Since his inauguration in January 2025, President Donald Trump’s Administration has waged figurative war on the administrative state. The Administration has moved swiftly to lay off tens of thousands of government workers and even to shutter—or effectively shutter—entire agencies such as USAID. Efforts by the Administration to effectively shut down the Consumer Financial Protection Bureau (CFPB) have proven particularly contentious, including physically closing the building to staff, removing its name from the building’s face, and laying off as much as 90 percent of the staff. Moreover, the “Big Beautiful Bill” that Congress passed in July 2025 dramatically reduces the formula used to set the CFPB’s budget, nearly halving the agency’s budget but does not defund it or alter its powers or responsibilities.^[1]

That CFPB has been the centerpiece of high-stakes legal and policy challenges is nothing new. Over its mere decade of existence, the CFPB has been at the center of several important constitutional questions implicating issues from President’s power to make recess appointments (National Labor Relations Board v. Noel Canning)^[2], to the President’s power to remove the heads of “independent” regulatory agencies at-will (Seila Law LLC v. Consumer Financial Protection Bureau )^[3], and finally, in the most recent case decided in 2024, with respect to the ability to fund an Executive agency’s operations without seeking appropriations from Congress (Consumer Financial Protection Bureau v. Community Financial Services Association of America (CFSA)),^[4] which is at the root of the current fights over the President’s efforts to dramatically scale back the CFPB.

Beyond the particulars of CFPB and its future, this trilogy of cases touches on fundamental questions of the separation of powers and the proper role of the judiciary in maintaining the separation of powers under the Constitution. Deep at the root of these three cases rests a question that goes to the heart of the structural Constitution^[5]: to what extent can judges treat the separation of powers as largely self-enforcing in the sense that each branch will have sufficient incentive and ability to protect themselves from intrusions on their legitimate power and authority by the other branches? Or can courts assume that the separation of powers will be largely self-enforcing, in the sense that each branch, and by implication its respective members, will fiercely guard its powers from encroachment from the other branches.

In this sense, it is thought that judicial intervention to referee disputes between the President and Congress should be largely unnecessary. But this ignores a key feature of the separation of powers and checks and balances—they are not merely an end in themselves, a zero-sum power struggle between the branches for power. Rather they are intended as means to the end of preserving individual liberty. By dividing power among the three branches, and further subdividing Congress into two houses, the Framers intended to distribute authority and require the cooperation of all those actors in order to act. We can refer to this idea of the separation of powers as a means to the preservation of individual liberty, and particularly the idea that it will be self-enforcing as each branch will have the incentives and ability to protect their authority, as the “Madisonian Constitution.”

But Madison’s theory of self-enforcing checks and balances, while a valuable starting point for analysis, fails to contemplate the situations under which this system breaks down and the implications for individual liberty. To understand the Supreme Court’s opinion in CFPB v. CFSA it is necessary to revisit Noel Canning and Seila Law. In both of those earlier cases, the Supreme Court intervened in the decision to preserve fundamental principles of the separation of powers—in Noel Canning by blocking President Obama’s efforts to make recess appointments to the NLRB (and Richard Cordray as Acting Director of CFPB) and in Seila Law by striking down Dodd-Frank’s limit on the President’s power to remove the CFPB Director to “for cause” removal. In CFSA, however, the Supreme Court rebuffed a similar challenge to the CFPB’s funding structure.

This article argues that the Supreme Court erred in CFSA by refusing to accept CFSA’s challenge to the financing structure of the CFPB as a violation of the Constitution’s Appropriations Clause. Instead, all three cases share certain fundamental characteristics that necessitate Supreme Court intervention to preserve the separation of powers and, by implication, to shore up the structural protections for individual liberty in the Constitution. Moreover, in none of the three cases did the Supreme Court fully understand the underlying causes of the breakdown in the Madisonian Constitution, perhaps accounting for why it failed in CFSA to see the fundamental conceptual similarity in the cases.

In fact, all three cases illustrate the limits of the Madisonian Constitution and the need for judicial enforcement of separation of powers. In short, the break downs in the Madisonian Constitution are predictable and while potentially self-correcting, highly uncertain. Failing to act to preserve the structural Constitution can imperil individual liberty, potentially permanently, but at least for an extended period. This flaw in the Madisonian Constitution arises from two sources. The first was a conceptual error at the time the Constitution itself was written; the second arose soon thereafter with the rapid emergence of political parties. Madison’s analysis, while largely correct as a first approximation has been revealed to be incomplete as an analytical framework. By acting to strengthen Presidential control over the CFPB in Seila Law while blocking the effort to reassert Congress’s authority over the budget in CFSA, the Supreme Court has effectively combined the powers of purse and sword in the President’s unified hand, enabling a largely unconstrained agency with vast powers and functionally unlimited budget to bring enforcement actions, issue rules, and otherwise take actions against private individuals and companies with virtually no effective constraints from Congress.^[6] As will be discussed below, the Framers pointed to the historical example of English Kings who had waged wars funded through irregular means and without reliance on Parliament’s acquiescence and financing. But the principle was broader and seemingly extended to all activities and with equal force to allowing Congress to exercise executive authority. Requiring the President’s activities to be financed through congressional appropriations was seen as a check on this exercise of authority. 

Dodd-Frank Creation of the CFPB’s Radical Independence

The CFPB was created as part of 2010’s Dodd-Frank financial regulation act that overhauled the US financial regulatory system. The idea of the CFPB was germinated in a 2007 Democracy magazine article by then-Harvard Law Professor Elizabeth Warren, written in response to the growth of subprime mortgage products that eventually contributed to the global financial crisis.^[7] The concept then became a centerpiece of the Obama Administration’s Financial Regulatory Reform agenda following the 2008 Financial Crisis, eventually making its way into the Dodd-Frank law.^[8]

As originally proposed by the Obama Administration, the agency was to be a standalone multi-member commission modeled after the Consumer Products Safety Commission with the qualification that “at least one seat on the Board should be reserved for the head of a prudential regulator.” The Obama Administration’s original proposal was somewhat vague on its funding but contemplated a stable and independent funding stream “which could come in part from fees assessed on entities and transactions across the financial sector.”

In the end after all political negotiations, instead of creating a new agency led by a bipartisan multi-member commission, Dodd-Frank instead established a single-member independent agency headed by a Director appointed by the President subject to the advice and consent of the Senate.^[9] Moreover, once confirmed, the CFPB Director was to serve a five-year term and could be removed from office only “for cause,” defined in Dodd-Frank as “inefficiency, neglect of duty, or malfeasance in office.”^[10] Yet Dodd-Frank went still further, providing that even after the expiration of the five-year term the Director could continue to lead the agency “until a successor has been appointed and qualified,” meaning that any delay in confirming a successor would permit the incumbent Director to remain in place. Indeed, this could even mean that if the Democrats controlled a majority of the Senate they could refuse to confirm the President’s nominee and permit the incumbent Director to remain in place even after the term expires. This latter provision tied the hands of a subsequent president from removing the Director and replacing him with the new President’s choice.

Moreover, although the CFPB would technically be housed within, and draw its operating budget from, the Federal Reserve, its operations would be largely unchecked by the Federal Reserve Board.^[11] For example, Dodd-Frank Section 1012(c), labeled “Autonomy of the Bureau” specifically prohibits the Board of Governors to “intervene in any matter or proceeding before the Director,” “appoint, direct, or remove any officer or employee of the Bureau,” or “merge or consolidate the Bureau” or any of its functions with any division or office of the Fed. In this statutory grant of autonomy in its operations, the independent authority wielded by the Consumer Financial Protection Bureau are distinct from that of, for example, the Bureau of Consumer Protection of the FTC which has no authority to act on its own but may do so only through majority vote of the Federal Trade Commission. The CFPB also would not have to impose its own assessments on regulated entities, but instead would draw its budget from the Fed’s assessments and other sources of revenues.

Dodd-Frank provides that, each quarter, “the Board of Governors [of the Federal Reserve] shall transfer to the Bureau the amount determined by the Director to be reasonably necessary to carry out the authorities of the Bureau under Federal consumer financial law” subject to a cap of 12 percent of the “total operating expenses of the Federal Reserve System.” To further insulate the Bureau from congressional oversight, Dodd-Frank Section 1017(a)(2)(c) specifically provided that these funds “shall not be subject to review” by the House or Senate Appropriations Committees. Dodd-Frank further provided that this provision “may not” be construed as implying an obligation for the Director to “consult with or obtain the consent or approval” of the Office of Management and Budget.

The purpose of these twin pillars of the CFPB’s radical independence structure were a single director who would be confirmed by the Senate but removable only for cause, combined with an independent funding structure that permitted the CFPB to draw its operating budget directly from the Federal Reserve outside of the Congressional appropriations process and President’s budget review. The goal was to simultaneously insulate the Bureau from a feared effort of a future Republican Congress cutting the agency’s budget and a future Republican President from firing the Director and replacing him with a more industry-friendly leader. As Justice Ketanji Brown Jackson noted in her concurring opinion in CFPB v. CFSA, the CFPB’s extreme independence was intentional, as Congress “designed the funding scheme to protect the Bureau from the risk that powerful regulated entities might capture the annual appropriations process.”

As originally designed, the CFPB was seen as arguably the most powerful and least accountable agency in the history of the country.^[12] The original structure of the CFPB created a system with a single Director appointed for a five-year term, subject to confirmation by the Senate, and after that removable by the President only “for cause.” This insulation from both Congress and the President was backed by the unique funding mechanism designed for the CFPB. While nominally identified as a “Bureau” of the Federal Reserve, the CFPB was not subject to annual appropriations by Congress, as is, say, the Federal Trade Commission. Instead, the CFPB is entitled to bypass Congress’s annual appropriations process and draw funds directly from the Federal Reserve Board, subject to a statutory percentage-based cap (which the CFPB has never approached) based on the Fed’s total annual operating expenses.^[13] As of fiscal year 2024, the funding cap for the CFPB is estimated at $785.4 million.^[14] The Fed, in turn, draws its funds from the sale of securities, and from assessments on its member banks.

