Before Willett, Engelhardt, and Wilson, Circuit Judges. Cory T. Wilson, Circuit Judge: “An elective despotism was not the government we fought for; but one which should not only be founded on free principles, but in which the powers of government should be so divided and balanced . . . , as that no one could transcend their legal limits, without being effectually checked and restrained by the others.” The Federalist No. 48 (J. Madison) (quoting Thomas Jefferson’s Notes on the State of Virginia (1781)). In particular, as George Mason put it in Philadelphia in 1787, “[t]he purse & the sword ought never to get into the same hands.” 1 The Records of the Federal Convention of1787, at 139–40 (M. Farrand ed. 1937). These foundational precepts of the American system of government animate the Plaintiffs’ claims in this action. They also compel our decision today. Community Financial Services Association ofAmerica and Consumer Service Alliance of Texas (the “Plaintiffs”) challenge the validity of the Consumer Financial Protection Bureau’s 2017 Payday Lending Rule. The Plaintiffs contend that in promulgating that rule, the Bureau acted arbitrarily and capriciously and exceeded its statutory authority. They also contend that the Bureau is unconstitutionally structured, challenging the Bureau Director’s insulation from removal, Congress’s broad delegation of authority to the Bureau, and the Bureau’s unique, double-insulated funding mechanism. The district court rejected these arguments. Weagree that, for the most part, the Plaintiffs’ claims miss their mark. But one arrow has found its target: Congress’s decision to abdicate its appropriations power under the Constitution, i.e., to cede its power of the purse to the Bureau, violates the Constitution’s structural separation of powers. We thus reverse the judgment of the district court, render judgment in favor of the Plaintiffs, and vacate the Bureau’s 2017 Payday Lending Rule. I. A. In response to the 2008 financial crisis, Congress enacted the Consumer Financial Protection Act, 12 U.S.C. §§ 5481–5603. The Act created the Bureau as an independent regulatory agency housed within the Federal Reserve System. See id. § 5491(a). The Bureau is charged with “implement[ing]” and “ enforce[ing]” consumer protection laws to “ ensur[e] that all consumers have access to markets for consumer financial products and services” that “are fair, transparent, and competitive.” Id. § 5511(a). Congress transferred to the Bureau administrative and enforcement authority over 18 federal statutes which prior to the Act were overseen by seven different agencies. See id. §§ 5512(a), 5481(12), (14). Those statutes “cover everything from credit cards and car payments to mortgages and student loans.” Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2200 (2020). In addition, Congress enacted a sweeping new proscription on “any unfair, deceptive, or abusive act or practice” by certain participants in the consumer-finance industry. 12 U.S.C. § 5536(a)(1)(B). “Congress authorized the [Bureau] to implement that broad standard (and the 18 pre-existing statutes placed under the agency’s purview) through binding regulations.” Seila Law, 140 S. Ct. at 2193 (citing 12 U.S.C. §§ 5531(a)–(b), 5581(a)(1)(A), (b)). Congress placed the Bureau’s leadership under a single Director to be appointed by the President with the advice and consent of the Senate. 12 U.S.C. § 5491(b)(1)–(2). The Director serves a term of five years, with the potential of a holdover period pending confirmation of a successor. Id. § 5491(c)(1)–(2). The Act originally limited the President’s ability to remove the Director, id. § 5491(c)(3), but the Supreme Court invalidated that provision while this litigation was pending, see Seila Law, 140 S. Ct. at 2197. The Director is vested with authority to “prescribe rules and issue orders and guidance, as may be necessary or appropriate to enable the Bureau to administer and carry out the purposes and objectives of the Federal consumer financial laws, and to prevent evasions thereof.” 12 U.S.C. § 5512(b)(1). This includes rules “identifying as unlawful unfair, deceptive, or abusive acts or practices” committed by certain participants in the consumer-finance industry. Id. § 5531(b). The Bureau’s funding scheme is unique across the myriad independent executive agencies across the federal government. It is not funded with periodic congressional appropriations. “Instead, the [Bureau] receives funding directly from the Federal Reserve, which is itself funded outside the appropriations process through bank assessments.” Seila Law, 140 S. Ct. at 2194. Each year, the Bureau simply requests an amount “determined by the Director to be reasonably necessary to carry out the” agency’s functions. Id. § 5497(a)(1). The Federal Reserve must then transfer that amount so long as it does not exceed 12% of the Federal Reserve’s “total operating expenses.” Id. § 5497(a)(1)–(2). For the first five years of its existence (i.e., 2010–2014), the Bureau was permitted to exceed the 12% cap by $200 million annually so long as it reported the anticipated excess to the President and congressional appropriations committees. Id. § 5497(e)(1)–(2). B. In 2016, Director Richard Cordray, who was appointed by President Barack Obama, proposed a rule to regulate payday, vehicle title, and certain high-cost installment loans (the “Payday Lending Rule”). After a public notice-and-comment period, Director Corday finalized the Payday Lending Rule in November 2017, during the first year of the Trump administration. See Payday, Vehicle Title, and Certain High-Cost Installment Loans, 82 Fed. Reg. 54472 (Nov. 17, 2017). The rule became effective on January 16, 2018, and had a compliance date of August 19, 2019. Id. The Rule had two major components, each limiting a practice the Bureau deemed “unfair” and “abusive.” See id. First, the “Underwriting Provisions” prohibited lenders from making covered loans “without reasonably determining that consumers have the ability to repay the loans according to their terms.” 12 C.F.R. § 1041.4 (2018); 82 Fed. Reg. at 54472. The Underwriting Provisions have since been repealed and are not at issue in this appeal. See 85 Fed. Reg. 44382 (July 22, 2019). Second, and relevant here, the “Payment Provisions” limit a lender’s ability to obtain loan repayments via preauthorized account access. See 12 C.F.R. § 1041.8. The Bureau determined that absent a new and specific authorization, it is “unfair and abusive” for lenders to attempt to withdraw payments for covered loans from consumers’ accounts after two consecutive withdrawal attempts have failed due to a lack of sufficient funds. Id. § 1041.7; 82 Fed. Reg. at 54472. The Payment Provisions accordingly prohibit lenders from initiating additional payment transfers from consumers’ accounts after two consecutive attempts have failed for insufficient funds unless “the additional payment transfers are authorized by the consumer.” 12 C.F.R. § 1041.8(b)(1), (c)(1). The Payment Provisions cast a wide net. So long as the purpose of the attempted transfer is to collect payment due on a covered loan, the two-attempt limit applies to “any lender-initiated debt or withdrawal of funds from a consumer’s account.” Id. § 1041.8(a)(1). This includes checks, debit and prepaid card transfers, preauthorized electronic fund transfers, and remotely created payment orders. See id.; 82 Fed. Reg. at 54910. In April 2018, the Plaintiffs sued the Bureau on behalf of payday lenders and credit access businesses, seeking an “order and judgment holding unlawful, enjoining, and setting aside” the Payday Lending Rule. The Plaintiffs alleged that the rule exceeded the Bureau’s statutory authority and otherwise violated the Administrative Procedure Act (APA). They further alleged that the rule was invalid because the Act’s for-cause removal provision, self-funding mechanism, and delegation of rulemaking authority each violated the Constitution’s separation of powers. Around this time, the Bureau, now led by Acting Director Mick Mulvaney, announced that it intended to engage in notice-and-comment rulemaking to reconsider the Payday Lending Rule. Due to that ongoing effort, the parties filed a joint request to stay both the litigation and the rule’s effective date. The district court entered a stay pending further order of the court. Cmty. Fin. Servs. Ass’n of Am., Ltd. v. CFPB, 2018 WL 6252409, at *2 (W.D. Tex. Nov. 6, 2018). While the Bureau engaged in rulemaking, President Trump nominated and the Senate confirmed Kathleen Kraninger as Director, replacing Acting Director Mulvaney. In early 2019, the Bureau issued a proposed rule rescinding the Underwriting Provisions but leaving the Payment Provisions intact. 