “so as to result in a final fiscal year 2020 appropriation from the general fund estimated at $0”).

Finally, the dissent suggests we need not address congressional intent at all, because the Government actually promised these respondents a refund. See, e.g., post, at 1, 5–7, 18–19, n. 7. Once again, the dissent relies on an argument respondents have not advanced in this Court. And, once again, the dissent might have done better following respondents’ cue. The relied-upon passage in the Government’s bankruptcy court filing is nothing more than a request by the Government not to be forced to provide any remedy until after it has exhausted all appeals. See Objection of the United States to Debtor’s Motion To Determine Extent of Liability for Quarterly Fees Payable in No. 16–21142 (Bkrtcy. Ct. Kan., Apr. 27, 2020), ECF Doc. 2868, pp. 59–61. Read fairly, the Government promised only what you would expect: that it would comply with a final judgment. See ibid.; see also Reply Brief 7, n. 1 (“Although the government does not believe a refund is the appropriate remedy, if it is subject to a judgment directing it to pay a refund, it will of course comply”).

In sum, while the dissent invents new arguments to arrive at its favored outcome, we prefer to stick with the parties and our controlling precedent.[3]

IV

Respondents and the dissent ask us to override Congress’s clear intent because, they claim, due process requires it. See post, at 14–18. To advance this argument, they rely on a series of cases involving unconstitutional state taxes. See, e.g., McKesson Corp. v. Division of Alcoholic Beverages and Tobacco, Fla. Dept. of Business Regulation, 496 U. S. 18 (1990); Harper v. Virginia Dept. of Taxation, 509 U. S. 86 (1993); Reich v. Collins, 513 U. S. 106 (1994); Newsweek, Inc. v. Florida Dept. of Revenue, 522 U. S. 442 (1998) (per curiam). In respondents’ view, these cases stand for the proposition that “due process requires ‘meaningful backward-looking relief ’ unless an ‘exclusive’ predeprivation remedy is both ‘clear and certain.’ ” Brief for Respondents 22 (first quoting Brief for Petitioner 29, in turn quoting McKesson, 496 U. S., at 31, then quoting Newsweek, 522 U. S., at 443–444; capitalization and boldface deleted). Respondents claim that because the predeprivation remedy here was neither exclusive nor clear and certain, they are entitled to a refund. See Brief for Respondents 22–28.

We disagree. To start, we note that the tax cases arrived at their holdings only after scrutinizing close to a century of tax-specific jurisprudence and carefully analyzing the unique interests the taxation context involves, including the Government’s “exceedingly strong interest in financial stability” and the attendant need for prompt payment and postdeprivation protections. See McKesson, 496 U. S., at 37; see also id., at 32–38. The dissent does not dispute this, nor does it adequately explain why we deemed such history and context so central to our holdings in the tax cases. See post, at 17–18. For their part, respondents simply ignore this context entirely. Instead, they replace the word “tax” with “fee,” see Brief for Respondents 22, and assert that the constitutional holding of the tax cases applies to any case involving “monetary injury,” including those arising from the voluntary, fee-for-service bankruptcy system, id., at 9.

No matter, though. Even assuming that the tax cases apply, respondents are not entitled to relief under them. We held in those cases that if there was “ ‘a meaningful opportunity for taxpayers to withhold contested tax assessments and to challenge their validity in a predeprivation hearing,’ the ‘availability of a predeprivation hearing constitute[d] a procedural safeguard . . . sufficient by itself to satisfy the Due Process Clause.’ ” Harper, 509 U. S., at 101 (quoting McKesson, 496 U. S., at 38, n. 21). Here, respondents acknowledge that they had the opportunity to challenge their fees before they paid them. See Brief for Respondents 25 (“[T]he same bankruptcy procedures are open and available before or after paying an invalid fee. Both are equally acceptable for a party to assert and preserve its rights”). Under the tax cases, then, respondents are not entitled to any particular remedy.

Respondents and the dissent misread our later decisions as displacing that basic holding. In subsequent cases, we addressed situations where a State “reconfigure[d] its scheme, unfairly, in mid-course—to ‘bait and switch’ ” taxpayers out of a refund remedy guaranteed under state law. Reich, 513 U. S., at 111; see also Newsweek, 522 U. S., at 444. We held that States could not hold open a postdeprivation refund remedy to encourage payment and then take it away after taxpayers paid. See Reich, 513 U. S., at 111–112; Newsweek, 522 U. S., at 444–445. In this case, though, there was neither a guaranteed refund remedy nor a bait and switch. Nothing in the Bankruptcy Code promises a refund to those who successfully challenge their fees. And respondents point to no previously available statutory remedy of which the Government has now deprived them. So those later cases do not help respondents either.