In addition, rather than being required to remit any transferred but unspent funds to the Treasury, the CFPB is authorized to maintain an additional reserve fund to use at its discretion. By 2024, that reserve had accumulated to approximately $711 million.^[15] In addition, Dodd-Frank provided that the CFPB’s budget need not be submitted to the Office of Management and Budget for review by the President. Together, this means that the CFPB can spend over $1 billion annually with no required review by either Congress or the White House. By comparison, the 2024 budget of the Federal Trade Commission, which possesses most of the Federal Government’s consumer protection jurisdiction and shares antitrust enforcement authority with the Department of Justice, was $425.7 million, roughly half the size of the CFPB’s budget.^[16]

In ruling that this unique funding scheme violated the Constitution, the Fifth Circuit characterized the CFPB’s funding structure as a “double insulation” from Congress’s Appropriations power: the Fed itself derives its funding from its own operations (not from Congress) and the CFPB in turn derives its funding from the Fed (and thus also not from Congressional appropriations).^[17]

As a result, in the years following the CFPB’s creation in 2010, it operated as essentially a perpetual motion machine, with a roving commission that reached to the far corners of the American economy. Once confirmed, the director would remain in place for at least a five-year term (and under the statute would continue until a replacement was named) and during that period the Director would have a functionally unlimited budget to support operations with no need to answer to Congress or even the president. Within that scope, the Director had the power to hire and fire every member of the CFPB staff with no further Congressional input or consent.

In Seila Law LLC v. Consumer Financial Protection Bureau, decided in 2020, the Supreme Court reined in one element of this apparatus by holding that the “for cause” removal limitation was unconstitutional and severed that provision from the CFPB’s statute, making the Director removable by the President.^[18] CFPB v. CFSA challenged the second leg of this novel structure, attacking the CFPB’s funding design.

The First Shoe Drops: Seila Law

The initial legal challenges to the CFPB’s structure raised concerns about both of these elements of the agency’s structure: the limitation on the President’s removal power as well as an Appropriations challenge to its budget.^[19] For reasons that are unclear, however, the removal power challenge to the single-member independent agency structure that was decided in Seila Law was divided from the appropriations challenge and advanced more quickly through the courts.

The first major case to vindicate a challenge to the constitutionality of the CFPB’s structure was PHH Corp. v. Consumer Financial Protection Bureau, decided in 2016.^[20] In an opinion written by then-DC Circuit Judge Brett Kavanaugh, the court held that the limits on the President’s power to remove the CFPB Director rendered the agency’s structure unconstitutional. Kavanaugh’s opinion was later reversed by the DC Circuit sitting en banc. Kavanaugh’s opinion in PHH, however, provided a template for the Supreme Court to decide the issue when it was raised a few years later in Seila Law v. CFPB.

In Seila Law, the Court ruled that the limitation on the President’s removal power violated the Constitution because it “lack[ed] a foundation in historical practice and clashes with constitutional structure by concentrating power in a unilateral actor insulated from Presidential control.” Although the Court recognized that the concept of multi-member independent agencies, whose members are protected from removal by the President, was recognized as legitimate in Humphrey’s Executor v. United States (1935),^[21] the concept of a single-member independent agency was highly uncommon and highly contested through history. Quoting Federalist No. 51, the Court stressed that a foundational goal of the Constitution was to fragment power and to prevent the “gradual concentration” of power in the same department by giving the members of each branch “the necessary constitutional means and personal motives to resist encroachment of the [other branches].” Chief Justice Roberts’s opinion stressed that not only was the Director of the CFPB not elected by the people “nor meaningfully controlled through the threat of removal by someone who is,” but, he noted “The Director does not even depend on Congress for annual appropriations” and so was beyond the effective oversight of elected legislative officials as well. Quoting Federalist No. 58, the Court endorsed Madison’s observation that “power over the purse may, in fact, be regarded as the most compleat and effectual weapon” in assuring accountability to the people. As a result of this systemic lack of accountability to the democratic process, the Court declared that (at least with respect to agencies that exercise the broad powers of the CFPB) limits on the President’s removal power for single-member independent agencies to be unconstitutional. Exceptions to the contrary, according to the Court, were for highly circumscribed and relatively ministerial duties (such as the Social Security Administration or the Customs Office) or were highly contested as improper (as was the case with CFPB from its inception).

The Supreme Court’s Opinion in CFPB v. CFSA

Justice Thomas wrote the 7-2 majority opinion in CFPB v. CFSA, joined by concurring opinions by Justice Kagan (joined by Justices Sotomayor, Kavanaugh, and Barrett), and a separate concurring opinion by Justice Jackson. Justice Alito wrote a dissenting opinion joined by Justice Gorsuch. All the opinions were remarkably short by the standards of recent Supreme Court opinions, including Justice Thomas’s 22-page majority opinion.

The range of reasoning among the four opinions is striking, in that it reveals a variety of jurisprudential approaches taken by the Supreme Court to questions of separation of powers on these issues. At root, however, the opinions for the Justices in the majority reflect an implicit assumption that the Court should defer to the political compromise struck between the President and Congress, rather than intervening to override it.

The brevity of Justice Thomas’s opinion is likely explained by the fact that the majority upheld the CFPB’s funding structure, largely because of a lack of any compelling textual, historical, or functional reason to strike it down. Focusing on the original understanding of the Appropriations Clause, Justice Thomas concluded the Constitution’s text and original understanding places no clear limits on the Appropriations power beyond the requirement that Congress 1) “identify a source of public funds” and 2) “authorize the expenditure of those funds for designated purposes to satisfy the Appropriations Clause.” And in the case of the CFPB, Congress had done so, identifying both the source (the Federal Reserve) and authorization to spend money (on whatever the Director believed “reasonably necessary to carry out the authorities of the Bureau”).^[22] As a result, the funding scheme passed constitutional muster for the Appropriations Clause.

Beyond that bare minimum requirement imposed by the text and limited history surrounding the Appropriations Clause, Thomas suggests the questions of appropriations are essentially political questions to be resolved through negotiations between Congress and the President. Thomas points to a handful of examples from history of agencies funded in perpetuity with no effective budget cap, such as the Customs Office and Post Office, which were largely funded by assessments from users of the service (such as charging postage to mail a letter). Perhaps most important, however, Thomas implies that the fatal flaw in the challengers’ argument is that, although they argue that to be valid under the Appropriations Clause in terms of amount and duration (time), they were unable to point to any basis for those requirements in text or history or where a Court might draw a judicial line.^[23] Thus, even though this particular funding scheme was largely unprecedented, there was no basis on which to conclude that Congress had gone too far. Considering any basis for imposing those requirements or discernible principle for how to apply them, Thomas implied that the Framers left those questions to be sorted out between the President and Congress. As Justice Alito noted in his dissent, Justice Thomas’s test was so broad that Congress could authorize a government agency to be funded entirely by private sources, thereby enabling the agency to set its own budget without any popular control.

Writing in concurrence, Justice Kagan added to Thomas’s originalist opinion an analysis of the tradition and practice of accommodation between Congress and the President with respect to matters of appropriations. She particularly points to the longstanding financial independence of the financial regulatory agencies (such as the Federal Reserve, Office of the Comptroller of the Currency, and others) which have long been self-funding and subsisted largely on assessments from the parties they regulate.

The lineup of Justices signing on to Justice Kagan’s opinion is intriguing, both for those who joined and those who did not. Kagan’s opinion gained four votes, just one shy of the votes necessary to add this analysis to the majority opinion. Justice Sotomayor joined her opinion, but Justice Jackson did not. This suggests some degree of disagreement between Justice Jackson and the analysis in Kagan’s opinion; Jackson, however, joined Justice Thomas’s majority opinion. In addition, Kagan’s concurrence regarding the use of tradition by the Court was joined by Justices Kavanaugh and Barrett. Justice Kavanaugh’s assent is particularly noteworthy, in that he had rejected a similar line of argument by Justice Kagan in Seila Law.

Finally, Justice Jackson’s concurring opinion, writing for herself alone, added a functionalist justification to uphold the CFPB’s funding scheme, arguing that the budgetary independence of the CFPB was an essential part of Congress’s scheme in Dodd-Frank “to protect the Bureau from the risk that powerful regulated entities might capture the annual appropriations process” and that the plaintiffs in the case (a trade association of payday lenders), “represent exactly the type of entity the Bureau’s progenitors sought to regulate and whose influence Congress may have feared.” In her view the Supreme Court should not lightly reject that policy judgment by Congress that such insulation from democratic accountability was necessary to protect the agency’s mission.

Writing in dissent, Justice Alito joined by Justice Gorsuch rejected these rationales. First, pointing to the text of the Appropriations Clause, he argued that it should be read as a “term of art” within the historical context in which it was written. And that historical context is well understood — it arose from the centuries-long struggle between Parliament and the King over control of the purse which was, more fundamentally, control over the governance of policy itself, including the war power. US history after the Constitution’s ratification reinforced this reading, as agencies — with very limited exceptions — were funded by annual appropriations from the Treasury. Third, “While there have been departures from this dominant model,” such as the Customs Office and Post Office, “nothing like the CFPB’s funding scheme,” of double-insulation from Congress’s appropriations power, “has previously been seen.” Moreover, all those agencies had narrow, circumscribed power and none of those agencies possessed the vast scope of policymaking and enforcement authority possessed by the CFPB. Finally, where those agencies subsisted on self-funding mechanisms, those were grounded in services provided to those who use them, such as the Customs and Post Offices charging for their services. Similarly, financial regulatory agencies such as the Office of the Comptroller of the Currency and Federal Deposit Insurance Corporation charge for their services they provide in regulating the solvency of those entities. Fees are assessed on those entities for examination and other services because of the indirect benefit they receive in exchange — such as access to deposit insurance and other regulatory advantages.     . CFPB’s activities, by contrast, are designed to benefit consumers and the larger public, not the regulated entities, which undermines this justification for funding CFPB’s duties (either directly or indirectly) through assessments. Moreover, none of those agencies were funded largely through transfers from another agency that is also removed from the appropriations process, as is the CFPB, and permitted to retain any unspent funds rather than remitting them to the Treasury.

Adding all this up, Alito concludes the Court need not be able to draw a precise outer perimeter on Congress’s delegation of the Appropriations power to recognize in this case it goes too far. Agreeing with Justice Jackson, Alito agrees that Congress created the CFPB with “maximum unaccountability.” But Alito notes that lack of accountability, combined with the CFPB’s vast powers, is exactly the problem. And while Seila Law pared back some of that independence, it left the budgetary issue unaddressed. In fact, while Seila Law fixed half of the constitutional problem by making the agency more accountable to the President, that case actually “worsens the appropriations problem” and “increased the power of the Executive over appropriations.” Because Seila Law vests the President with authority to remove the CFPB Director at will, now the President can instruct the Director to requisition the amount of money the President thinks is desirable, then spend it as directed.