84 Fed. Reg. 4252. In July 2020, following the Supreme Court’s decision in Seila Law, the Bureau finalized its revised rule. 85 Fed. Reg. 44382. The Bureau simultaneously issued a separate “Ratification,” in which it “affirm[ed] and ratifie[d] the [P]ayment [P]rovisions of the 2017 [Payday Lending] Rule.” 85 Fed. Reg. 41905-02. In August 2020, the district court lifted the stay, and the Plaintiffs amended their complaint to challenge, among other things, the Bureau’s ratification of the Payment Provisions. Thereafter, the parties filed cross-motions for summary judgment. The district court granted summary judgment for the Bureau on each of the Plaintiffs’ claims. Cmty. Fin. Servs. Ass’n of Am., Ltd. v. CFPB, 558 F. Supp. 3d 350 (W.D. Tex. 2021). The court concluded, inter alia, that: (1) the promulgating Director’s insulation from removal did not render the Payment Provisions void ab initio, id. at 358; (2) the Bureau’s “ratification of the Payment Provisions was a solution tailored to the constitutional injury sustained by the [Plaintiffs],” id. at 365; (3) the “Payment Provisions [were] consistent with the Bureau’s statutory authority and not arbitrary and capricious,” id.; (4) the Bureau’s self-funding mechanism did not violate the Appropriations Clause because it was expressly authorized by statute, id. at 367; and (5) there was no nondelegation issue because the Bureau was vested with an “intelligible principle” to guide its discretion, id. The Plaintiffs now appeal. We allowed the Third-Party Payment Processors Association, a national non-profit association of payment processors and their banks, to appear as amicus curiae in support of the Plaintiffs’ arbitrary-and-capricious challenge. II. We”review a district court’s judgment on cross motions for summary judgment de novo, addressing each party’s motion independently, viewing the evidence and inferences in the light most favorable to the nonmoving party.” Morgan v. Plano Indep. Sch. Dist., 589 F.3d 740, 745 (5th Cir. 2009). Summary judgment is appropriate “if the movant shows that there is no genuine dispute as to any material fact and themovant is entitled to judgment as a matter of law.” Fed. R. Civ. P. 56(a). Constitutional issues are also reviewed de novo. Huawei Techs. USA, Inc. v. FCC, 2 F.4th 421, 434 (5th Cir. 2021). The Plaintiffs raise four overarching issues on appeal. They contend that the Payment Provisions of the Payday Lending Rule are invalid because: (1) the rule’s promulgation violated the APA; (2) the rule was promulgated by a Director unconstitutionally insulated from presidential removal; (3) the Bureau’s rulemaking authority violates the nondelegation doctrine; and (4) the Bureau’s funding mechanism violates the Appropriations Clause of the Constitution. We address each argument in turn. A. The APA instructs courts to “hold unlawful and set aside agency action[s]” that are “ arbitrary, capricious,an abuse of discretion, or otherwise not in accordance with law,” or “in excess of statutory jurisdiction, authority, or limitations.” 5 U.S.C. § 706(2). The Plaintiffs lodge two arguments under the APA. First, they contend that the Bureau exceeded its statutory authority by declaring more than two successive preauthorized withdrawals to be “unfair” and “abusive.” Second, they assert that the Payment Provisions are arbitrary and capricious in their entirety or, alternatively, as applied to two specific contexts—installment loans and debit and prepaid card payments. 1. The Act grants the Bureau broad authority to prescribe rules prohibiting “unfair, deceptive, or abusive actsor practicesin connection with any transaction with a consumer for a consumer financial product or service, or the offering of a consumer financial product or service.” 12 U.S.C. § 5531(b). This authority is not without limitation, however. Congress included specific definitions that govern when an act or practice may be deemed “unfair,” id. § 5531(c)(1), or “abusive,” id. § 5531(d). And unless those definitions are met, the Bureau “shall have no authority” to regulate conduct on either ground. See id. § 5531(c)–(d). In devising the Payment Provisions, the Bureau assessed the statutory definitions and determined that it was both “unfair” and “abusive” for lenders to attempt additional withdrawals from consumers’ accounts after two consecutive attempts failed due to insufficient funds unless the lender acquired “new and specific authorization.” 12 C.F.R. § 1041.7; see also 82 Fed. Reg. at 54472. The Plaintiffs assert that the Bureau lacked authority to regulate the number of unsuccessful withdrawal attempts because this practice falls outside the Act’s definitions of “unfair” and “abusive.” Our review begins (and ends) with unfairness.[1] Under the Act, an act or practice is “unfair” if “the Bureau has a reasonable basis to conclude that [1] the act or practice causes or is likely to cause substantial injury to consumers [2] which is not reasonably avoidable by consumers; and [3] such substantial injury is not outweighed by the countervailing benefits to consumers or to competition.” 12 U.S.C. § 5531(c)(1). The Bureau evaluated each element in its 2017 rulemaking record and concluded that the proscribed practice satisfied all three. The Plaintiffs challenge only the first two elements on appeal. As to the first, the Bureau determined that lenders’ excessive withdrawal attempts cause or are likely to cause consumers substantial injury in the form of repeated fees, including insufficient fund fees, overdraft fees, and lender-imposed return fees. 82 Fed. Reg. at 54732–34. It also found that “consumers who experience two or more consecutive failed lender payment attempts appear to be at greater risk of having their accounts closed by their account-holding institution.” Id. at 54734. The Plaintiffs do not dispute the occurrence or substantiality of these injuries. Rather, they challenge the Bureau’s finding that the proscribed practice either causes or is likely to cause them. The Plaintiffs assert that “[c]onsumers’ banks—not lenders— cause failed-payment fees or bank-account closures” because they are the ones who “impose, collect, or otherwise control [them].” We are unpersuaded. The presence of an “independent causal agent[]” does not “ erase the role” lenders play in bringing about the contemplated harm. FTC v. Neovi, Inc., 604 F.3d 1150, 1155 (9th Cir. 2010). Though not the “most proximate cause,” a lender’s repeated initiation of unsuccessful payment transfers is both a but-for and a proximate cause of any resulting fees or closures. FTC v. Wyndham Worldwide Corp., 799 F.3d 236, 246 (3d Cir. 2015) (“[The fact] that a company’s conduct was not the most proximate cause of an injury generally does not immunize liability from foreseeable harms.”). The Plaintiffs also challenge the Bureau’s finding that these injuries are not reasonably avoidable by consumers. Few courts have meaningfully addressed this second element of “unfairness” under the Act. E.g., CFPB v. Navient Corp., No. 3:17-CV-101, 2017 WL 3380530, at *20–21 (M.D. Pa. Aug. 4, 2017); CFPB v. D & D Mktg., No. CV 15-9692, 2016 WL 8849698, at *10 (C.D. Cal. Nov. 17, 2016); CFPB v. ITT Educ. Servs., Inc., 219 F. Supp. 3d 878, 916–17 (S.D. Ind. 2015). In doing so, these courts relied on our sister circuits’ interpretations of “reasonably avoidable” from the analogous standard in the Federal Trade Commission Act (FTCA). See 15 U.S.C. § 45(n).[2] We do the same.