With all that said, nothing we say here should be taken to diminish or depart from the holdings of the tax cases as they apply in the tax context. Nor do we mean to suggest that congressional intent is an entirely unchecked guide in our remedial analysis for constitutional violations. We cannot remedy an old constitutional problem by creating a new one, so due process and other constitutional protections undoubtedly will limit the possible remedies in many cases. See Barr, 591 U. S., at 633. Here, though, due process does not mandate any particular remedy. Therefore, as the tax cases themselves advise, we must, “within the bounds of [our] institutional competence, . . . implement what the legislature would have willed.” Levin, 560 U. S., at 427.

*  *  *

Faced with the unconstitutional nonuniformity we recognized in Siegel, Congress would have provided for uniform fees going forward. That remedy cures the constitutional violation, and due process does not require another result. The judgment of the Court of Appeals for the Tenth Circuit is reversed, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.

SUPREME COURT OF THE UNITED STATES

_________________

No. 22–1238

_________________

OFFICE OF THE UNITED STATES TRUSTEE, PETITIONER v. JOHN Q. HAMMONS FALL 2006, LLC, et al.

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

[June 14, 2024]

Justice Gorsuch, with whom Justice Thomas and Justice Barrett join, dissenting.

What’s a constitutional wrong worth these days? The Court’s answer today seems to be: not much. Between 2018 and 2020, the government charged fees to bankruptcy debtors that varied arbitrarily from region to region, leaving some debtors millions of dollars worse off than others. Two years ago, we held that this geographically discriminatory treatment violated the Constitution’s Bankruptcy Clause—a provision that, we stressed, was not “toothless.” Siegel v. Fitzgerald, 596 U. S. 464, 468 (2022). Today, however, the Court performs a remedial root canal, permitting the government to keep the cash it extracted from its unconstitutional fee regime.

The path the Court follows is as striking as its destination. Never mind that a refund is the traditional remedy for unlawfully imposed fees. Never mind that the government promised to supply precisely that relief if the debtors in this case prevailed, as they have, in their constitutional challenge. Never mind that backtracking on that promise raises separate due process concerns. As the majority sees it, supplying meaningful relief is simply not worth the effort. Respectfully, that alien approach to remedies has no place in our jurisprudence.

I

A

Certainty is the lifeblood of bankruptcy. For the system to function, a debtor must be certain that putting all his assets on the table for creditors will afford him a fresh start. So too must a creditor have certainty about what priority his loan may or may not enjoy in the event of a borrower’s bankruptcy. Recognizing as much, our Constitution grants Congress power to establish “uniform Laws on the subject of Bankruptcies throughout the United States.” Art. I, §8, cl. 4; see 3 J. Story, Commentaries on the Constitution of the United States §§1101–1103, pp. 4–8 (1833). That provision affords Congress some “flexibility” in drafting bankruptcy laws, but it does not tolerate laws that treat parties in bankruptcy differently based on the “arbitrary” happenstance of their “geograph[y].” Siegel, 596 U. S., at 476. Laws like those, this Court has held, do not apply “uniform[ly] . . . throughout the United States.”

Our case arises from a violation of that uniformity requirement. In much of the country, the United States Trustee Program, housed in the Department of Justice, handles administrative tasks once handled by bankruptcy courts. Id., at 468. The Trustee Program is funded by quarterly fees paid principally “by debtors who file cases under Chapter 11 of the Bankruptcy Code.” Id., at 469; see 28 U. S. C. §1930(a)(6)(A). Thanks to a quirk of history, however, six federal judicial districts are not in the Trustee Program. Instead, they are part of the so-called Administrator Program, overseen by the Judicial Conference of the United States and “funded by the Judiciary’s general budget.” Siegel, 596 U. S., at 469. In those districts, Congress did not require debtors to pay fees “at all.” Ibid. That is, until a lower court highlighted the disparity and held it violated the Bankruptcy Clause. St. Angelo v. Victoria Farms, Inc., 38 F. 3d 1525, 1529–1532 (CA9 1994).

In 2000, Congress implemented a fix. It provided that “the Judicial Conference of the United States may require” debtors in Administrator Program districts “to pay fees equal to those” debtors pay in Trustee Program districts. 114Stat. 2412 (enacting §1930(a)(7)). Although the statutory language (“may require”) was permissive, the Judicial Conference took the hint and began charging the same fees as those levied in Trustee Program districts, thus putting all debtors on equal footing. Siegel, 596 U. S., at 470.