Analyzing CFPB v. CFSA

The flaw in Justice Thomas’s originalist majority opinion, as well as Justice Kagan’s appeal to historical practice, is evident when comparing the Court’s approach to separation of powers disputes in two prior cases involving the CFPB: National Labor Relations Board v. Noel Canning^[24] and Seila Law. Both cases exhibit a structural understanding of the Constitution and the applicability of those principles to the modern age that is lacking in Justice Thomas’s approach to originalist constitutional interpretation and Justice Kagan’s confused approach in CFSA.

In Noel Canning, the named party in the case was the National Labor Relations Board and involved an effort by President Obama to make two purported “recess” appointments to the NLRB, even though Congress was holding pro forma sessions and therefore was not actually “in recess” according to its own rules. Nevertheless, President Obama purported to make two recess appointments to the NLRB. On the same date he made the NLRB appointments, President Obama also named Richard Cordray as a recess appointment as Director of the CFPB under the same theory. Republicans had been filibustering a vote on Cordray’s nomination for months, seeking certain structural reforms to the CFPB as a condition for allowing Cordray’s nomination to go through.

In Noel Canning, the Supreme Court refused to step aside and permit President Obama to unilaterally declare Congress in recess, which would have effectively allowed the President to circumvent the Senate’s advice and consent role with respect to nominations. In Seila Law, the Supreme Court refused to allow the Executive to cede its power to remove the CFPB Director even though President Obama agreed to that loss of Executive authority in Dodd-Frank. In both cases, the Court noted that upholding the separation of powers and checks and balances is not an end in itself, but rather a means to the end of preserving individual liberty. As a result, the Supreme Court has an obligation to uphold constitutional structure even when Congress or the President gives away its powers, if doing so threatens individual liberty.

Yet, in CFPB v. CFSA, the Supreme Court upheld the unprecedented funding scheme of the agency on the ground that some future Congress could choose to rescind that open-ended delegation of power. The ruling implicitly held that checks on the exercise of the appropriations power are political, not legal, in that what one Congress gave away a subsequent Congress could reclaim.

The approach taken by the Court in CFPB v. CFSA that views the conflict over the Appropriations Clause as primarily a political question misses a larger point in the modern discussion over the separation of powers and methods of originalist constitutional interpretation. Justice Thomas’s decision not to impose judicial limits on the Appropriations Clause of the Constitution follows the logic laid out initially by James Madison and the Framers — that the separation of powers should be viewed as a zero-sum struggle for power between the legislative and executive branches.^[25] It follows that neither branch would be willing to cede power unless there is some good reason (in principle or prudence) to do so; hence, historical practice provides a reliable guide to interpreting the structural Constitution. As a result of this assumed mutual checks and balances, it is implied that judicial oversight is not necessary.

As the Supreme Court has noted, the Constitution’s separation of powers and checks and balances are not intended as ends in themselves but means to achieve the ends of the Constitution — namely, to preserve individual liberty and to frustrate the power of “factions” (majority or minority special interests) from commandeering the power of the federal government to promote their own narrow purposes, rather than the public good. Madison explains the logic most clearly in Federalist No. 51. He notes that the fundamental source of government and protection against tyranny is “the people.” Nevertheless, “experience has taught mankind the necessity of auxiliary precautions” such as separation of powers and federalism. The logic of the Constitution is that by distributing power among the various branches, each responsible for different core functions of government, the various branches will check each other, thereby preventing any one of them from becoming sufficiently powerful to oppress the people. Although dependence on the people is the fundamental source of protection against tyranny, the primary, ongoing source of protection is seen as this automatic checking scheme. The clash between the different branches will lead them to keep each in its place and give each of them a formidable tool by which to defend themselves: the legislature holds the purse, the executive the sword, and the judiciary “neither force nor will, but merely judgment.” Each of these elements of the exercise of power is incomplete standing alone but is dependent on the cooperation of the others to achieve its purpose.

According to Madison, the animating energy of this self-checking system is individual ambition. Each branch will be determined to defend itself from encroachments by others and, as a result, Madison posits that those who compose those branches will feel similarly. He writes, “We see it particularly displayed in all the subordinate distributions of power, where the constant aim is to divide and arrange the several offices in such a manner as that each may be a check on the other that the private interests of every individual may be a sentinel over the public rights.” (emphasis added). Through this alignment of individual self-interest with the functions of each branch, “Ambition must be made to counteract ambition. The interest of the man must be connected with the constitutional rights of the place.”

Madison’s assumption that individual self-interest in pursuing power would be “connected” with the “constitutional rights of the place” has long been accepted as a matter of deductive truth.^[26] But, in fact, it is an aspiration, and there remains an empirical question whether the Constitution actually does align the interests of the man with the constitutional rights of the place. As noted by law professors Darryl Levinson and Richard Pildes, despite the dominance of Madison’s vision of competitive branches in constitutional thought, “it has never been clear exactly how the Madisonian machine was supposed to operate.” They note government institutions or composite branches of the federal government do not have wills or interests of their own, divorced from the wills or interests of the individuals that compose them. Thus, while Madison recognizes in Federalist 51 that for the system of check and balances to operate properly and each branch to resist the encroachments of the others, it was necessary to connect the “interest of the man… with the constitutional rights of the place” so that through the pursuit of self-interest each individual member would have an incentive to protect the prerogatives and powers of the body in which he serves.

The debate between Thomas and Alito highlights a larger debate within originalist constitutional interpretation. Recognizing that the purpose of the Constitution is to pursue certain identifiable ends, not merely a list of terms and clauses to be interpreted in isolation, requires an exercise in constitutional construction not mere textual analysis. More to the point, Madison’s assumptions about how the branches will automatically check one another describes their behavior much of the time. But this design can fail when the incentives of individual politicians diverge from the constitutional role contemplated by the body. Analogously, economic markets work to align individual incentives with the overall general welfare most of the time to promote the ends of individual liberty and economic prosperity. However, markets can also fail in predictable fashion, which is where law and regulation step in. Similarly, Madison’s design typically works well to achieve the Constitution’s ends, but not invariably so. Sometimes individual incentives, whether in markets or politics, are the problem not the solution to the challenge of promoting overall social welfare.

Thomas’s constitutional analysis in CFSA reflects what Professor Michael Greve characterizes as “clause-bound” approach to originalist constitutional interpretation.^[27] Under this approach, the judge should seek to give meaning to a particular term by engaging in a detailed linguistic understanding of that particular term’s original public meaning—in CFSA, the term “Appropriations.” Dictionaries are to be consulted; contemporary uses of the word in linguistic context are to be read.

Yet the Constitution is not just a collection of words. Those precise words carry not just linguistic context but also context within the Constitution’s structure and purpose. Greve argues that proper originalist interpretation should take into account the purpose and structure of the overall Constitution to understand what those particular terms mean in their historical and structural context. Greve’s particular focus is on the proper originalist understanding of the Commerce Clause and so-called dormant Commerce Clause, but the insight is generalizable.

The admonition to construe certain clauses within the Constitution’s overall structure is especially important in cases such as CFSA that implicate the Constitution’s structure, rather than individual rights (which may be more susceptible to linguistic analysis). This is especially so where, as here, individual rights are affected indirectly by structural provisions and the failure of structural protections leaves individuals with no effective means of redress. Violations of enumerated individual rights effectively have two sources of vindication. First, the structural protections that tend toward protection. But where those fail, the ability to vindicate one’s rights through an individual cause of action. Unlike harms arising from alleged violation of one’s rights to freedom of speech or religion which can be vindicated by individual lawsuits, however, the Court’s opinion in CFSA leaves the plaintiffs with no effective individual remedy to vindicate their harm. As a result, the plaintiffs in CFSA are effectively left at the mercy of the administrative state and Congress’s vigilance in protecting its powers and limiting agency overreach, a thin reed indeed.

Madison’s Design Flaw

The mechanism Madison envisions is akin to Adam Smith’s “invisible hand” in the market whereby self-interested individuals are led through the pursuit of their own economic self-interest to promote the public good as a byproduct of those self-interested efforts.^[28] With respect to the Constitution’s checks and balances, the analogous idea is that the innate desire of politicians to protect and enlarge their power will lead them to jealously guard the authority of the respective branches in which they operate. The most obvious way to do this is to align the politician’s electoral prospects (“interest of the man”) with protection of the interests of the constituency or function for which the body was constituted. For example, the Framers channeled this concept in the original design of the United States Senate, providing for the election of Senators by state legislatures. As noted in Federalist No. 62, a primary purpose of this mode of selection was to give the state legislatures in their political capacity a voice in the federal government. To do so, the election of Senators was vested in the state legislatures themselves, not in the individual citizens of those states.^[29]

Conceptually, however, Congress is more like an economic firm — a composite of individual actors cooperating to pursue a common goal — not a single individual with a unitary will.^[30] And a fundamental characteristic of firms is the presence of what are called “agency costs” within the firm, where the incentives of the individual are not aligned with the overall interests of the firm. For example, while the desire of the firm’s owners (the shareholders) is to maximize the firm’s profitability, the interests of the firm’s employees will not automatically be aligned to this end. Instead, employees’ interest will be to pursue their own interests, even if that conflicts with the interests of the shareholder, such as by trying to increase their compensation (even if that reduces overall firm profitability) or to work less, activities known as “shirking” in economic parlance. The deviation in incentives between the pursuit of self-interest by the employee pursuing her self-interest, on one hand, and the desire of the firm’s owners on the other creates agency costs within the firm. As an initial matter, the firm will attempt to reduce this self-serving behavior through complicated contractual arrangements designed to align the employee’s incentives with the firm’s ends. But because such means are imperfect in accomplishing their purposes, there has arisen a body of external judicial checks on employee behavior, such as the requirement of fiduciary duty that imposes legal sanctions for failures to put the firm’s interests ahead of one’s own.

Analogously, as a matter of self-interest the goal of individual members of Congress is to be reelected. This individual goal will be often, but not invariably, aligned with preserving the power of Congress as a whole against incursion from other branches. Thus, Madison’s assumption that the interest of the man will automatically align with the constitutional rights of the place to protect Congress’s authority against infringement by other branches (such as by avoiding excessive delegation of authority to the executive branch) is unfounded. The question cannot be resolved as a matter of a priori reasoning, as Madison seemingly implies, but will require assessing the degree to which a Congressman’s individual goals align with that of the body of a whole and whether the incentives to do so are properly aligned. As discussed below, political parties (unanticipated by the Framers in their constitutional design) exacerbate this problem, as partisanship provides still another degree of distraction that complicates politicians’ individual incentives. But the underlying problem is woven into the Framers’ original design—the alignment between the “interests of the man” (reelection) cannot be assumed to be aligned with the “constitutional rights of the place” (to preserve Congress’s power and not to overdelegate the authority of the body as a whole). To understand whether the self-interest of its members will be sufficient to check the expansionist impulses of other branches, therefore, it will be necessary to examine the incentives of individual members of Congress to do so, not merely Congress as a whole.