[3] To determine whether an injury was “reasonably avoidable” under the FTCA, courts generally “look to whether the consumers had a free and informed choice.” Neovi, 604 F.3d at 1158; accord Am. Fin. Servs. Ass’n v. FTC, 767 F.2d 957, 976 (D.C. Cir. 1985). “An injury is reasonably avoidable if consumers ‘have reason to anticipate the impending harm and the means to avoid it,’ or if consumers are aware of, and are reasonably capable of pursuing, potential avenues toward mitigating the injury after the fact.” Davis v. HSBC Bank Nev., N.A., 691 F.3d 1152, 1168–69 (9th Cir. 2012) (quoting Orkin Exterminating Co. v. FTC, 849 F.2d 1354, 1365–66 (11th Cir. 1988)). The Plaintiffs contend that consumers can reasonably avoid injury associated with successive withdrawal attempts by (1) “not authorizing automatic withdrawals,” (2) “sufficiently funding [their] account[s],” (3) “negotiating revised payment options,” (4) “invoking [their] rights under federal law to issue stop-payment orders or rescind account access,” or (5) “declining to take out the loan” and “pursuing alternative[] sources of credit.” Each of these concerns was raised during the public comment period of the Bureau’s rulemaking process. See, e.g., 82 Fed. Reg. at 54736–37. The Bureau found none of them sufficient to constitute a reasonable means of avoiding injury. Id. at 54737. The rulemaking record prefaces that many borrowers resort to payday loans because they are in financial distress and lack other viable options for financing. Id. at 54571, 54735. Addressing the Plaintiffs’ first point, the Bureau explained that since “leveraged payment mechanisms” are “a central feature of these loans,” borrowers typically do not have the ability to shop for loans without them. Id. at 54737. The Bureau also found that simply funding their accounts is not a reasonable means for borrowers to avoid injury because “[m]any borrowers [do] not have the funds” after two unsuccessful withdrawal attempts, and “subsequent [withdrawals] can occur very quickly, often on the same day, making it difficult to ensure funds are in the right account before the [next withdrawal] hits.” Id. For the same reason, the Bureau found negotiating repayment options to be too slow a solution to mitigate against fees incurred on additional withdrawal attempts. See id. at 54736–37. Regarding the Plaintiffs’ fourth point, the Bureauexplained that costs, “[c]omplexities in payment processing systems[,] and the internal procedures of consumers’ account-holding institutions, combined with lender practices, often make it difficult for consumers to stop payment or revoke authorization effectively.” Id. Finally, the Bureau concluded that “the suggestion that a consumer can simply decide not to participate in the market is not . . . a valid means of reasonably avoiding the injury.” Id. at 54737. By that logic, the Bureau reasoned, “no market practice could ever be determined to be unfair.” Id. The Bureau’s explanations are fully fleshed out in the Payday Lending Rule’s 519-page rulemaking record, where they are supported by a variety of data and industry-related studies. Reviewing that record as it undergirds the Payment Provisions, we find the Bureau had “a reasonable basis to conclude” that the harms associated with three or more unsuccessful withdrawal attempts are “not reasonably avoidable by consumers.” 12 U.S.C. § 5531(c)(1). Because the proscribed practice thus satisfies the elements of an “unfair” practice under the Act, we conclude that the Bureau acted within its statutory authority in promulgating the Payment Provisions. 2. Next, the Plaintiffs contend that the Payment Provisions are arbitrary and capricious, either as a whole or as applied. “The APA’s arbitrary-and-capricious standard requires that agency action be reasonable and reasonably explained. Judicial review under that standard is deferential, and a court may not substitute its own policy judgment for that of the agency.” FCC v. Prometheus Radio Project, 141 S. Ct. 1150, 1158 (2021). Still, we must ensure that an agency “examine[s] the relevant data and articulate[s] a satisfactory explanation for its action including a rational connection between the facts found and the choice made.” Motor Vehicle Mfrs. Ass’n of U.S., Inc. v. State Farm Mut. Auto. Ins. Co., 463 U.S. 29, 43 (1983) (quotation omitted). A rule is arbitrary and capricious if the agency relied on “impermissible factors, failed to consider important aspects of the problem, offered an explanation for its decision that is contrary to the record evidence, or is so irrational that it could not be attributed to a difference in opinion or the result of agency expertise.” BCCA Appeal Grp. v. U.S. EPA, 355 F.3d 817, 824 (5th Cir. 2003). Here, the Plaintiffs first contend that the Payment Provisions are arbitrary and capricious in their entirety because they rest on stale data from four-to-five years prior to their promulgation, and the Bureau failed to consider the provisions’ important countervailing effects. As to the first point, the Plaintiffs forfeited their stale data argument by failing to raise it in the district court. See Rollins v. Home Depot USA, Inc., 8 F.4th 393, 398 (5th Cir. 2021). And forfeiture aside, the Bureau offered a reasoned explanation in its 2017 rulemaking record for relying on data collected from 2011–2012. See 82 Fed. Reg. at 54722, 54729. As to the second point, the only countervailing effect the Plaintiffs allege the Bureau failed to consider is “the increased likelihood that a loan will enter into collections sooner than it would have (if it would have at all).” But the Bureau persuasively responds that “[i]f the borrower is unable to obtain the funds, it is unclear why the borrower (or the lender) would be better off if the lender could initiate failed withdrawal attempts—and, in the process, pile additional fees onto the borrower—before the loan enters collections.” Even if the Payment Provisions’ limit on repeated withdrawal attempts might send some loans to collections sooner, that possibility is not so “important” that the Bureau had to consider it specifically. See Motor Vehicle Mfrs., 463 U.S. at 43 (explaining “an agency rule would be arbitrary and capricious if the agency . . . entirely failed to consider an important aspect of the problem”). Turning to their as-applied challenge, the Plaintiffs assert that the Payment Provisions are arbitrary and capricious as applied to debit and prepaid card payments and as to separate installments of multi-payment installment loans. Amicus joins them with respect to debit and prepaid cards. Together, they contend that the Payment Provisions “arbitrarily treat[] debit and prepaid card payments the same as check and [account clearinghouse] payments, even though the former do not give rise to the fees that, in the Bureau’s assessment, justify the Rule.” The Bureau acknowledged in the rulemaking record that debit and prepaid card transactions “present somewhat less risk of harm to consumers,” but it declined to exclude them for several reasons. 82 Fed. Reg. at 54750. For one, theBureaufound thatthough failed debit and prepaid card transactions may not trigger insufficient fund fees, “some of them do trigger overdraft fees, even after two failed attempts.” Id. And as with other payment-transfer methods, consumers would still be subject to “return payment fees and late fees charged by lenders.” Id. at 54723, 54734. The Bureau also explained that a carve out for these transactions “would be impracticable to comply with and enforce.” Id. at 54750. These considerations suffice to establish a “rational connection between the facts found and choice made.” Motor Vehicle Mfrs., 463 U.S. at 43 (quotation omitted). Therefore, the Payment Provisions are not arbitrary and capricious as applied to debit and prepaid card transfers.[4] Similarly, we cannot say that the Bureau acted arbitrarily and capriciously by extending the Payment Provisions’ two-attempt limit across all scheduled installment payments on the same loan. The Plaintiffs contend that the Bureau failed to support its decision with “reasoned analysis or record evidence.” But again, the rulemaking recordprovesotherwise. Citing its own study, the Bureau explained that a third withdrawal attempt, even as applied to a different scheduled payment, would still likely fail “even if two weeks or a month has passed.” 