The solution didn’t last. Come 2017, Congress enacted temporary measures to boost Trustee Program funding. There, Congress directed that, whenever Trustee Program funds dropped below $200 million, certain bankruptcy estates had to pay new and much higher quarterly fees (where some once paid $30,000, for example, the law now required them to pay up to $250,000). §1004, 131Stat. 1232; see Siegel, 596 U. S., at 470. The 2017 Act “applied to all pending cases” in Trustee Program districts. Id., at 471. But for reasons not entirely clear from the record before us, the Judicial Conference didn’t immediately follow suit. It waited until October 2018 to implement those changes in Administrator Program districts—and even then applied them “only to newly filed cases.” Ibid.

Ultimately, Congress had to intercede again. At the close of 2020, Congress withdrew its direction to the Judicial Conference providing that it “may require” debtors in Administrator Program districts to pay the same fees as debtors in Trustee Program districts. In its place, Congress issued a more emphatic instruction, telling the Judicial Conference that it “shall” ensure that quarterly fees remain “consistent across all Federal judicial districts.” §§2–3, 134Stat. 5086, 5088.

But if that solved the problem going forward, it left another question unanswered: what to do about Trustee Program debtors who had paid more in fees between 2018 and 2020 than did their similarly situated Administrator Program counterparts. Many Trustee Program debtors brought challenges alleging that the fees they had paid violated the uniformity requirement of the Bankruptcy Clause. And in 2022, we agreed with them, holding that the debtors before us had been subject to “arbitrary geographically disparate” fees in violation of the Constitution. Siegel, 596 U. S., at 476. After reaching that conclusion, we remanded the case then before us for a lower court to determine “the appropriate remedy . . . in the first instance.” Id., at 480–481.

B

John Q. Hammons Hotels & Resorts found itself in the middle of this mess. In 2016, various entities affiliated with Hammons filed Chapter 11 bankruptcy petitions in the District of Kansas, a Trustee Program district. In re John Q. Hammons Fall 2006, LLC, 15 F. 4th 1011, 1018 (CA10 2021). The cases remained pending after the 2017 Act kicked in and before the 2020 Act mandated fee uniformity across the Nation. So Hammons was charged higher quarterly fees than debtors in Administrator Program districts.

Hammons did not challenge the fee disparity immediately. That would have come at a heavy cost: Until Hammons paid its fees in full, the bankruptcy court could not confirm Hammons’s plan of reorganization, a vital step in the Chapter 11 process. See 11 U. S. C. §1129(a)(12). Worse, as a debtor defaulting on its fees, Hammons would also have run the risk of being kicked out of the Chapter 11 process entirely. §§1112(b)(1), (b)(4)(K).

So Hammons waited until early 2020. By that time the bankruptcy court had confirmed its plan. See Debtors’ Motion To Determine Extent of Liability for Quarterly Fees in No. 16–21142 (Bkrtcy. Ct. Kan., Mar. 3, 2020), ECF Doc. 2823, p. 5. But by that time Hammons had also “paid over $2.5 million more in quarterly fees than [it] would have paid had [it] filed in” an Administrator Program district. 15 F. 4th, at 1018. Arguing that this discriminatory treatment was unconstitutional under the Bankruptcy Clause, Hammons sought a refund of those excess payments. ECF Doc. 2823, at 8.

The U. S. Trustee opposed the request. But he promised that “[i]f [Hammons] prevail[ed] after all levels of review on [its] claim that [the fee disparity] is unconstitutional, the government [would] refund fees to the extent they were overpaid.” Objection of the United States to Debtor’s Motion to Determine Extent of Liability for Quarterly Fees in No. 16–21142 (Bkrtcy. Ct. Kan., Apr. 27, 2020), ECF Doc. 2868, p. 59. As reassurance, the U. S. Trustee stressed that Congress had “authorized payments of refunds . . . in its most recent annual appropriation law.” Id., at 59–60 (citing 133Stat. 2398).

This long-promised payment eventually came due. Anticipating our decision in Siegel by a year, in 2021 the Tenth Circuit held that Hammons had been subjected to an arbitrary and geographically disparate fee forbidden by the Bankruptcy Clause. 15 F. 4th, at 1023. By way of remedy, that court held the Trustee to its promise, ordering him to pay Hammons a refund of the fees it had paid in excess of those it would have owed in an Administrator Program district during the same period. Id., at 1026. This Court granted certiorari to address what remedy is due debtors, like Hammons, who were charged unconstitutional fees between 2018 and 2020—the question we left open in Siegel. 600 U. S. ___ (2023).