Among the three branches of the federal government, this agency costs problem will be particularly predominant among Congress and its 535 members. The “unitary” nature of the executive branch headed by a sole President raises fewer problems of incentives. Although the vast size of the executive branch creates an opportunity for agency costs within the executive branch (especially among career employees), the corporate structure of the executive branch and incentives will be for the President to minimize these costs. Moreover, as suggested by William Niskanen’s influential theory of agency behavior, laid out in his famous 1971 book Bureaucracy and Representative Government, most agency employees will share a motivation to increase the power and budgets of their respective agencies, because doing so increases their own power and prestige.^[31] In this sense, the general incentive structure for actors within the executive branch generally fits Madison’s vision of the interests of the man being connected with the institutional rights of the place.^[32] As a result, while agency costs might exist between the President and those who work under his supervision in terms of policy goals, those conflicts are more likely to take the form of infighting within the executive branch, as different agencies vie for jurisdictional power and budgetary resources, rather than externally with respect to Congress or the judiciary.

Congressional Incentives Contrasted

The incentives of individual members of Congress are more complicated than the President’s. The fundamental imperative of every member of Congress is to be reelected. This reality creates incentives for members of Congress to permit broad delegation to the Executive Branch and to fail to protect Congress’s powers from infringement by the President. First, in the modern administrative state, individual Congressmen may acquire more power even through a process that makes Congress as a body less powerful. Second, the overbearing role played by political parties, especially as shaped by the centrality of the President in the modern political system, will often create incentives for individual Congressmen to acquiesce in Presidential power grabs to promote their individual electoral prospects. As a result, it is not possible to simply assume with Madison that the incentives of individual members of Congress will be oriented toward checking the ambitions of the executive branch, nor to defend the power of Congress as an institution against Presidential incursion on Congress’s turf.

These factors create a power imbalance between Congress and the President—because Congress will be weakened by internal collective action problems and the President acts as a unitary force in dealing with Congress, the President will generally have the initiative in any power struggle between them. This intrinsic power imbalance has been exacerbated by partisan party dynamics, which as is discussed below, tends to undermine the willingness of congressional members of the President’s party to support Congress’s abstract authority against the President’s wishes.

Then-Professor Neomi Rao captured the first dynamic in distinguishing between the incentives of the “collective Congress” versus those of individual members of Congress.^[33] She notes that in some instances, Congress can act only collectively, such as by passing legislation, and it will often be in the interest of both Congress as a whole (the collective Congress) and individual members of Congress to preserve Congressional power and to resist infringement by the President on their authority and will try to place limits on the President (such as in the line-item veto case of INS v. Chadha^[34]).

In other cases, however, it often will be in interest of individual members of Congress to make broad delegations to the Executive, or more generally to accede to Presidential infringements on their power, when it advances their own political (and electoral) prospects. As a result, individual members of Congress may favor broad delegations of authority to the Executive branch because that increases the power of the individual member of Congress, even though it reduces the power of the collective Congress as a body. As a result, members of Congress face a sort of prisoner’s dilemma, where each individual member may increase his personal power as a result of delegation, while at the same time diminishing the larger institutional power of Congress.

Rao identifies several reasons why an individual member of Congress might support broad delegations that increase the power of the President relative to Congress as an institution. First, delegation reduces the cost of legislating by enabling Congress to avoid having to reach agreement on the details of legislation and instead delegating the details to the agency. Second, delegation permits Congress to participate in a credit-claiming, blame-passing game whereby it can claim credit for “doing something” in response to democratic demand, but foist responsibility for the decisions onto the agency that has to decide the details of how the law will be implemented, including blame for any costs it imposes. Rao’s analysis focuses on ordinary delegations of authority to agencies. But the analysis is equally applicable to larger structural delegations, such as the delegation to CFPB to essentially establish and fund its own budget; or, as is effectively the case after Seila Law, to allow the President to exercise indirect control over the agency’s budget without congressional interference.

It is generally accepted that the decisions and actions of members of Congress can be best modeled as being consistent with the goal of reelection. Regardless of what other objectives a member of Congress seeks, whether ideological, to serve the public, or to promote one’s narrow self-interest, successful election and reelection to Congress is a necessary proximate goal for accomplishing those other goals. Moreover, electoral politics is a competitive business — a politician that shirks on this primary goal of obtaining reelection is likely to be out-competed in the electoral market by another politician who accomplishes those ends more effectively. Broad delegations of authority that weaken Congress collectively can nevertheless strengthen the authority of an individual member of Congress and thereby promote his electoral prospects.

The exercise of general oversight authority that protects the interests of Congress as a whole can be understood as a public good from the perspective of Congress. It is extremely difficult for any individual member of Congress to claim credit from constituents for simply making sure that Congress does not over-delegate its authority to executive agencies, especially where the exercise of that authority by the agency is broadly popular to the politicians’ constituents. Similarly, general oversight of the agency’s actions to ensure that it is acting efficiently and consistent with its legislative purpose is a public good within Congress, in that the individual member of Congress bears the opportunity cost of time and energy that could be spent on other activities, but the entire Congress is benefited by the agency’s improved performance. Moreover, it will be difficult for the politician to even demonstrate the connection between his efforts and the agency’s outcomes. As a result, the individual politician who provides a collective benefit from more effective oversight will have minimal ability to claim credit for these efforts or to convert them into tangible political support from interest groups or the public at large. A politician’s time and energy will be better spent on projects for which he can claim credit from special interests, such as intervening directly with regulators to shape regulations or intervening on particular issues that benefit political supporters. Because it is difficult for the politician to claim individual credit for these actions taken to protect Congress’s collective authority, it is also difficult for the politician to convert time and effort spent on those activities into direct political support.

Broad delegations that reduce the power of the collective Congress, however, can increase the power of individual members of Congress and enable them to convert their interactions with agencies into private benefits that can aid their political prospects. A common political strategy is the following: Congress can pass a broad bill that provides a vast scope of discretion to an agency, thereby claiming credit for “doing something” and forcing the agency to engage in the challenging task of figuring out how to implement the details of the policy, which invariably will mean allocating the costs and benefits of the policy to competing interest groups. At this point, individual members of Congress can intervene in the regulatory process on behalf of various interest groups to try to minimize the costs (or maximize the benefits) flowing from the agency’s action to the interest group. As opposed to high-profile oversight hearings, these targeted interventions are often informal and largely unobserved by the public at large, such as letters, phone calls, and other contacts designed to sway the agency’s decisions. Although informal, these techniques of congressional control over agency action can be highly effective at shaping agency policy.^[35]

More important for current purposes, however, is that targeted interventions on behalf of specific interest groups enable politicians to “privatize” the benefits flowing from congressional oversight of agency decisionmaking, thereby converting those actions into financial and electoral support. As Rao describes the dynamic, “When an agency has regulatory discretion, members can ‘rescue’ constituents from regulatory burdens. In a vast regulatory state, legislators may be most effective by securing waivers or exemptions for influential groups.”^[36] Rao notes as an example that “immediately after the passage of the Affordable Care Act, nearly 20% of the exemptions granted were to businesses located in House Speaker Nancy Pelosi’s district.”^[37]

The power of individual members of Congress to influence agency decision-making has grown as a result of the increasingly common practice over time of appointing former congressional staffers as commissioners of nominally independent agencies.^[38] As shown by Professors Brian Feinstein and Todd Henderson, around 1980, 10 percent of agency commissioners had previously served as congressional staff, increasing to about 20 percent by 2000. By 2020, however, over 40 percent of all commissioners had Hill experience, including 53 percent in 2009.

This increase in the number of Hill staffers serving as commissioners can be explained by several non-exclusive explanations. Regardless of the causes, the effect has been to increase the already-substantial influence of individual Congressmen over regulatory agencies. For example, an increasingly dysfunctional confirmation process in Congress since the George W. Bush Administration has placed a premium on nominees being backed by a Congressional Sherpa, who will be willing to exert efforts to get a nominee through the confirmation process.^[39] In addition, Feinstein and Henderson argue that in the era of growing Presidential power, placing former congressional staffers as agency leaders is an informal means for members of Congress to exercise influence over the agency’s actions. Moreover, congressional staffers are more likely to have ideological and policy positions that are more well-known to members of Congress, and so may be more predictable in how they will exercise delegated policymaking discretion in general.

Exercising these informal means of individual congressional influence can be a more potent means to affect policy than attempting to persuade Congress as a whole to pass legislation, which is a difficult task in itself, but will also likely require a series of policy compromises and other checks. This will especially be the case during periods of divided government when the President’s party is in the minority in Congress, but individual members of Congress can influence like-minded regulators directly. Rao, for example, quotes former Senator Al Franken’s claim that CFPB is accountable to Congress, stating, “I’ve gone to CFPB on mortgages, rules on mortgages in rural areas, and gotten them to change their rules. So, I can go to them all the time and get changes.”^[40] Rao also notes that it is common knowledge that Senator Warren and her staff “exercise substantial influence and control over the CFPB.”^[41] News reports dubbed Warren the “most influential voice” over appointments and policy in the Biden Administration.^[42] Richard Shelby, the longtime Chairman and one of the top Republicans on the Senate Banking Committee, played a similar role over financial regulatory policy during the first Trump Administration.^[43]

Rao observes, “Legislators may find it expedient in the short term to trade real, formal, lawmaking power for functional influence over administration. Yet in the long term, this may further weaken Congress as an institution — individual congressmen may feel the only way to gain short-term influence and power is to relinquish even more formal lawmaking power.”^[44] As a result, delegations that diminish the institutional power of Congress can occur and will not necessarily be self-correcting. Moreover, the ability to claim credit for exerting direct influence over the regulatory process will be at a premium during periods of divided government (and typically low levels of legislative output) because when Congress is stalemated, no legislation occurs at all that can be used to assess a member’s accomplishments.