82 Fed. Reg. at 54753. The Bureau also found that “the tailoring of individualized requirements for each discrete payment practice would add considerable complexity to the rule.” Id. Further, the Bureau determined that distinguishing between re-presentments of the same payment and new presentments for new installments would invite evasion by lenders. The Bureau referenced a rule imposed by the National Automated Clearinghouse Association (NACHA), a self-governing private organization, that is similar to the Payment Provisions (except that it only applies after three attempts). See id. at 54728–29. The Bureau noted that the NACHA rule’s distinction between attempts to collect a new payment and re-initiation of a prior one had led companies to manipulate data fields so that it would appear as if a withdrawal attempt was for a new installment. See id. at 54728 n.985 & 54729. In sum, we conclude that the Payment Provisions are not arbitrary and capricious, either in their entirety or in their two contested applications. As Plaintiffs fail to show that the Payday Lending Rule’s promulgation violated the APA, summary judgment in favor of the Bureau on this claim was warranted. B. The Plaintiffs next contend that the Payment Provisions must be invalidated because the Payday Lending Rule was initially promulgated by a director who was unconstitutionally shielded from removal. 1. The Act states that the Bureau’s Director may be removed only “for inefficiency, neglect of duty, or malfeasance in office.” 12 U.S.C. § 5491(c)(3). In Seila Law, the Court held that this limitation on the President’s removal power violated the Constitution’s separation of powers. 140 S. Ct. at 2197. But the Court declined to find that the Director’s unconstitutional insulation from removal rendered the remainder of the Act invalid. Id. at 2208–11. Instead, the Court concluded that the infirm removal provision was severable and remanded the case for a determination of the appropriate relief. Id. at 2211. Like Seila Law, Collins v. Yellen, 141 S. Ct. 1761 (2021), involved a challenge to actions taken by an independent agency, the Federal Housing Finance Agency (FHFA), that was headed by a single officer removable only for cause. See 141 S. Ct. at 1784. The Collins petitioners asserted that the FHFA Director’s for-cause removal protection violated the separation of powers, and therefore the agency actions at issue “must be completely undone.” Id. at 1787. The Court agreed that the for-cause removal provision was unconstitutional, finding Seila Law “all but dispositive.” Id. at 1783. But it refused to hold that an officer’s insulation from removal, by itself, rendered all agency action taken under that officer void. Id. at 1787–88. Unlike cases “involv[ing] a Government actor’s exercise of power that the actor did not lawfully possess,” the Court explained, a properly appointed officer’s insulation from removal “does not strip the [officer] of the power to undertake the other responsibilities of his office.” Id. at 1788 & n.23. Thus, to obtain a remedy, the challenging party must demonstrate not only that the removal restriction violates the Constitution but also that “the unconstitutional removal provision inflicted harm.” Id. at 1788–89. While the Plaintiffs acknowledge Collins, they argue the case is distinguishable on several grounds. None are persuasive. First, they assert that Collins applies only to retrospective relief. But Collins did not rest on a distinction between prospective and retrospective relief. As the Sixth Circuit recently explained, Collins‘s remedial inquiry “focuse[d] on whether a ‘harm’ occurred that would create an entitlement to a remedy, rather than the nature of the remedy, and our determination as to whether an unconstitutional removal protection ‘inflicted harm’ remains the same whether the petitioner seeks retrospective or prospective relief.” Calcutt v. FDIC, 37 F.4th 293, 316 (6th Cir. 2022).[5] The Plaintiffs also contend that Collins “does not apply to rulemaking challenges.” This distinction is similarly without a difference. To the contrary, in Collins, the Court explicitly stated that “the unlawfulness of the removal provision does not strip the Director of the power to undertake the other responsibilities of his office.” 141 S. Ct. at 1788 n.23. Because the Bureau’s Director’s “other responsibilities” include rulemaking, see 12 U.S.C. §§ 5511(a), 5512(b), Collins is directly on point, and the Plaintiffs must demonstrate that the unconstitutional removal provision caused them harm. 2. Joining the issue, the Plaintiffs assert that “even if Collins does inform the analysis here, its framework plainly requires setting aside the [Payment Provisions]” because the Plaintiffs have made a sufficient showing of harm. As noted above, after Collins, a party challenging agency action must show not only that the removal restriction transgresses the Constitution’s separation of powers but also that the unconstitutional provision caused (or would cause) them harm. 141 S. Ct. at 1789. The Court chose to remand Collins‘s remedy question and stopped short of articulating a precise statement as to how a party may prove harm. See id. at 1788–89. Instead, the Collins majority concluded with several hypotheticals: Although an unconstitutional provision is never really part of the body of governing law (because the Constitution automatically displaces any conflicting statutory provision from the moment of the provision’s enactment), it is still possible for an unconstitutional provision to inflict compensable harm. And the possibility that the unconstitutional restriction on the President’s power to removea Director . . . could have such an effect cannot be ruled out. Suppose, for example, that the President had attempted to remove a Director but was prevented from doing so by a lower court decision holding that he did not have “cause” for removal. Or suppose that the President had made a public statement expressing displeasure with actions taken by a Director and had asserted that he would remove the Director if the statute did not stand in the way. In those situations, the statutory provision would clearly cause harm. Id. We distill from these hypotheticals three requisites for proving harm: (1) a substantiated desire by the President to remove the unconstitutionally insulated actor, (2) a perceived inability to remove the actor due to the infirm provision, and (3) a nexus between the desire to remove and the challenged actions taken by the insulated actor. This is borne out by the concurring Justices’ opinions as well. See id. at 1792–93 (Thomas, J., concurring); id. at 1801 (Kagan, J., concurring in part); id. at 1803 n.1 (Sotomayor, J., concurring in part and dissenting in part). As Justice Kagan emphasized, “plaintiffs alleging a removal violation are entitled to injunctive relief—a rewinding of agency action—only when the President’s inability to fire an agency head affected the complained-of decision.” Id. at 1801 (Kagan, J., concurring in part) (emphasis added). It is thus not enough, as the Plaintiffs would have us hold, for a challenger to obtain relief merely by establishing that the unconstitutional removal provision prevented the President from removing a Director he wished to replace. As we read Collins, to demonstrate harm, the Plaintiffs must show a connection between the President’s frustrated desire to remove the actor and the agency action complained of. See id. at 1789. Without this showing, the Plaintiffs could put themselves in a better place than otherwise warranted, by challenging decisions either with which the President agreed, or of which he had no awareness at all. Id. at 1802 (Kagan, J., concurring in part). Applying Collins‘s framework, we conclude the Plaintiffs fail to show that the Act’s removal provision inflicted a constitutional harm. Though they state “[i]t is uncontested that, but for the later-invalidated removal restriction, President Trump would have replaced [Director] Cordray before he finalized the [Payday Lending Rule],” their only support for this assertion consists of a few carefully selected statements from Director Cordray’s book, see, e.g., Richard Cordray, Watchdog: How Protecting Consumers Can Save Our Families, Our Economy, and Our Democracy185 (2020) (“[T]he threat that I would be fired as soon as President Trump took office loomed over everything.”), and an online article, see Kate Berry, In Tell-All, Ex-CFPB Chief Cordray Claims Trump Nearly Fired Him, American Banker (Feb. 27, 2020) https://www. americanbanker.com/news/in-tell-all-ex-cfpb-chief-cordrayclaims-trump-nearly-fired-him (stating “President Trump was advised to hold off on firing Corday because the Supreme Court had not yet weighed in on [the] ‘for cause’ provision”). These secondhand accounts of President Trump’s supposed intentions are insufficient to establish harm. The Director’s subjective belief that his firing might be imminent does not in itself substantiate that the President would have removed the Director but for the unconstitutional removal provision. Regardless, the record before us plainly fails to demonstrate any nexus between the President’s purported desire to remove Cordray and the promulgation of the Payday Lending Rule or, specifically, the Payment Provisions. In short, nothing the Plaintiffs proffer indicates that, but for the removal restriction, President Trump would have removed Cordray and that the Bureau would have acted differently as to the rule. Because the Plaintiffs have failed to demonstrate harm, we need not address the Bureau’s alternative argument that any alleged harm was cured by Director Kraninger’s ratification of the Payment Provisions. See CFPB v. CashCall, Inc., 35 F.4th 734, 743 (9th Cir. 2022) (finding “it unnecessary to consider ratification” where the challenger could not establish harm). Summary judgment in favor of the Bureau on this claim was proper. C. We next consider the Plaintiffs’ argument that the Bureau’s rulemaking authority violates the Constitution’s separation of powers by running afoul of the nondelegation doctrine.[6] The Constitution provides that “[a]ll legislative Powers herein granted shall be vested in a Congress of the United States.” U.S. Const. art. I, § 1. Inherent in “that assignment of power to Congress is a bar on its further delegation.” Gundy v. United States, 139 S. Ct. 2116, 2123 (2019) (plurality opinion). “Under the nondelegation doctrine, Congress may not constitutionally delegate its legislative power to another branch of government.” United States v. Jones, 132 F.3d 232, 239 (5th Cir. 1998) (citing Mistretta v. United States, 488 U.S. 361, 372 (1989)). But the Supreme Court has long delimited this general principle: “So long as Congress ‘lay[s] down by legislative act an intelligible principle to which the person or body authorized to [act] is directed to conform, such legislative action is not a forbidden delegation of legislative power.’” Touby v. United States, 500 U.S. 160, 165 (1991) (quoting J.W. Hampton, Jr., & Co. v. United States, 276 U.S. 394, 409 (1928)). It is “constitutionally sufficient if Congress clearly delineates the general policy, the public agency which is to apply it, and the boundaries of this delegated authority.” Am. Power & Light Co. v. SEC, 329 U.S. 90, 105 (1946); see also Gundy, 139 S. Ct. at 2129 (explaining that “[t]hose standards . . . are not demanding”). Through the Act, Congress gave the Bureau authority “to prescribe rules . . . identifying as unlawful unfair, deceptive, or abusive acts or practices.” 12 U.S.C. § 5531(b). This constituted a delegation of legislative power because “the lawmaking function belongs to Congress.” Loving v. United States, 517 U.S. 748, 758 (1996). The question is whether Congress also “supplied an intelligible principle to guide the [Bureau's] discretion.” Gundy, 139 S. Ct. at 2123. The Plaintiffs assert that “[t]here is no intelligible principle” behind the Bureau’s “vague and sweeping” rulemaking authority. We disagree. In the Act, Congress articulated its general policy preferences, established the Bureau as the agency to apply them, and set boundaries—albeit broad ones— on the Bureau’s rulemaking authority. Am. Power & Light Co., 329 U.S. at 105. Given that the Supreme Court “has over and over upheld even very broad delegations,” Gundy, 139 S. Ct. at 2129, the Act’s delegation of rulemaking authority to the Bureau passes muster. Congress’s general policy is distilled in the Bureau’s purpose and objectives. 12 U.S.C. § 5511(a)–(b). The Bureau’s “purpose” is “to implement and, where applicable, enforce Federal consumer financial law consistently for the purpose of ensuring that all consumers have access to markets for consumer financial products and services and that markets for consumer financial products and services are fair, transparent, and competitive.” Id. § 5511(a). That purpose is accompanied by five “objectives” toward which “[t]he Bureau is authorized to exercise its authorit[y.]” Id. § 5511(b). One of those is to “ensur[e] that . . . consumers are protected from unfair, deceptive, or abusive acts and practices.” Id. § 5511(b)(2). In line with that objective, Congress empowered the Bureau to “prescribe rules applicable to a covered person or service provider identifying as unlawful unfair, deceptive, or abusive acts or practices in connection with any transaction with a consumer for a consumer financial product or service, or the offering of a consumer financial product or service.” Id. § 5531(b). Congress then circumscribed that authority by including specific criteria that must be met before the Bureau can label a practice “unfair” or “abusive.” See id. § 5531(c)–(d).[7] Far from an “open-ended delegation” that offers “no guidance whatsoever,” Jarkesy v. SEC, 34 F.4th 446, 462 (5th Cir. 2022) (emphasis omitted), Congress’s grant of rulemaking authority to the Bureau was accompanied by a specific purpose, objectives, and definitions to guide the Bureau’s discretion. This was more than sufficient to confer an “intelligible principle.” See Whitman v. Am. Trucking Ass’n, 531 U.S. 457, 474–75 (2001) (compiling the various directives the Supreme Court has deemed sufficient to constitute an “intelligible principle”). D. Finally, the Plaintiffs contend that the Payday Lending Rule is invalid because the Bureau’s funding structure violates the Appropriations Clause of the Constitution and the separation of powers principles enshrined in it. Though the constitutionality of the Bureau has been heavily litigated, this issue has yet to be definitively resolved. In Seila Law, the Supreme Court determined that the Act’s presidential removal restriction violated the Constitution’s separation of powers, but the Court did not confront whether the Bureau’s unique funding scheme does. 140 S. Ct. at 2197. And a majority of this court recently concluded that the issue was not properly before us in another case challenging the Bureau’s structure and authority. See CFPB v. All Am. Check Cashing, Inc., 33 F.4th 218, 220 & n.2 (5th Cir. 2022)(en banc). However, Judge Jones, in a magisterial separate opinion joined by several of our colleagues, disagreed and addressed the parties’ Appropriations Clause challenge. See id. at 221 (Jones, J., concurring). Methodically analyzing the question, she concluded that the Bureau’s funding mechanism contravenes the Constitution’s separation of powers. Id. at 242. The issue is squarely raised here. We reach the same conclusion. 1. Our “system of separated powers and checks and balances established in the Constitution was regarded by the Framers as ‘a self-executing safeguard against the encroachment or aggrandizement of one branch at the expense of the other.’” Morrison v. Olson, 487 U.S. 654, 693 (1988) (quoting Buckley v. Valeo, 424 U.S. 1, 122 (1976)). “If there is one aspect of the doctrine of Separation of Powers that the Founding Fathers agreed upon, it is the principle, as Montesquieu stated it: ‘To prevent the abuse of power, it is necessary that by the very disposition of things, power should be a check to power.’” United States v. Cox, 342 F.2d 167, 190 (5th Cir. 1965) (Wisdom, J., concurring) (quoting Baron de Montesquieu, The Spirit of the Laws bk. XI, ch. IV (1772)). On that foundation, the Framers erected the three branches of government—legislative, executive, and judicial—and endowed each with “the necessary constitutional means and personal motives to resist encroachments of the others.” The Federalist No. 51 (J. Madison); see U.S. Const. art. I, § 1; id. art. II, § 1, cl. 1; id. art. III, § 1. Drawing on the British experience, the Framers “carefully separate[d] the ‘purse’ from the ‘sword’ by assigning to Congress and Congress alone the power of the purse.” Tex. Educ. Agency v. U.S. Dep’t of Educ., 992 F.3d 350, 362 (5th Cir. 2021).