II

A

Where does that leave us? Before this Court, the U. S. Trustee does not question Hammons suffered a constitutional injury in having to pay nonuniform fees. That much was settled by Siegel. Nor does the U. S. Trustee dispute he promised to refund Hammons its overpayments should it prevail—as it has now prevailed—on the merits of its constitutional claim. Everyone agrees those fees total approximately $2.5 million. Even more than that, it is undisputed Congress has already taken the affirmative step of appropriating funds for refunds in cases just like this one. With all that beyond dispute, the next step should be too: Just as the Tenth Circuit held, the U. S. Trustee should be ordered to make Hammons whole for its injury and pay the promised refund.

Traditional remedial principles command that result. No one argues, for example, that sovereign immunity bars this suit or others like it. Nor is there a question Hammons sought a refund in a timely fashion. As the U. S. Trustee puts it, Congress has allowed “[t]he amounts of the payments [to] be litigated at the time of the budget submission; by filing an adversary proceeding to challenge fees at any time while the bankruptcy case is ongoing; or by filing a district court action after the case has terminated.” Brief for Petitioner 5–6. And Hammons brought its fee challenge while its bankruptcy case was still ongoing. It is long since settled, too, that where (as here) Congress has provided “a general right to sue” for the invasion of a legal right but has not specified any particular form of relief, “federal courts may use any available remedy to make good the wrong done.” Barnes v. Gorman, 536 U. S. 181, 189 (2002) (internal quotation marks omitted). And where (as here), someone pays money—or has money withheld from him—because of invalid government action, the most appropriate remedy is monetary relief.

Centuries of judicial practice confirm as much. This Court has long said that the “[a]ppropriate remedy” for “duties or taxes erroneously or illegally assessed . . . is an action of assumpsit for money had and received.” Philadelphia v. Collector, 5 Wall. 720, 731 (1867). We have held that “the law . . . will compel restitution or compensation” “if a county obtains the money or property of others without authority.” City of Louisiana v. Wood, 102 U. S. 294, 299 (1880) (internal quotation marks omitted). And on the theory that “the appropriate remedy” for unconstitutional discrimination “is a mandate of equal treatment,” Heckler v. Mathews, 465 U. S. 728, 740 (1984) (emphasis deleted), we have “regularly . . . affirmed District Court judgments ordering that welfare benefits be paid to members of an unconstitutionally excluded class,” Califano v. Westcott, 443 U. S. 76, 90 (1979).

Our longstanding precedents should make short work of this case. Hammons remitted to the U. S. Trustee more than $2.5 million in “overpayments.” Siegel, 596 U. S., at 472 (internal quotation marks omitted). Those overpayments were exacted in violation of the Bankruptcy Clause. To remedy the violation, Hammons is entitled to a refund—the relief the U. S. Trustee promised from the start.

B

Despite all this, the government now tries to backtrack. Yes, it promised to pay should Hammons prove a constitutional injury. Yes, Hammons has now done exactly that, consistent with Siegel. Yes, Congress has appropriated sums to make Hammons and others like it whole. And, yes, traditional remedial principles would seem to dictate just that form of relief. Still, the government insists, it should not be forced to pay. It’s an astonishing claim, made all the more astonishing by the fact a majority of the Court goes along with it.

How do they get there? To determine the appropriate remedy for Hammons’s constitutional injury, the government and majority reason, we “must adopt the remedial course Congress likely would have chosen had it been apprised of the constitutional infirmity.” Brief for Petitioner 14 (internal quotation marks omitted). And, they continue, had Congress known in 2017 that the disparate fee arrangement was unconstitutional, it would have responded by imposing higher fees on debtors in the Administrator Program districts. Id., at 14–15. And, the government and majority say, “[t]he most appropriate way to effectuate that remedy is on a purely prospective basis”—ensuring that fees are “uniform going forward.” Id., at 20. Of course, Congress already provided just this prospective relief in the 2020 Act. So really, the government and majority conclude, that means “no further relief is required.” Ibid.; see ante, at 7–11. Presto: No refund for Hammons. It is a line of reasoning as bold as it is untenable.[1]

1

Start with the government and majority’s major premise: the notion that our only proper role is to speculate about—and then give effect to—the course of action Congress would have taken to address the constitutional injury its fee regime imposed had it been warned in advance. Consider what that would mean in a more familiar context. Suppose you suffered some form of arbitrary and unlawful discrimination in the workplace and sued your employer for damages. In response, suppose your employer reassured you that, had it known beforehand what the incident would mean for its wallet, it would have taken steps to avoid the incident—and it promises to do better in the future. In what world does your employer’s promise of a prospective-only remedy do anything to redress your past injuries? And why would it matter what the employer might have done differently?