Agency actors can also be expected to actively participate in this dynamic. Agencies can selectively further the ideological and political interests of some key Congress members. Agencies have a wide range of discretion that they can use to boost those politicians that support the agency as opposed to those who do not, ranging from relatively minor decisions like where to have field hearings that spotlight local politicians, to broader questions such as where to spend discretionary funds and invest enforcement resources, or which policies to pursue. Politicians that promote and defend the agency may benefit from the exercise of this discretion in a manner that can boost a politician’s electoral prospects. In addition, agency heads may have political ambitions of their own, such as was the case with CFPB Director Richard Cordray, who resigned from the position to run (unsuccessfully) for Governor of Ohio. Unsurprisingly, Cordray’s bid was backed by Senator Elizabeth Warren, a powerful Democratic personality and fundraiser, who Rao noted is generally recognized as asserting special influence over the CFPB.^[45]

The Madisonian vision of institutional checks and balances between the branches is flawed for a second important reason — the dominant role of political parties in the modern environment, a phenomenon not contemplated by the Madisonian system. It has long been recognized that the Framers’ Constitution simply had a blind spot when anticipating the development of political parties. Party loyalty, not institutional loyalty to one’s branch, has long served as a dominant influence on legislators. As Levinson and Pildes have noted, for most practical purposes today, relationships between Congress and the President are defined by a separation of parties, not powers.^[46]

Levinson and Pildes note that the modern structure of democratic politics “effectively channels [self-interested] ambitions into a… set of activities that has nothing to do with aggrandizing their departments or defending them against encroachments.” They continue, “Individual politicians gain and exercise power by winning competitive elections and effectuating political or ideological goals. Neither of these objectives correlates in any obvious way with the interests or power of branches of government as such.” The Framers’ hope that the new government could avoid the factionalism and conflicts of party politics proved short-lived, as such divisions emerged early in American political history.

The substantial role played by party loyalty relative to institutional loyalties was demonstrated by the political reaction to the Supreme Court’s decision in NLRB v. Noel Canning. As noted above, the Supreme Court invalidated those recess appointments as improper, as Congress was not actually in recess. The issue of confirming a nominee to serve as Director was especially significant because Congress had ceded its power of the purse over the agency; as a result, the only significant power remaining with Congress to supervise the agency was the power to confirm the President’s nominee to the position. Leveraging this power, Senate Republicans had filibustered Cordray’s nomination to the position, demanding structural changes to the agency to make it more accountable. President Obama attempted to circumvent this authority by providing a recess appointment for Cordray.

Given this dynamic, it would be a rudimentary application of the Madisonian conception of the Constitution to assume that the Senate would be united in its opposition to President Obama’s unprecedented power grab. That was not the case. Instead, Senatorial opinions on Obama’s move broke along party lines. Forty-five Senators filed an amicus brief in the case opposing the action — every party to the brief was a Republican. No Democratic Senators opposed the action and Democratic Senate Majority Leader Harry Reid subsequently criticized the Supreme Court decision, stating “President Obama did the right thing when he made these appointments on behalf of American workers.”^[47] It would be hard to imagine a clearer example that rejects the Madisonian “separation of powers” vision of the Constitution in favor of the “separation of parties” vision.

A corollary to the importance of parties is that the dynamics of relationships between the branches will differ significantly during periods of unified versus divided government. In the modern political system, the President is the political center of gravity in the US electoral system, and individual members of Congress are defined by where they stand in relation to the President’s agenda. As demonstrated by Senator Reid’s obeisant support for President Obama’s breathtaking incursion into the legislative branch’s autonomy — to literally declare Congress to have recessed contrary to Congress’s own rules — congressional members of the President’s party generally will be exceedingly deferential to the President’s demands, as doing so is seen as aiding the electoral prospects of individual members (except in unusual situations where a senator or congressman finds it in his interest not to follow the President on certain issues, such as was the case in recent years with West Virginia Senator Joe Manchin, a Democrat who held a Senate seat in very Republican West Virginia).

It follows that, for political reasons, during periods of unified government the governing majority will defer to Presidential incursions on their authority, especially if a congressman has reason to believe that he can exert informal influence over the agency. By contrast, the non-governing party will be more resistant to those incursions and during periods of divided government will resist. In fact, empirical evidence suggests that during periods of divided government, Congress is more likely to delegate to independent agencies, over which they have greater influence, than to executive agencies.^[48] Moreover, because of the ascendancy of the President as the central figure in the American political system, this deference during periods of divided government is not symmetrical: the electoral fortunes of individual members of Congress are far more closely tied to that of the President than the President is to any single member of Congress. Thus, the President has little reason to defer to Congressional expansions of authority, unlike the contrary. Congress as an institution will be far less willing to check the President — to have “ambition counteract ambition”—during periods of unified government than during periods of divided government. Even then, the source of any pushback will result from partisan, rather than institutional, dynamics.

Incentives and Constitutional Interpretation

Thus, there is no basis for Madison’s assumption that invariably “interest of the man will be connected to the institutional rights of the place” in terms of preventing infringements on the power of co-equal branches. Just as judicial intervention may be necessary in the corporate setting to limit agency cost behavior by executives when contractual incentives prove insufficient, we could appropriately recognize that sometimes Courts will have to intervene to prevent Congress from surrendering core powers, such as the appropriations power, and allowing the President to unify purse and sword to infringe on individual liberty.

Equally problematic as a conceptual point is the reasoning of Justice Kagan’s concurrence (joined by Justices Kavanaugh and Barrett) that points to history and tradition as support for allowing Congress to effectively delegate the power of the purse to the President. The primary problem with the CFPB’s funding structure is not that it represents Congress ceding its authority over the purse to an unaccountable agency (now, even more surreal, an executive agency that draws its funding from an independent agency, the Federal Reserve, with no formal appropriation from Congress). The problem is that Congress in 2010, during a period of unified government, gave away not only its current ability to control the agency but also the authority of future Congresses to use the appropriations power to ensure accountability of the CFPB to elected officials. In this sense, it is identical to the problem presented in Seila Law that President Obama’s surrender of the power to remove the CFPB Director limited not just his power but also that of future Presidents who were not part of the original political bargain and essentially allow Obama extend his policy influence into the future. Relying on today’s political processes during a period of unified government will not protect the interests of future actors during periods of divided government or private actors harmed by the failure to abide by proper constitutional processes.

Why did Congress give away its appropriations power? For the same reason President Obama was willing to give away the President’s removal power. At the time, Washington was controlled by a unified Democratic government which saw the CFPB as an ideological ramrod for future political purposes. And this is what Justice Jackson implicitly acknowledges in observing that in 2010 the Democratic-controlled Congress intended to insulate the CFPB from pressure from special interests in the financial industry by limiting their ability to exert influence (particularly a future Republican Congress friendlier to the banking industry) to cut CFPB’s budget or otherwise influence its operations, including in the 2010 midterms when Democrats expected to (and in fact, did) lose at least one house of Congress. Similarly, the original arrangement for a Director with a five-year term, removable only for cause, was designed to tie the hands of a future Republican President who might be elected in 2012. And while Justice Jackson parrots the Obama Administration’s line that the extreme independence of the CFPB was necessary to protect the agency from “capture” by powerful special interests, this rationale is incomplete, as illustrated by the agency’s history since its founding. The CFPB’s structure does not insulate it from interest-group capture, it simply means that the agency will be captured by other interests, namely bureaucratic staff and favored interest groups such as trial lawyers and consumer advocacy organizations.^[49]

Periods of unified government can push through laws and governing arrangements that reduce the power of future Congresses for the benefit of individual members of the enacting Congress. Today, reversing or repealing a grant of power from Congress, even one as fundamental as the appropriations power, will require the establishment of not only unified partisan government, but typically a majority large enough to overcome a filibuster by the party that supported the initial legislation. This is an increasingly rare occurrence. Since 1969, the United States has produced unified government for only 7 Congresses, divided government in 19 Congresses, and one (the first Congress of George W. Bush’s Presidency) that began as unified government but became divided mid-Congress because of party defections.^[50] In some instances of technically unified government, the majority was so small in one or both houses as to vitiate the prospect of repeal of any major legislation as a practical matter. Moreover, other than for a brief period following the 2008 elections when the Democrats held 60 seats in the Senate, neither party has held a filibuster-proof (60 vote) partisan majority in the Senate since 1979, when the Democrats held 61 seats in the post-Watergate era.^[51] When prior laws are rolled back, such initiatives are typically relatively small and targeted at particular favored constituencies, rather than fundamental reforms, such as the modest bipartisan 2018 amendments to Dodd-Frank that granted regulatory relief to small banks but left most major portions of Dodd-Frank untouched.^[52]

More daunting is that any such relief will require the President to acquiesce in restoring limits on his authority rather than vetoing the legislation. For precisely the reasons that Madison articulates so well, powers acquired by the President are unlikely to be ceded back to Congress. Equally implausible is the idea that Congress would find a sufficient number of votes to override a Presidential veto.

It is a conventional rhetorical shorthand to refer to legislation as being passed “by Congress.” But while this convention can be a useful rhetorical shortcut in many instances, it should not be allowed to mislead clear thinking. During the oral argument in CFPB v. CFSA,^[53] Justice Kavanaugh asked Solicitor General Prelogar, “So Congress could change it tomorrow?” To which she replied, “Absolutely, Congress could change it tomorrow.” Even then, if Democrats continued to control either house of Congress or even held a sufficient minority in the Senate to block legislation through a filibuster, “Congress” will be unable to act, thereby locking in the status quo.

But, of course, Congress could not change it tomorrow. Congress with the agreement of the President could change it tomorrow. But the partisan alignment of Congress in recent years is likely to be the new norm for the foreseeable future—divided government or narrow majorities for one party or another. For example, prior to the 2024 elections, Democrats controlled the Presidency and held a narrow majority in the Senate while Republicans hold only a narrow and fractious majority in the House of Representatives. Following the 2024 elections, Republicans held the White House and narrow majorities in both Houses of Congress. Most observers expect narrow divided government and narrow majorities for the foreseeable future, frequently producing legislative gridlock and difficulties in achieving major legislative changes. Presidential acquiescence in surrendering this power is especially unlikely in the case of an agency as powerful and sweeping in regulatory authority as the CFPB. Unlike the historical examples of virtually all prior non-appropriated agencies, such as the Customs Service, Post Office, and financial regulatory agencies, the CFPB wields broad policy-making authority that directly affects the welfare of all American consumers, businesses, and the overall American economy. Congress’s      authority over an agency’s budget is its primary tool for oversight and to shape the agency’s policy priorities. As a general rule, regardless of whether a President is a Republican or Democrat, few Presidents are likely to be principled or magnanimous enough to yield this power back to Congress to frustrate the President’s regulatory priorities. Instead, most Presidents would simply prefer to choose their own Director, who can then set their budget unilaterally without Congressional oversight or approval. Again, this is especially the case since Seila Law effectively enables the President to control the CFPB’s budget by threatening to remove any Director that does not follow the President’s instructions.