[8] The Framers’ reasoning was twofold. First, they viewed Congress’s exclusive “power over the purse” as an indispensable check on “the overgrown prerogatives of the other branches of the government.” The Federalist No. 58 (J. Madison). Indeed, “the separation of purse and sword was the Federalists’ strongest rejoinder to Anti-Federalist fears of a tyrannical president.” Josh Chafetz, Congress’s Constitution, Legislative Authority and the Separation of Powers 57 (2017). The Framers also believed that vesting Congress with control over fiscal matters was the best means of ensuring transparency and accountability to the people. See The Federalist No. 48 (J. Madison) (“[T]he legislative department alone has access to the pockets of the people.”).[9] As James Madison explained, the “power over the purse may, in fact, be regarded as the most complete and effectual weapon with which any constitution can arm the immediate representatives of the people, for obtaining a redress of every grievance, and for carrying into effect every just and salutary measure.” The Federalist No. 58 (J. Madison).[10] The text of the Constitution reflects these foundational considerations. First, even before enumerating how legislation becomes law (i.e., passage by both houses of Congress and presentment to the President for signature), the Constitution provides that “[a]ll Bills for raising Revenue shall originate in the House of Representatives . . . .” U.S. Const. art. I, § 7, cl. 1. It then grants the general authority “[t]o lay and collect Taxes” and spend public funds for various ends—the first power positively granted to Congress by the Constitution. Id. art. I, § 8, cl. 1. Importantly though, that general grant of spending power is cabined by the Appropriations Clause and its follow-on, the Public Accounts Clause: “No money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law; and a regular Statement and Account of the Receipts and Expenditures of all public Money shall be published from time to time.” Id. art. I, § 9, cl. 7. The Appropriations Clause’s “straightforward and explicit command” ensures Congress’s exclusive power over the federal purse. OPM v. Richmond, 496 U.S. 414, 424 (1990). Critically, it makes clear that “[a]ny exercise of a power granted by the Constitution to one of the other branches of Government is limited by a valid reservation of congressional control over funds in the Treasury.” Id. at 425. Of equal importance is what the clause “takes away from Congress: the option not to require legislative appropriations prior to expenditure.” Kate Stith, Congress’ Power of the Purse, 97 Yale L.J.1343, 1349 (1988). Given that the executive is forbidden from unilaterally spending funds, the actual exercise by Congress of its power of the purse is imperative to a functional government. The Appropriations Clause thus does more than reinforce Congress’s power over fiscal matters; it affirmatively obligates Congress to use that authority “to maintain the boundaries between the branches and preserve individual liberty from the encroachments of executive power.” All Am. Check Cashing, 33 F.4th at 231 (Jones, J., concurring). The Appropriations Clause thus embodies the Framers’ objectives of maintaining “the necessary partition among the several departments,” The Federalist No. 51 (J. Madison), and ensuring transparency and accountability between the people and their government. The clause’s role as “a bulwark of the Constitution’s separation of powers” has been repeatedly affirmed. U.S. Dep’t of Navy v. Fed. Lab. Rels. Auth., 665 F.3d 1339, 1347 (D.C. Cir. 2012) (Kavanaugh, J.); see id. (“The Appropriations Clause prevents Executive Branch officers from even inadvertently obligating the Government to pay money without statutory authority.”) (citations omitted); see also, e.g., Sierra Club v. Trump, 929 F.3d 670, 704 (9th Cir. 2019) (“The Appropriations Clause is a vital instrument of separation of powers . . . .”); City of Chicago v. Sessions, 888 F.3d 272, 277 (7th Cir. 2018) (discussing the power of the purse as an important aspect of the separation of powers created by “[t]he founders of our country”); United States v. McIntosh, 833 F.3d 1163, 1175 (9th Cir. 2016) (“The Appropriations Clause plays a critical role in the Constitution’s separation of powers among the three branches of government and the checks and balances between them.”). As Justice Story said: The object is apparent upon the slightest examination. It is to secure regularity, punctuality, and fidelity, in the disbursements of the public money . . . . If it were otherwise, the executive would possess an unbounded power over the public purse of the nation; and might apply all its moneyed resources at his pleasure. The power to control and direct the appropriations, constitutes a most useful and salutary check upon profusion and extravagance, as well as upon corrupt influence and public peculation. 2 Joseph Story, Commentaries on the Constitution of the United States § 1348 (3d ed. 1858). Justice Scalia similarly observed that, while the requirement that funds be disbursed in accord with Congress’s dictate and Congress’s alone may be inconvenient, “clumsy,” or “inefficient,” it “reflect[s] ‘hard choices . . . consciously made by men who had lived under a form of government that permitted arbitrary governmental acts to go unchecked.’” NLRB v. Noel Canning, 573 U.S. 513, 601–02 (2014) (Scalia, J., concurring) (quoting INS v. Chadha, 462 U.S. 919, 959 (1983)). In short, the Appropriations Clause expressly “was intended as a restriction upon the disbursing authority of the Executivedepartment.” Cincinnati Soap Co. v. United States, 301 U.S. 308, 321 (1937). 2. All that in mind, we turn to the Bureau’s structure. The Bureau “wields vast rulemaking, enforcement, and adjudicatory authority over a significant portion of the U.S. economy.” Seila Law, 140 S. Ct. at 2191. “The agency has the authority to conduct investigations, issue subpoenas and civil investigative demands, initiate administrative adjudications, and prosecute civil actions in federal court.” Id. at 2193. The Bureau “may seek restitution, disgorgement, and injunctive relief, as well as civil penalties of up to $1,000,000 (inflation adjusted) for each day that a violation occurs.” Id. Unlike nearly every other administrative agency, Congress placed this “staggering amalgam of legislative, judicial, and executive power in the hands of a single Director” rather than a multimember board or commission. All Am. Check Cashing, 33 F.4th at 221–22 (Jones, J., concurring); see 12 U.S.C. § 5491(b). Most anomalous is the Bureau’s self-actualizing, perpetual funding mechanism. While the great majority of executive agencies rely on annual appropriations for funding, the Bureau does not. See 12 U.S.C. § 5497(a). Instead, each year, the Bureau simply requisitions from the Federal Reserve an amount “determined by the Director to be reasonably necessary to carry out” the Bureau’s functions.[11] Id. The Federal Reserve must grant that request so long as it does not exceed 12% of the Federal Reserve’s “total operating expenses.” 12 U.S.C. § 5497(a)(1)–(2).[12] The funds siphoned by the Bureau, in effect, reduce amounts that would otherwise flow to the general fund of the Treasury, as the Federal Reserve is required to remit surplus funds in excess of a limit set by Congress. See 12 U.S.C. § 289(a)(3)(B). The Bureau thus “receives funding directly from the Federal Reserve, which is itself outside the appropriations process through bank assessments.” Seila Law, 140 S. Ct. at 2194; see 12 U.S.C. § 5497(a).