None of that comports with traditional remedial principles. A promise of fee uniformity going forward may prevent future discrimination between debtors. But it does nothing to remedy fees unlawfully exacted in the past. Far from an “appropriate remedy,” the majority’s prospective remedy for a past injury is no remedy at all. By overlooking the (obvious) distinction between prospective and retrospective relief, the majority defies this Court’s teaching that, in cases like this one, “effective relief consists of damages, not an injunction.” Tanzin v. Tanvir, 592 U. S. 43, 51 (2020).

Nor is it sensible to ask what remedy the government might prefer. This Court has long held that, in our legal system, it is the plaintiff, not the defendant, who “has a right to choose” what form of legally permissible relief he will seek. Twist v. Prairie Oil & Gas Co., 274 U. S. 684, 689 (1927). And for just as long we have considered irrelevant a defendant’s plea that, if he had known what he was doing was wrong, “he would have pursued a different course of action within the law.” Corsicana Nat. Bank of Corsicana v. Johnson, 251 U. S. 68, 88 (1919). Entertaining that kind of “hypothesis,” we have explained, “would be an unwarranted resort to fiction in aid of a wrongdoer, and at the expense of the party injured.” Ibid.

Seeking a way around these problems and following the government’s lead, the majority points to cases in which plaintiffs sought prospective equitable relief from an unconstitutional law. See Brief for Petitioner 14–15.[2] And in that posture, those cases indicate, the Court has sometimes thought it appropriate to ask how much of the challenged statute it should declare inoperative going forward: Should the whole statute, or only parts of it, be held unenforceable in the future?

That question, the Court has sometimes said, poses one of “severability.” Barr v. American Assn. of Political Consultants, Inc., 591 U. S. 610, 614 (2020) (opinion of Kavanaugh, J.); see Ayotte v. Planned Parenthood of Northern New Eng., 546 U. S. 320, 331–332 (2006). Sometimes, Congress will include an express severability clause providing that the unconstitutionality of any one provision will not preclude the enforcement of others going forward. Barr, 591 U. S., at 623. But what happens when a statute contains no such provision? In cases like that, this Court has, from time to time, resorted to asking the hypothetical question: What would Congress “have willed” about the law’s future application had it foreseen its constitutional defect? Levin v. Commerce Energy, Inc., 560 U. S. 413, 427 (2010).

So, for example, in Sessions v. Morales-Santana, 582 U. S. 47 (2017), the Court faced a statute that supplied a faster path to citizenship for children born abroad to American mothers than for those born abroad to American fathers. Id., at 51. The Court held that law violated the Equal Protection Clause. Id., at 72. To resolve how the law should operate going forward consistent with the Constitution, the Court asked whether Congress would have preferred the “ ‘withdrawal of benefits’ ” from children of American mothers or the “ ‘extension of benefits’ ” to children of American fathers, and chose the former option. Id., at 73.[3]

None of that, however, has anything to do with our case. Hammons seeks damages to remedy a past violation. The company does not seek from us any form of prospective relief. As a result, we have no occasion to take a scalpel to Congress’s work. We do not face anything like the question whether to extend or withdraw benefits to ensure a statute’s constitutional operation going forward. Indeed, attempting to do so in this case would be utterly pointless, for in 2020 Congress already modified the challenged provision to remove its constitutional infirmity going forward. And just because future parties will not be injured does nothing to erase the fact that parties injured by past misconduct are entitled to relief.

The decisions the majority relies upon only confirm the point. Take Morales-Santana. While the Court consulted hypothetical legislative intent to resolve a question about the scope of prospective relief, it also acknowledged limits on the propriety of that course. It observed, for example, that legislative intent is “irrelevant” when “a defendant [is] convicted under a law classifying on an impermissible basis”; for that past harm, he is entitled to relief “without regard to the manner in which” Congress might have wanted to “cure the infirmity.” Id., at 74, n. 24. The Court stressed, too, that we “loo[k] to Justice Harlan’s concurring opinion in Welsh v. United States,” 398 U. S. 333 (1970), when considering remedies for discriminatory treatment. 582 U. S., at 75. And that opinion is wholly inconsistent with the majority’s approach today. Guessing how the legislature would have fixed a statute had it known of a constitutional defect might be appropriate “in an action for a declaratory judgment or an action in equity,” Justice Harlan wrote. Welsh, 398 U. S., at 363–364 (opinion concurring in result) (internal quotation marks omitted). But, he added, that course is not “appropriate” in cases, like the one before him, where the plaintiff sought relief for a past harm and the result of guesswork about legislative intentions could leave him “remediless.” Id., at 362.