Events at the CFPB following President Trump’s election demonstrate the difficulties of restoring the separation of powers through the political process. The winding details of the ongoing struggle between the Trump Administration and the CFPB go beyond the scope of this article. What does matter is the unusual nature of the current situation and the obstacles the Trump Administration has faced to carrying out its plan. Briefly put, the Trump Administration appears to have concluded that given the entrenchment of the permanent bureaucracy, or “deep state,” at the CFPB it would be largely impossible to reform the agency to align it with the Administration’s priorities by appointing a new Director.^[54] As a result, the Administration has articulated a desire to eliminate the CFPB and transfer its authorities to other agencies—notably not to other independent agencies such as the FDIC (a bipartisan commission), but instead to the Treasury Department and the OCC, an executive agency also funded by assessments on regulated entities, and thus also controlled by the Trump Administration. At the same time, President Trump and his advisors have suggested a desire to eliminate the FDIC and transfer its responsibilities to the OCC, as well as challenges to the political independence of the Federal Reserve. Finally, the Trump Administration has presented a high-profile effort to overturn the Humphrey’s Executor case and bring independent agencies such as the FTC, under Presidential control. In short, the Trump Administration has tried to bring about its reforms through executive action and litigation instead of through the legislative process.

Given the sweep of this reform agenda and its centrality to the Trump Administration’s priorities, what is striking is the near-abandonment of any efforts to bring this about legislatively, even with the size of President Trump’s mandate and prevailing Republican majorities in both houses of Congress. While the so-called One Big Beautiful Bill domestic tax and spending bill did make it through Congress and reduced the ceiling on the maximum amount the agency could requisition from the Federal Reserve, it made no effort to actually remove the self-funding nature of the agency or substantive changes to the agency’s powers. This limited focus was because the legislation was enacted through Congress’s “reconciliation” process for budget bills, which under Senate rules are exempt from the filibuster. Any broader structural reforms to the CFPB, including most notably subjecting the agency to the congressional appropriations process, would require normal legislation and would be subject to standard legislative processes, including the filibuster.

This recognizes the thrust of Justice Alito’s dissent that the combination of Seila Law with CFSA exacerbates the separation of powers issues of either standing alone. In Seila Law the Court held that the President should have power to remove the CFPB Director at will, rather than for cause, but chose to sever the removal provision from the rest of the statute. The Court’s holding in CFSA would enable the President to instruct the CFPB Director on how much to draw from the Federal Reserve and how to allocate it, and then to remove any Director that did not follow the President’s orders. In fact, it is not clear why the President could not simply instruct the CFPB Director to draw the maximum amount permitted by Dodd-Frank to the extent that the Director deems such amount to be “reasonably necessary to carry out” the agency’s functions. There is no obvious, judicially enforceable limit on the Director’s determination as to what amount of funding might be considered “reasonably necessary” to carry out the CFPB’s functions, especially given that the CFPB Director does not have to provide any justification for her request. Even if such a standard existed, it is doubtful that anyone would have legal standing to challenge the CFPB’s practical judgment that a certain budget request was “reasonably necessary” to carry out the agency’s mission or that the particular party was harmed by the excessive request (as opposed to, for example, the source of the funding as was the issue in CFPB v. CFSA).^[55] Even if a party did have standing it is unlikely that a court would veto the CFPB’s request as excessive in light of what was “reasonably necessary.”

Ironically, in the oral argument in the case Chief Justice Roberts (author of the Seila Law opinion) came tantalizingly close to recognizing the role played by parties today and its potential to upset the suppositions of the Madisonian model. Addressing the Solicitor General, Roberts observed, “legend has it there have been times when the same party controlled both houses of Congress and the White House, and in that situation, you can see Congress empowering the President in a way that might seem unusual to the Framers.”^[56] As noted, his claim is not merely “legend,” it is now well-supported by theoretical understanding and empirical evidence. Nevertheless, Roberts let the matter drop and it apparently played no role in any of the opinions in the case that relied on traditional Madisonian principles.

Seila Law Contra CFSA

The Supreme Court’s unwillingness to look to structural constitutional considerations to protect the power of future Congresses to exercise the appropriations power is in marked contrast to the Court’s prior decision in Seila Law. As the Court noted there, the CFPB director wields vast enforcement, regulatory, and adjudicatory powers with minimal supervision by any elected official. Moreover, in that case the Court noted, “The CFPB’s receipt of funds outside the appropriations process further aggravates the agency’s threat to Presidential control.” The Chief Justice held that given these extensive powers, the CFPB’s insulation from Presidential control — and particularly the inability of the President to remove the Director at will — rendered the limits on removal unconstitutional.

But the tension between Seila Law and CFSA presents itself immediately. The majority and concurring opinions in CFSA suggest that Congress’s perpetual delegation of its appropriations power to a now-Executive Official to be funded from the proceeds of a self-funding independent agency (plus the CFPB’s own endowment) is not problematic because “Congress” could always change the law and reclaim that authority. The assumption is that what “Congress” gives, Congress could also take away. As discussed above, this argument is simply confused as a practical reality in the modern political system. As a logical matter, if Congress could simply change the law to reclaim the appropriations power, couldn’t it also have amended CFPB’s enabling statute to eliminate the limits on the President’s removal power? Nevertheless, it appears that the Court never even considered this possibility in Seila Law, because it recognized that the possibility that a future Congress could change the law was not a defense to the unconstitutionality of the statute, because individual parties suffered direct harm from the defective constitutional structure.

A moment’s reflection also indicates why President Barack Obama was willing to sign a law that restricted the President’s ability to remove a senior agency official — because Obama implicitly understood that the law was not really tying his hands, the five-year term imposed a limit on future Presidents’ ability to remove the Director and replace him with his own choice and extending Obama’s effective control over the agency into the next President’s term.^[57] Moreover, it is well-understood that as a matter of political reality that if President Obama had become dissatisfied with Cordray’s performance, Obama could request Cordray’s resignation and Cordray would almost certainly comply and the formal limits on President Obama’s power to control the agency through the threat of removing the Director were largely irrelevant.^[58] Therefore, the limits on removal operated as sort of a one-way ratchet—it placed no practical limits on President Obama’s ability to control the agency’s operations but it would effectively hamstring any future efforts by Republican Presidents to control the agency’s operations.

For example, if Mitt Romney had won the 2012 election, Director Cordray would have remained in office until late in Romney’s term leaving him little opportunity to influence the agency’s activities. If reelected in 2012 (as he was), Obama could appoint a new Director (or reappoint the existing Director) toward the end of his term that would continue into the next President’s term, which is exactly what President Obama did.^[59] President Obama clearly understood that whomever he selected as Director would act consistently with his preferred policy objectives; indeed, unless a President is utterly incompetent (which Obama was not) the President will select an appointee because she shares the President’s policy priorities. Moreover, as noted above, Cordray later resigned to run for Governor of Ohio, during which he won Obama’s endorsement, which would be expected to have further influenced his loyalty to the President’s agenda. Thus, it is largely meaningless as a practical matter whether Obama held the legal right to remove his hand-picked appointee, as Cordray was understood to be sympathetic to Obama’s policy objectives and Cordray himself had good reason to seek to remain in Obama’s good graces.

This tension between Seila Law and CFSA is even more glaring, given that there is no textual basis for the supposed constitutional removal requirement on which Seila Law turns (a point stressed by Justice Kagan in her Seila Law opinion).^[60] In addition, the validity of limits on the President’s ability to remove officers of independent agencies has been reaffirmed repeatedly since Humphrey’s Executor, whereas the appropriations power has been rarely tested and never to the degree of CFPB. The Appropriations Clause, however, is expressly enumerated in the Constitution and was seen as an essential power possessed by the legislature after centuries of struggle against the King and executive authority. Ironically, the Court seems more vigilant in protecting a power not expressly mentioned in the Constitution (the power of the President to remove an Executive branch official) than in protecting Congress’s constitutionally enumerated authority over the appropriations power.

As Justice Alito recognizes in his dissent in CFSA, the Court’s decision in Seila Law “addressed part of the problem posed” by the design of the CFPB for “maximum unaccountability” by making the Director accountable to the President. Alito continues, pointing on the significant relationship between the two cases:

[B]ut that decision [Seila Law] did nothing to protect Congress’s power of the purse. Indeed, standing alone, Seila Law worsens the appropriations problem. The appropriations requirement developed to ensure that the Executive (in England, the monarch) would be accountable to the people’s elected representatives. Seila Law, however, increased the power of the Executive over appropriations. By brandishing or wielding the threat of removal, a President may push the CFPB director to requisition the amount of money that the President thinks is appropriate and to spend that money as the President wishes.

Thus, the Court struck the limits on the President’s removal power that was included in Dodd-Frank, but not the restrictions on Congress’s appropriations power. Not only does this fail to address the accountability and separation of powers issues, as Alito observes, it actually exacerbates them. In fact, in arguing against severability in Seila Law, petitioners argued that Congress would have been unlikely to have insulated the CFPB from its appropriations control if it had known that CFPB would be brought under Presidential control. Chief Justice Roberts’s opinion in the case acknowledges that, but did not address the argument.

Moreover, in Seila Law, Justice Kagan’s dissent focused on arguments based on tradition and history, just as her concurring opinion in CFSA. But in Seila Law, the Court rejected her historical evidence, characterizing it as sparse and contested over time. As Justice Alito notes in his dissent in CFSA, the historical examples upon which the majority and Justice Kagan’s concurrence rely are hardly more fulsome. Moreover, the provided historical examples — the post office, National Mint, customs office, and certain financial regulators — are easily distinguishable from the CFPB, in that they are agencies with narrow and highly circumscribed missions (in contrast to the vast scope of authority wielded by the CFPB). In those instances, the funding scheme of the agency was tied to the services it provided to the parties it regulated. For example, the Post Office is funded through charging postal rates to those who mail letters and packages; the customs office is funded in part through collection of tonnage fees. In addition, where the fees collected exceed the amounts necessary for operations, any surplus is remitted to the Treasury.