[13] So Congress did not merely cede direct control over the Bureau’s budget by insulating it from annual or other time limited appropriations. It also ceded indirect control by providing that the Bureau’s self-determined funding be drawn from a source that is itself outside the appropriations process—a double insulation from Congress’s purse strings that is “unprecedented” across the government. All Am. Check Cashing, 33 F.4th at 225 (Jones, J., concurring). And where the Federal Reserve at least remains tethered to the Treasury by the requirement that it remit funds above a statutory limit, Congress cut that tether for the Bureau, such that the Treasury will never regain one red cent of the funds unilaterally drawn by the Bureau. This novel cession by Congress of its appropriations power—its very obligation “to maintain the boundaries between the branches,” id. at 231— is in itself enough to give grave pause. But Congress went to even greater lengths to take the Bureau completely off the separation-of-powers books. Indeed, it is literally off the books: Rather than hold funds in a Treasury account, the Bureau maintains “a separate fund, . . . the ‘Bureau of Consumer Financial Protection Fund,’” which “shall be maintained and established at a Federal [R]eserve bank.” 12 U.S.C. § 5497(b)(1). This fund is “under the control of the Director,” and the monies on deposit are permanently available to him without any further act of Congress. Id. § 5497(c)(1). Thus, contra the Federal Reserve, id. § 289(a)(3)(B), the Bureau may “roll over” the self-determined funds it draws ad infinitum. To underscore the point, the Act explicitly states that “[f]unds obtained by or transferred to the Bureau Fund shall not be construed to be Government funds or appropriated monies.” Id. § 5497(c)(2). To underscore it again, Congress expressly renounced its check “as a restriction upon the disbursing authority of the Executive department,” Cincinnati Soap, 301 U.S. at 321, by legislating that “funds derived from the Federal Reserve System . . . shall not be subject to review by the Committees on Appropriations of the House of Representatives and the Senate.” Id. § 5497(a)(2)(C). So the Bureau’s funding is double-insulated on the front end from Congress’s appropriations power. And Congress relinquishedits jurisdiction to review agency funding on the back end. In between, Congress gave the Director its purse containing an off-books charge card that rings up “[un]appropriated monies.” Wherever the line between a constitutionally and unconstitutionally funded agency may be, this unprecedented arrangement crosses it.[14] The Bureau’s perpetual insulation from Congress’s appropriations power, including the express exemption from congressional review of its funding, renders the Bureau “no longer dependent and, as a result, no longer accountable” to Congress and, ultimately, to the people. All Am. Check Cashing, 33 F.4th at 232 (Jones, J., concurring); see id. at 234 (detailing examples showing that the Bureau’s “lack of accountability is not just a theoretical worry”). By abandoning its “most complete and effectual” check on “the overgrown prerogatives of the other branches of the government”—indeed, by enabling them in the Bureau’s case—Congress ran afoul of the separation of powers embodied in the Appropriations Clause. See The Federalist No. 58 (J. Madison). The constitutional problem is more acute because of the Bureau’s capacious portfolio of authority. “It acts as a mini legislature, prosecutor, and court, responsible for creating substantive rules for a wide swath of industries, prosecuting violations, and levying knee-buckling penalties against private citizens.” Seila Law, 140 S. Ct. at 2202 n.8. And the “Director’s newfound presidential subservience exacerbates the constitutional problem[] arising from the [Bureau's] budgetary independence.” All Am. Check Cashing, 33 F.4th at 234 (Jones, J., concurring). An expansive executive agency insulated (no, double-insulated) from Congress’s purse strings, expressly exempt from budgetary review, and headed by a single Director removable at the President’s pleasure is the epitome of the unification of the purse and the sword in the executive—an abomination the Framers warned “would destroy that division of powers on which political liberty is founded.” 2 The Works of Alexander Hamilton 61 (Henry Cabot Lodge ed., 1904). The Bureau’s arguments to the contrary are unconvincing. First, it contends that there is no constitutional infirmity because its funding scheme was enacted by Congress. In essence, the Bureau contends that because Congress spun the agency’s funding mechanism into motion when it passed the Act, voila!—the Appropriations Clause is satisfied. The Bureau’s argument misreads not only Supreme Court precedent but also the plain text of the Appropriations Clause. Start with the clause’s text: “No money shall be drawn from the Treasury, but in Consequence of Appropriations made by law.” U.S. Const. art I, § 9, cl. 7 (emphasis added). A law alone does not suffice—an appropriation is required. Otherwise, why not simply travel under the general procedures for enacting legislation provided elsewhere in Article I? The answer is that spending only “in Consequence of Appropriations made by law” is additive to mere enabling legislation; appropriations are required to meet the Framers’ salutary aims of separating and checking powers and preserving accountability to the people. The Act itself tacitly admits such a distinction in its decree that “[f]unds obtained by or transferred to the Bureau Fund shall not be construed to be . . . appropriated monies.” 12 U.S.C. § 5497(c)(2). We take Congress at its word. But that is the rub. The Bureau relies on the Supreme Court’s statement that the Appropriations Clause “means simply that no money can be paid out of the Treasury unless it has been appropriated by an act of Congress.” Richmond, 496 U.S. at 424 (quoting Cincinnati Soap, 301 U.S. at 321). But neither Richmond nor Cincinnati Soap purported definitively to map the contours of the Appropriations Clause. Regardless, Congress’s mere enactment of a law, by itself, does not satisfy the clause’s requirements. Otherwise, the Bureau’s position means that no federal statute could ever violate the Appropriations Clause because Congress, by definition, enactsthem. As discussed supra,our Constitution’s structural separation of powers teaches us that cannot be so. Cf. New York v. United States, 505 U.S. 144, 182 (1992) (“The Constitution’s division of power among the three branches is violated where one branch invades the territory of another, whether or not the encroached-upon branch approves the encroachment.”). The converse argument, that Congress can alter the Bureau’s perpetual self-funding scheme anytime it wants, curing any infirmity, is likewise unavailing. “Congress is always capable of fixing statutes that impinge on its own authority, but that possibility does not excuse the underlying constitutional problems. Otherwise, no law could run afoul of Article I.” All Am. Check Cashing, 33 F.4th at 238 (Jones, J. concurring); cf. PHH Corp. v. CFPB, 881 F.3d 75, 158(D.C. Cir. 2018) (en banc) (Henderson, J., dissenting) (“[A]n otherwise invalid agency is no less invalid merely because the Congress can fix it at some undetermined point in the future.”), abrogated on other grounds by Seila Law, 140 S. Ct. 2183. The Bureau also contends that because every court to consider its funding structure has deemed it constitutionally sound, we should too.