The few decisions the majority cites addressing requests for retrospective relief make a similar point. Consider Los Angeles Dept. of Water and Power v. Manhart, 435 U. S. 702 (1978), a case alleging unlawful discrimination under Title VII of the Civil Rights Act of 1964. See ante, at 7. That statute, the Court observed, provides that “retroactive relief ‘may’ be awarded if it is ‘appropriate.’ ” 435 U. S., at 719. Despite the permissive statutory language, the Court recognized the traditional “presumption in favor of ” money damages to remedy past discrimination. Ibid. This presumption, Manhart continued, was so strong it “can seldom be overcome.” Ibid. Exactly so.[4]

2

Turn now to the minor premise of the majority’s argument and a second, independent problem emerges. Relying on severability precedents, the majority reasons that Congress would not have wanted to issue refunds in cases like this one. But even assuming speculation about Congress’s wishes has anything to do with the scope of retrospective relief, it would still require a refund here.

When searching for congressional intent, we have said, there is no better place to look than “existing statutory text.” Lamie v. United States Trustee, 540 U. S. 526, 534 (2004). Even in severability cases, we have taken pains to stress that courts may not elevate judicial guesswork about “Congress’s hypothetical intent” over “statutory text,” which is “the definitive expression of Congress’s will.” Barr, 591 U. S., at 624–625 (opinion of Kavanaugh, J.).

Follow those directions here and we end up at a refund. As the government has admitted, existing statutory text reveals that “Congress [has] authorized payments of refunds” from appropriated funds. ECF Doc. 2868, at 59–60; see 133Stat. 2398. This fact is as sure a sign as any that Congress didn’t believe refunds would cause the sort of “ ‘disruption of the statutory scheme’ ” the majority worries over. Ante, at 7. The law gives us our answer—refunds—no guesswork necessary.

How does the majority respond? It points to Congress’s decision in the 2020 Act to “ ‘mandat[e] equal fees prospectively.’ ” Ante, at 10. And that decision, the majority asserts, is “[t]he best evidence that Congress did not intend” for us to permit refunds. Ibid. But the majority never explains why that inference is a good, let alone the best, inference to draw from the 2020 Act’s silence about retroactive relief. Given that Congress had already legislated to provide for refunds, why would it need to repeat itself in the 2020 Act? Cf. Bowen v. Michigan Academy of Family Physicians, 476 U. S. 667, 681 (1986) (“We ordinarily presume that Congress intends the executive to obey its statutory commands”). And, particularly in those circumstances, why wouldn’t the better inference be that Congress assumed courts would apply their ordinary rules and recognize that refunds are the appropriate remedy for illegal fees already exacted? [5]

III

A

Traditional remedial principles guarantee Hammons a refund. Nothing the majority offers begins to suggest otherwise. Still, even if we could somehow put all that aside, this Court’s due process precedents would demand the same result.

Those precedents contemplate cases like this one. We have held that, if an individual “reasonably relie[s] on the apparent availability of a postpayment refund when paying” a contested fee, the government may not later “declare, only after the disputed [fees] have been paid, that no such remedy exists.” Newsweek, Inc. v. Florida Dept. of Revenue, 522 U. S. 442, 444–445 (1998) (per curiam) (internal quotation marks omitted). This due process rule holds true even when the individual had the option of pursuing a “prepayment remedy” but chose instead to take the “apparent[ly] availab[le]” postpayment route. Id., at 443, 445. It does because due process prevents the government from engaging in a “ ‘bait and switch’ ” by later refusing to honor any remedial path it previously held open to the plaintiff. Reich v. Collins, 513 U. S. 106, 111 (1994).

The majority’s failure to supply a refund violates that rule. Start with the bait the government offered. As constitutional challenges like Hammons’s began trickling in, U. S. Trustees across the country urged courts against awarding injunctive relief or setoffs to parties contesting their disparate fee assessments. That kind of relief was unnecessary, the government contended, precisely “because the statute appropriating funds to the United States Trustee Program . . . permits refunds from the U. S. Trustee System Fund . . . according to standard procedures.” Memorandum of Law in Support of Defendants’ Motion for Summary Judgment in In re MF Global Holdings Ltd., No. 19–01379 (Bkrtcy. Ct. SDNY, Nov. 21, 2019), ECF Doc. 13, pp. 48–49. With representations like these—representations the government would repeat in Hammons’s own bankruptcy proceeding, see Part I–B, supra—who could doubt that the opportunity to seek a postpayment refund was anything less than “ ‘clear and certain’ ”? Reich, 513 U. S., at 111. Or that Hammons’s decision to choose this route rather than delay its plan confirmation to pursue a prepayment challenge was anything other than “reasonabl[e]”? Newsweek, 522 U. S., at 445.