Kagan’s focus on a history of accommodation between the branches misses the point of the Constitution itself, which is to protect individual liberty. As Justice Scalia noted in his concurrence in Noel Canning, “Since the separation of powers exists for the protection of individual liberty, its vitality ‘does not depend’ on ‘whether “the encroached-upon branch approves the encroachment.”’” Similarly, the history of limits on the President’s removal power was not a barrier to striking the limit in Seila Law, as the Court concluded that the CFPB’s single-member structure, combined with the agency’s very broad discretionary authority, necessitated making the agency head accountable to the President. The structural provisions of the Constitution are designed to protect individual liberty, not to look after the interests of the respective branches. As Scalia noted there, to allow such traditions to effectively amend the Constitution through acquiescence by one body, in the incursions by another, amounts to an “adverse-possession theory of executive authority.” Scalia explained that “the political branches cannot by agreement alter the constitutional structure.”

A major purpose of the Constitution is to create precommitments to durable relationships, in order to constrain the government from threatening liberty through short-term expedients.^[61] For example, a major purpose of the Constitution is to limit democratic passions from infringing on individual rights, and the Constitution’s structural protections are intended in large part to constrain majoritarian sentiments.

More important than one branch acquiescing in incursions on its power by another is the systemic problem of a potential collusion of powers by the branches of the federal government to enlarge all of their powers. The Madison concept of the separation of powers rests on the implied presumption that the various branches of the government are locked in a sort of zero-sum, prisoner’s dilemma where they are fighting over a fixed “pie” of federal power, such that an increase in power by one branch can only come at the expense of reducing the power of another, and hence the power of the individuals that comprise them. In theory, this means that conflict between the various branches will thwart efforts to infringe upon individual liberty by checking each branch’s grasping for power.

But this ignores the potential that the branches can cooperate to increase the overall power of the federal government and thereby potentially increase the power of all the branches of the federal government (and their respective members) simultaneously. Indeed, it is a well-known feature of prisoner’s dilemma games that while refusal to cooperate is the dominant strategy in a single-round game, cooperation is the dominant strategy for the repeated prisoners’ dilemma game.^[62] Thus, while in the short-run the President and Congress might see any increase in the power of a rival branch as a loss of power for themselves, over time both members of Congress and the President can develop practices of cooperation that can increase the total amount of power available to both of them and thereby to increase the power of each.^[63] Quite obviously, the interactions between Congress and the White House over the 250-year history of the US Constitution are best understood as a repeated prisoner’s dilemma interaction, not a single-shot, thereby providing many opportunities for this sort of mutually-beneficial cooperation between the branches as opposed to the rivalrous behavior anticipated by Madison.^[64]

Rather than allowing such arrangements to effectuate an informal amendment to the Constitution, such practices of cooperation in violations of the formal separation of powers actually should be viewed with skepticism, not deference. In this sense, traditions regarding the structural Constitution are fundamentally distinct from traditions regarding the scope of individual rights, which hew more closely to the traditional common law method. The Court’s failure to appreciate this distinction in CFSA, as compared to Seila Law, led the Court astray, allowing Congress in 2010 to effectively cede its appropriations authority to the Executive Branch, thereby unleashing the largely unconstrained powers of the President, acting through the CFPB, to bring enforcement actions, issue rules, and take other steps that infringe individual rights with no effective check from Congress.

Conclusion

In its short but eventful existence, the CFPB has been at the center of many of the most fundamental questions of the structural Constitution. The unprecedented combination of substantive power and democratic accountability has made clashes over the agency’s constitutional legitimacy inevitable. Noel Canning, Seila Law, and CFSA all addressed, directly or indirectly, various questions of the structural Constitution as they were raised by the unprecedented independence of the CFPB. While Seila Law increased the CFPB’s accountability to the President, CFSA failed to impose similar protections to ensure the agency’s accountability to Congress. As such, the case represents a massive shift in power from Congress to the President, who can now effectively wield the vast powers of the CFPB free from Congress’s primary tool for control, the power of the purse.

By ignoring the dangers of combining the power of the sword and purse in one hand, the Supreme Court’s decision in CFSA fails to grapple with the way in which the principles of the structural Constitution apply to the modern world of political partisanship and the incentives of individual members of Congress. The Madisonian constitution rests on the assumption that the separation of powers and checks and balances will be more or less self-enforcing as each of the branches of the federal government will resist efforts of the other branches to expand their power, that ambition would counteract ambition. But time has shown that although this may be true as a viable starting point, it is highly incomplete. In part, the limits were built in from the outset—Madison failed to appreciate that while the collective Congress might have an incentive to resist excessive delegations of authority to the President, individual members might increase their own power (and thereby their reelection prospects) through delegations that reduce the overall power and authority of Congress. In part, these cracks emerged later, with the rise of the party system and the increasing shift from Congress to the President as the motivating force in the political system and the incentives that created for individual members of Congress to align themselves with the President’s desires, even where those actions increase the President’s authority at the expense of Congress. For related reasons, these extraconstitutional innovations frequently will not be self-correcting.

Under such pressures it is inevitable that the self-enforcing nature of the structural constitution will break down, threatening individual liberty and calling forth a need for judicial engagement. In the Noel Canning and Seila Law cases the Court did not simply punt to the political process but intervened to protect the structural Constitution. Mere “clause-bound” reading of Constitutional provisions, as illustrated by Justice Thomas’s majority opinion in CFSA, isolated from their structural contexts ignores the constitutional and structural principles that underlie them. Ironically, the Supreme Court was more protective of the structural Constitution in the absence of any clear textual foundation in Seila Law and thus had to appeal to the deep structural principles of the Constitution than in CFSA where the Court relied on its narrow reading of the constitutional text without appreciating the underlying structural goals that text represents.

*George Mason University Foundation Professor of Law, Antonin Scalia Law School and Former Chair, Consumer Financial Protection Bureau Taskforce on Federal Consumer Financial Law (2020-21).

References

[1] See Greg Iacurci, “Trump’s ‘Big Beautiful Bill’ Slashes CFPB Funding: What it Means for You,” CNBC.com (Jul. 9 2025), https://www.cnbc.com/2025/07/09/trump-big-beautiful-bill-slashes-cfpb-funding-what-it-means.html.

[2] NLRB v. Noel Canning, 573 U.S. 513 (2014).

[3] Seila Law LLC v. Consumer Financial Protection Bureau, 591 U.S. 197 (2020).

[4] Consumer Financial Protection Bureau v. Community Financial Services Association of America, Ltd., 601 U.S. __ (2024).

[5] Throughout this article, the term “structural Constitution” will refer to those provisions that define the framework and organization of the government, such as separation of powers, federalism, and rules governing amendments and the like, in contrast to those provisions such as the Bill of Rights, which focus on the direct protection of individual rights. The structural Constitution is viewed as an indirect way to preserve individual liberty and balance the spheres of individual liberty and democracy through procedural mechanisms rather than direct definition and enforcement of individual rights.

[6] The Framers were greatly concerned to prevent the accumulation of the “purse and sword” in the same government actor, especially enabling the executive to exercise its power without needing to resort to the people and their representatives to raise the necessary funds. References to the need to separate the “purse” and “sword” were ubiquitous during the ratification debates and surrounding period and referred to the idea that the same actor should hold both the power to exercise executive authority, such as to wage war, and fund the operations of government, which was seen as a proper legislative function. Some Anti-Federalists argued the principle should require separation of taxation power and executive authority into different governments (such as state versus federal). See Cornell Law School, Legal Information Institute, Art. I,S8.C11.2.2.5 “Declare War Clause and State Ratification Debates on the Constitution.” https://www.law.cornell.edu/constitution-conan/article-1/section-8/clause-1/declare-war-clause-and-state-ratification-debates-on-the-constitution. Federalists, by contrast, argued that it was sufficient to place the power to raise and spend money in different hands from executive power. See also James Madison, Powers of Congress to Regulate the Militia (June 14, 1788), https://founders.archives.gov/documents/Madison/01-11-02-0085 (“The only rational meaning, is, that the sword and purse are not to be given to the same member. Apply it to the British government, which has      been mentioned. The sword is in the hands of the British king. The purse in the hands of the parliament. It is so in America, as far as any analogy can exist.”).

[7] Elizabeth Warren, “Unsafe at Any Rate,” Democracy (Summer 2007, No. 5).

[8] United States Department of the Treasury, Financial Regulatory Reform: A New Foundation: Rebuilding Financial Supervision and Regulation (June 2009), https://www.wsj.com/public/resources/documents/finregfinal06172009.pdf.

[9] Dodd-Frank Sec. 1011(b)(2), codified as 12 U.S.C. §5491(b)(2).

[10] Dodd-Frank Sec. 1011(c), codified as 12 U.S.C. §5491(c).

[11] Section 1023 of Dodd-Frank provides the lone substantive check on the CFPB, permitting the Financial Stability Oversight Council (FSOC) to set aside by 2/3 vote a “final regulation prescribed by the Bureau” if the regulation “would put the entire safety and soundness of the United States banking system or the stability of the financial system at risk.” Notably, the CFPB Director is a voting member of the FSOC as well as a member of the board of the Federal Deposit Insurance Corporation, whose Chair also serves on the FSOC.

[12] See Todd J. Zywicki, “The Consumer Financial Protection Bureau: Savior or Menace?” George Washington Law Review 81, no. 3 (April 2013): 856-928.

[13] Dodd-Frank Wall Street Reform and Consumer Protection Act §1017(a)(1). A separate issue, not discussed here, is the provision that the transfer be funded by “earnings” of the Federal Reserve Board. It has been argued that for several years the Federal Reserve has not had “earnings” but has instead has suffered losses. See Paul H. Kupiec and Alex J. Pollock, LawandLiberty.org, Can the Fed Fund the CFPB? (June 5, 2024), https://lawliberty.org/can-the-fed-fund-the-cfpb/; Brian Johnson, Examining the New Debate on CFPB Funding, Patomak.com (July 9, 2024), https://patomak.com/2024/07/09/examining-the-new-debate-on-cfpb-funding/.

[14] Consumer Financial Protection Bureau, CFO Update Through the First Quarter of Fiscal Year 2024: October 1, 2023-December 31, 2023 (Issued March 1, 2024, revised July 31, 2024), https://files.consumerfinance.gov/f/documents/cfpb_cfo-update_report_fy-2024_q1.pdf. At the time the case was decided the budget cap was approximately $750 million.

[15] Evan Weinberger, “Vought Agrees to Delay CFPB Funding Slash as Layoffs Continue,” Bloomberglaw.com (Feb. 14, 2025), https://news.bloomberglaw.com/banking-law/vought-agrees-to-delay-cfpb-funding-slash-as-layoffs-continue.

[16] Federal Trade Commission, Budget and Strategy, https://www.ftc.gov/about-ftc/budget-strategy#:~:text=About%20the%20FTC,-Mission&text=We%20use%20our%20budget%20(%24,promote%20competition%20in%20the%20marketplace.