[15] But carefully considering those decisions, we must respectfully disagree with their conclusion. Those courts found the constitutional scale tipped in the Bureau’s favor based largely on one factor: a handful of other agencies are also self-funded. For instance, the D.C. Circuit emphasized that “Congress has consistently exempted financial regulators from appropriations: The Federal Reserve, the Federal Deposit Insurance Corporation, the Office of the Comptroller of the Currency, the National Credit Union Administration, and the Federal Housing Finance Agency all have complete, uncapped budgetary autonomy.” PHH Corp., 881 F.3d at 95. Such a comparison, focused only on whether other agencies possess a degree of budgetary autonomy, mixes apples with oranges. Or, more accurately, with a grapefruit. Even among self-funded agencies, the Bureau is unique. The Bureau’s perpetual self-directed, double-insulated funding structure goes a significant step further than that enjoyed by the other agencies on offer. And none of the agencies cited above “wields enforcement or regulatory authority remotely comparable to the authority the [Bureau] may exercise throughout the economy.” All Am. Check Cashing, 33 F.4th at 237 (Jones, J., concurring); see also William Simpson, Above Reproach: How the Consumer Financial Protection Bureau Escapes Constitutional Checks & Balances, 36 Rev. Banking & Fin. L. 343, 367–69 (2016).[16] Taken together, the Bureau’s express insulation from congressional budgetary review, single Director answerable to the President, and plenary regulatory authority combine to render the Bureau “an innovation with no foothold in history or tradition.” Seila Law, 140 S. Ct. at 2202. It is thus no surprise that the Bureau “brought to the forefront the subject of agency self-funding, a topic previously relegated to passing scholarly references rather than front-page news.” Charles Kruly, Self-Funding and Agency Independence, 81 Geo. Wash. L. Rev. 1733, 1735 (2013). We cannot sum up better than Judge Jones did: [T]he [Bureau]‘s argument for upholding its funding mechanism admits no limiting principle. Indeed, if the [Bureau]‘s funding mechanism is constitutional, then what would stop Congress from similarly divorcing other agencies from the hurly burly of the appropriations process? . . . [T]he general threat to the Constitution’s separation of powers and the particular threat to Congress’s supremacy over fiscal matters are obvious. Congress may no more lawfully chip away at its own obligation to regularly appropriate money than it may abdicate that obligation entirely. If the [Bureau]‘s funding mechanism survives this litigation, the camel’s nose is in the tent. When conditions are right, the rest will follow. All Am. Check Cashing, 33 F.4th at 241 (Jones, J., concurring). The Bureau’s funding apparatus cannot be reconciled with the Appropriations Clause and the clause’s underpinning, the constitutional separation of powers. 3. That leaves the question of remedy. Though Collins is not precisely on point, we follow its framework because, though that case involved an unconstitutional removal provision, we read its analysis as instructive for separation-of-powers cases more generally. See Collins, 141 S. Ct. at 1787– 88; cf. All Am. Check Cashing, 33 F.4th at 241 (Jones, J., concurring) (finding Collins “inapt” for determining a remedy for the Bureau’s “budgetary independence”). Collins clarified a dichotomy between agency actions that involve “a Government actor’s exercise of power that the actor did not lawfully possess” and those that do not. 141 S. Ct. at 1787–88. Examples of the former include actions taken by an unlawfully appointed official, see Lucia v. SEC, 138 S. Ct. 2044, 2055 (2018); a legislative officer’s exercise of executive power, see Bowsher v. Synar, 478 U.S. 714, 727–36 (1986); and the President’s exercise of legislative power, see Clinton v. City of New York, 524 U.S. 417, 438 (1998). The remedy in those cases, invalidation of the unlawful actions, flows “directly from the government actor’s lack of authority to take the challenged action in the first place.” All Am. Check Cashing, 33 F.4th at 241 (Jones, J., concurring). In contrast, the Court found the separation of powers problem posed by an official’s unlawful insulation from removal to be different. Collins, 141 S. Ct. 1787–88. Unlike theabove examples,sucha provision “does not strip” a lawfully appointed government actor “of the power to undertake the other responsibilities of his office.” Id. at 1788. Thus, as discussed supra in II.B., to obtain a remedy, a plaintiff must prove more than the existence of an unconstitutional provision; she must prove that the challenged action actually “inflicted harm.” Id. at 1789. Into which category does the Bureau’s promulgation of the Payday Lending Rule fall, given the agency’s unconstitutional self-funding scheme? The answer turns on the distinction between the Bureau’s power to take the challenged action and the funding that would enable the exercise of that power. Put differently, Congress plainly (and properly) authorized the Bureau to promulgate the Payday Lending Rule, see 12 U.S.C. §§ 5511(a), 5512(b), as discussed supra in II.A–C. But the agency lacked the wherewithal to exercise that power via constitutionally appropriated funds. Framed that way, the Bureau’s unconstitutional funding mechanism “[did] not strip the [Director] of the power to undertake the other responsibilities of his office,” Collins, 141 S. Ct. at 1788 & n.23, but it deprived the Bureau of the lawful money necessary to fulfill those responsibilities. This is a distinction with more than a semantical difference, as it leads us to conclude that, consistent with Collins, the Plaintiffs are not entitled to per se invalidation of the Payday Lending Rule, but rather must show that “the unconstitutional . . . [funding] provision inflicted harm.” Id. at 1788–89. However, making that showing is straightforward in this case. Because the funding employed by the Bureau to promulgate the Payday Lending Rule was wholly drawn through the agency’s unconstitutional funding scheme,[17] there is a linear nexus between the infirm provision (the Bureau’s funding mechanism) and the challenged action (promulgation of the rule). In other words, without its unconstitutional funding, the Bureau lacked any other means to promulgate the rule. Plaintiffs were thus harmed by the Bureau’s improper use of unappropriated funds to engage in the rulemaking at issue. Indeed, the Bureau’s unconstitutional funding structure not only “affected the complained-of decision,” id. at 1801 (Kagan, J., concurring in part), it literally effected the promulgation of the rule. Plaintiffs are therefore entitled to “a rewinding of [the Bureau's] action.” Id. In considering other violations of the Constitution’s separation of powers, the Supreme Court has rewound the unlawful action by granting a new hearing, see Lucia v. SEC, 138 S. Ct. 2044, 2055 (2018), or invalidating an order, see NLRBv. Noel Canning, 573 U.S. 513, 521, 557 (2014); see also 5 U.S.C. § 706(2)(A) (providing that, under the APA, a “reviewing court shall . . . hold unlawful and set aside agency action . . . found to be . . . not in accordance with law”). In like manner, we conclude that the district court erred in granting summary judgment to the Bureau and in denying the Plaintiffs a summary judgment “holding unlawful, enjoining and setting aside” the challenged rule. Accordingly, we render judgment in favor of the Plaintiffs on this claim and vacate the Payday Lending Rule as the product of the Bureau’s unconstitutional funding scheme. III. The Bureau did not exceed its authority under either the Act or the APAin promulgating its 2017 Payday Lending Rule. The issuing Director’s unconstitutional insulation from removal does not in itself invalidatethe rule, and the Plaintiffs fail to demonstrate cognizable harm from that injury. Nor does the Bureau’s rulemaking authority transgress the nondelegation doctrine. We therefore AFFIRM the district court’s entry of summary judgment in favor of the Bureau in part. But Congress’s cession of its power of the purse to the Bureau violates the Appropriations Clause and the Constitution’s underlying structural separation of powers. The district court accordingly erred in granting summary judgment in favor of the Bureau and denying judgment in favor of the Plaintiffs. We therefore REVERSE the judgment of the district court on that issue, RENDER judgment in favor of the Plaintiffs, and VACATE the Bureau’s Payday Lending Rule. AFFIRMED in part; REVERSED in part; and RENDERED.