Now the impermissible switch. Even as it continues to maintain that “[t]he amounts of the payments can be litigated . . . at any time,” Brief for Petitioner 5–6, the U. S. Trustee asks us to “declare, only after the disputed [fees] have been paid, that no such remedy exists,” Reich, 513 U. S., at 108. Try as litigants might, the government now insists, they cannot in fact secure “refunds from the U. S. Trustee System Fund” under any “procedures.” ECF Doc. 13, at 49.

That bait and switch violates due process, plain and simple. We should not be in the business of tolerating such “contrived and self-serving” changes in position. McKesson Corp. v. Division of Alcoholic Beverages and Tobacco, Fla. Dept. of Business Regulation, 496 U. S. 18, 42 (1990). Rather, our precedents “requir[e] the [government] to provide the remedy it has promised.” Alden v. Maine, 527 U. S. 706, 740 (1999); accord, Newsweek, 522 U. S., at 445; see McKesson, 496 U. S., at 31 (government “obligate[d]” to supply “meaningful backward-looking relief ”).

B

How does the majority answer this latest problem? On its telling, the only bait and switch our due process precedents guard against arises when the government holds open the possibility of a postpayment refund and then removes that option by statute or regulation after a party has paid the fee it wishes to contest. Ante, at 15–16. So, yes, the Trustee promised that litigants could pay now and litigate for a refund later. But, the majority insists, Hammons should have disregarded those representations and seized “the opportunity” always provided by statute to seek injunctive relief “before [it] paid” the challenged fees. Ante, at 15.[6]

This argument, too, misreads our precedents. The availability of “predeprivation remedies,” we have explained, is “beside the point” when a party reasonably relies on the apparent availability of a postpayment remedy. Reich, 513 U. S., at 113. Nor is it the case that an impermissible bait and switch can be accomplished only through statutory or regulatory changes. In Newsweek and Reich, for example, this Court held that a state-court decision violated due process by robbing the taxpayer of a postpayment remedy that appeared available until the court ruled otherwise. Newsweek, 522 U. S., at 443–445; Reich, 513 U. S., at 111–113. Indeed, Newsweek summarily reversed a lower court for “fail[ing] to consider” this point. 522 U. S., at 443. The case before us is therefore no different from those we’ve considered before, except in one respect: In Newsweek and Reich, this Court cured the lower courts’ due process violation; here, the Court itself creates one by robbing Hammons of a postpayment remedy that until this moment appeared available.

With nowhere left to go, the majority tries to suggest that our due process precedents are limited to the tax context. Ante, at 14. It’s the “[g]overnment’s exceedingly strong interest in” prompt tax payments, the majority reasons, that brings with it the “postdeprivation protections” discussed in our tax cases. Ibid. (internal quotation marks omitted). But the majority does not explain why, as a matter of due process, the government’s promises about the availability of postdeprivation procedures must be honored only in the tax context. Nor could it. If there’s anything unique about our tax decisions, it’s our treatment of “the field of taxation” as an area where we’ve “afforded [governments] great flexibility in satisfying the requirements of due process.” National Private Truck Council, Inc. v. Oklahoma Tax Comm’n, 515 U. S. 582, 587 (1995). In other words, we have long treated the procedural protections described in our tax cases as some of the most government-friendly due process will tolerate. See Londoner v. City and County of Denver, 210 U. S. 373, 385–386 (1908). And if a bait and switch is impermissible in the tax context, surely it must be in others.

This is hardly a new message. Reprimanding the Georgia Supreme Court for announcing there was no postpayment remedy only after the plaintiffs had paid a contested tax in reliance on that remedy, this Court in Reich explained that the case before it bore “a remarkable resemblance to NAACP v. Alabama ex rel. Patterson, 357 U. S. 449 (1958).” 513 U. S., at 112. And Patterson concerned a challenge to a state court’s contempt holding, not anything having to do with a tax. There, the Court held that, if “nothing ‘suggest[s]’ ” a particular procedural route “ ‘is the exclusive remedy,’ ” due process prohibits a government from later penalizing an individual for pursuing one available route rather than another. 513 U. S., at 113. Precisely the same reasoning and rule apply here—another inconvenient fact the majority prefers to ignore. See ante, at 15 (asserting there’s no “dispute” that McKesson and its progeny apply only to taxes). In choosing this path, however, the majority sends a clear message to lower courts and litigants: Next time the government asks you to hold off on pursuing a remedy on the promise you can always pursue it later, its representations are worth no more than the relief the Court awards Hammons today.[7]

IV

The government’s final salvo has to do with an appeal to public policy. Because there are fewer Administrator Program debtors who paid lower fees between 2018 and 2020 than there are Trustee Program debtors who paid arbitrarily higher fees during that period, the government reasons it is preferable either to try to recoup money from Administrator Program debtors or to do nothing at all. Brief for Petitioner 37–40. A refund to Trustee Program debtors, the government warns, would “transfe[r] to taxpayers substantial costs.” Reply Brief 2; see Brief for Petitioner 35. The majority echoes these concerns. Providing a refund, it says, would be “enormous[ly]” “disrupti[ve],” in part because reimbursing debtors in Trustee Program districts “would be expensive.” Ante, at 9.[8]

These concerns may be animated by prudent fiscal policy, but that is not how remedies work. Declining to pay an injured plaintiff will always be the cheapest option for the defendant. But when a refund is “otherwise available” as a matter of law, “the cost of [the] refund” cannot “justify a decision to withhold it.” McKesson, 496 U. S., at 51, n. 35. Consider how different, for example, our equality jurisprudence would look were it any other way. In the 1970s, pointing to the price tag associated with extending equal benefits to men and women was a favorite tactic of the federal government. See, e.g., Brief for Appellant in Weinberger v. Wiesenfeld, O. T. 1974, No. 73–1892, p. 22 (extending “ ‘mother’s benefits’ to fathers” might lead to “over $300 million” in costs, equivalent to many times more than that amount today). Should this Court have balked at the sticker price for remedying this “monetary disparity”? Ante, at 7. More recently, the government argued that a “damages remedy against federal employees” for religious discrimination was too costly to count as “ ‘appropriate relief,’ ” Brief for Petitioners in Tanzin v. Tanvir, O. T. 2020, No. 19–71, p. 30, even though damages were “the only form of relief that [could] remedy some . . . violations,” Tanzin, 592 U. S., at 51. Should we have stopped to perform a cost-benefit analysis there, too?[9]

V

I struggle to understand why today the majority so readily dismisses any remedy in this case—all to save the government from the trouble of issuing funds the Legislature has appropriated and the Executive has promised to pay. As I see it, two possible lines of thinking may explain this unusual outcome, neither reassuring.

One possibility is that the majority views Bankruptcy Clause violations as less worthy of relief than other constitutional violations. The majority nods in that direction when it compares today’s decision to others involving what it calls “far more serious dignitary harms.” Ante, at 11. But if that’s the reason, it is hardly a convincing one. After all, the majority describes its “What would Congress have done?” approach to remedies as universally applicable—governing questions of retrospective relief in sex discrimination and free exercise cases no less than those arising under the Bankruptcy Clause. See ante, at 7. Nor do we as judges have any warrant to play favorites among the Constitution’s provisions, exalting some while relegating others to the status of “a second-class right.” New York State Rifle & Pistol Assn., Inc. v. Bruen, 597 U. S. 1, 70 (2022) (internal quotation marks omitted).

The other possibility is no better. Perhaps the majority thinks supplying relief isn’t worth the trouble because the constitutional violation at issue here was, as the majority puts it, “short-lived and small.” Ante, at 12. After all, the violation began in 2018 and ended in 2020. But on what account does a multiyear violation of the Constitution count as “short-lived”? And how does that violation count as “small” when it cost Hammons $2.5 million and, as the majority itself emphasizes, cost others millions more? Cf. Culley v. Marshall, 601 U. S. 377, 411–412 (2024) (Sotomayor, J., joined by Kagan and Jackson, JJ., dissenting) (months-long deprivation of a car is a harm of constitutional proportions); Wellness Int’l Network, Ltd. v. Sharif, 575 U. S. 665, 703 (2015) (Roberts, C. J., dissenting) (insisting there is no “ ‘de minimis’ ” exception for constitutional “incursion[s]”). Consider, too, what that kind of thinking could mean for those seeking retrospective relief for other constitutional violations. It’s not hard to imagine today’s decision receiving a warm welcome from those who seek to engage in only a dash of discrimination or only a brief denial of some other constitutionally protected right. The rest of us can only hope that the Court corrects its mistake before it metastasizes too far beyond the bankruptcy context.[10]

Respectfully, I dissent.