[17] Community Financial Services Association of America, Ltd. V. Consumer Financial Protection Bureau, 51 F. 4^th 616 (5^th Cir. 2022).

[18] Seila Law LLC v. Consumer Financial Protection Bureau, 591 U.S. 197 (2020).

[19] See Competitive Enterprise Institute, “Challenging Dodd-Frank: State National Bank of Big Spring v. Mnuchin” (Sept. 7, 2018), https://cei.org/court_case/challenging-dodd-frank-state-national-bank-of-big-spring-v-mnuchin/.

[20] PHH Corp. v. Consumer Financial Protection Bureau, 839 F.3d 1 (D.C. Cir. 2016).

[21] Humphrey’s Executor v. United States, 295 U.S. 602 (1935).

[22] Dodd-Frank Sec. 1017(a)(1), codified as 12 U.S. §5497(a)(1).

[23] The Court also contrasted to the Constitution’s textual silence on the duration of appropriations with its express limit on the funding of the army for no more than two years as provided in Art. I, §8, cl. 12.

[24] NLRP v. Noel Canning, 573 U.S. 513 (2014).

[25] See A.C. Pritchard and Todd J. Zywicki, “Constitutions and Spontaneous Orders: A Response to Professor McGinnis,” North Carolina Law Review 77: 537-550 (1998-1999).

[26] By referring to the constitutional rights of “the place,” Madison is referring to the particular functions and responsibilities vested in each branch of the government.

[27] See Michael S. Greve, The Upside-Down Constitution (Cambridge, MA: Harvard University Press, 2012); Michael S. Greve, “The Originalism That Was, and the One That Will Be,” Yale Journal of Law & the Humanities 25: 101-111 (2013).

[28] “It is not from the benevolence of the butcher, the brewer, or the baker, that we expect our dinner, but from their regard to their own interest. We address ourselves, no to their humanity but to their self-love, and never talk to them of our own necessities but of their advantages.” Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, Book 1, Ch. 2 (1776).

[29] In fact, this incentive structure worked fairly well to preserve state authority and check the growth of the federal government, until the enactment of the Seventeenth Amendment during the Progressive Era adopted direct election. See Todd J. Zywicki, “Beyond the Shell and Husk of History: The History of the Seventeenth Amendment and Its Implications for Current Reform Proposals,” Cleveland State Law Review 45: 165-234 (1997).

[30] In the memorable short-hand by political scientist Kenneth Shepsle, “Congress is a ‘They,’ Not an ‘It.’” See Kenneth A. Shepsle, “Congress is ‘They,’ Not an “It’: Legislative Intent as Oxymoron,”International Review of Law and Economics 12, No. 2: 239-256 (June 1992).

[31] William Niskanen, Bureaucracy and Representative Government (New York: Routledge, 1971).

[32] As discussed below, a possible exception to this general rule might involve the increasingly common practice of appointing former congressional staffers to leadership positions within “independent” agencies. In such situations, the agency head may retain incentive to remain loyal to his or her congressional sponsor. For reasons discussed, however, this would likely take the form of increasing the influence of the individual member of Congress to influence policy rather than an increase in the power of Congress as a body relative to the executive branch of any given agency.

[33] See Neomi Rao, “Administrative Collusion,” New York University Law Review 90, No. 5: 1463-1526 (Nov. 2015); see also Donald Kochan, “Strategic Institutional Positioning: How We Have Come to Generate Environmental Law Without Congress,” Texas A&M Law Review 6, No. 2: 323-343 (Winter 2019) (discussing incentives of individual members of Congress to delegate broadly and incentives of agencies to exercise the delegations of power as expansively as possible).

[34] INS v. Chadha, 462 U.S. 919 (1983).

[35] See Barry R. Weingast and Mark J. Moran, “Bureaucratic Discretion or Congressional Control? Regulatory Policymaking by the Federal Trade Commission,” Journal of Political Economy 91, No. 5: 765-800 (Oct. 1983).

[36] See Rao, supra note 25, at p. 1481.

[37] Id.

[38] See Brian D. Feinstein and M. Todd Henderson, “Congress’s Commissioners: Former Hill Staffers at the S.E.C. and Other Independent Regulatory Commissions,” Yale Journal on Regulation 38: 175-245 (2021).

[39] See Feinstein and Henderson, at 177-78.

[40] The Administrative State v. The Constitution: Dodd-Frank at Five Years: Hearing Before the Subcomm. on the Constitution of the S. Comm. on the Judiciary, 114th Cong. (2015) (statement of Sen. Al Franken, Member, S. Comm. on the Judiciary).

[41] Rao, supra note 25, at p. 1483.

[42] Zachary Warmbrodt, “’Most influential voice’: Warren’s Network Spreads through Biden Administration,” Politico (Mar. 15, 2021).

[43] Andrew Ackerman, “’Shelby Mafia’ Is Helping Trump Deregulate Wall Street” (Dec. 12, 2017).

[44] Rao, supra note 25, at p. 1490.

[45] See Daniel Strauss, “Warren Goes All In for Cordray,” Politico Pro (Apr. 13, 2018), https://subscriber.politicopro.com/article/2018/04/warren-goes-all-in-for-cordray-479598.

[46] See Daryl J. Levison and Richard H. Pildes, “Separation of Parties, Not Powers, Harvard Law Review 119, No. 8: 2311-2386 (Jun. 2006).

[47] See Robert Barnes, “Supreme Court Rebukes Obama on Recess Appointments,” Washington Post (June 26, 2014), https://www.washingtonpost.com/politics/supreme-court-rebukes-obama-on-recess-appointments/2014/06/26/e5e4fefa-e831-11e3-a86b-362fd5443d19_story.html.

[48] See David Epstein and Sharyn O’Halloran, Delegating Powers: A Transaction Cost Politics Approach to Policy Making under Separation of Powers (Cambridge: Cambridge University Press, Dec. 2009).

[49] See Todd J. Zywicki, “The Consumer Financial Protection Bureau: Savior or Menace?” George Washington Law Review 81, No. 3: 856-928 (Apr. 2013).

[50] See United States of Representatives, “Party Government Since 1857,” https://history.house.gov/Institution/Presidents-Coinciding/Party-Government/.

[51] Following the 2008 election, Democrats held 58 seats in the U.S. Senate, augmented by the subsequent decision of Pennsylvania Senator Arlen Specter to switch parties and the resolution of the contested Minnesota Senate election in favor of Al Franken in July 2009. In August 2009, however, Democratic Massachusetts Senator Ted Kennedy passed away and the seat switched to Republican Scott Brown in a special election in January 2010, thereby ending the brief period during which Democrats held a filibuster-proof majority.

[52] See Jacot Pramuk and John W. Schoen, “Why 17 Democrats Voted with Republicans to Ease Bank Rules,” CNBC.com (Mar. 15, 2018), https://www.cnbc.com/2018/03/15/why-senate-democrats-voted-for-bank-bill-to-ease-dodd-frank-rules.html.

[53] Transcript of Oral Argument in Consumer Financial Protection Bureau v. Community Financial Services Association of America, Limited, Et al. (Oct. 3, 2023), https://www.supremecourt.gov/oral_arguments/argument_transcripts/2023/22-448_4f15.pdf.

[54] See John Heltman, “How Trump’s First 100 Days Changed the Game in Washington,” American Banker (Apr. 29, 2025) (quoting Todd Zywicki, “When Trump and those around them looked back at the first term, they concluded that what undermined them… was the bureaucracy—the deep state, the unaccountable bureaucrats.”).

[55] The legal requirement that a party have “standing” to sue requires that the party raising a challenge to the governmental action demonstrate: (1) the “plaintiff suffered or likely will suffer an injury in fact”; (2) the “defendant caused or likely will be the cause of the injury”; and (3) the “requested judicial relief likely will redress the injury.” Michael R. Hatcher, Michael D. Thomas, and Jennifer E. Burgess, “What U.S. Supreme Court Decision on Standing Tells Us About Challenges to Corporate DEI Initiatives,” JacksonLewis.com (July 17, 2024), https://www.jacksonlewis.com/insights/what-us-supreme-court-decision-standing-tells-us-about-challenges-corporate-dei-initiatives. Although the legal doctrine regarding standing is complex and confusing, it is clear that the first requirement (a claim of an injury in fact) “must be a specific and actual harm” not merely a “general complaint or harm.” It is unlikely that any party claiming the budget request is excessive relative to what is considered “reasonably necessary” above what is legitimate would be considered the sort of particularized harm necessary to establish injury in fact or that it could prove that limiting the agency to the amount that would be “reasonably necessary” would “redress” the claimed injury. As a result, it is doubtful that any party would have standing to sue on that ground.

[56] Transcript of Oral Argument in Consumer Financial Protection Bureau v. Community Financial Services Association of America, Limited, Et al. (Oct. 3, 2023) at p. 10.

[57] In fact, as noted above, Dodd-Frank’s provision that the incumbent Director could remain in office even after the expiration of his term until a successor is “appointed and qualified” could permit the Director to remain in place even beyond the expiration of the five-year term.

[58] This practical reality was especially the case with respect to someone such as Cordray who had future political aims for which remaining in good stead with the Obama Administration and Democratic Party generally were essential. As noted, he resigned as Director of CFPB to seek elective office (for which he would seek Obama’s endorsement) and then later returned to Washington during the Biden Administration as the head of the Federal Student Aid office of the Department of Education.

[59] The actual mechanics turned out to be a bit convoluted in that Cordray originally was appointed to an illegal “recess” appointment and then was appointed to a full term, which had the effect of extending Cordray’s term into the Trump Administration. Cordray was eventually confirmed to a five-year term in July 2013. Cordray resigned in November 2017 before the expiration of his term to pursue an unsuccessful run for the Ohio governorship.

[60] See Opinion of Justice Kagan concurring in the judgment with respect to the Court’s holding on severability of the for-cause removal provision but dissenting on the constitutionality of limiting the President’s removal power      (“The text of the Constitution allows these common for-cause removal limits. Nothing in it speaks of removal.”). 591 U.S. __, ___ (2020).  See also John C. Harrison, “The Unitary Executive Without Inherent Presidential Removal Power,” Center for the Study of the Administrative State CSAS Working Paper 22-03, https://administrativestate.gmu.edu/wp-content/uploads/2022/08/Harrison-Revised-FINAL.pdf (arguing that the power to remove officers is not inherent in the unitary executive); but see Aditya Bamzai and Saikrishna Bangalore Prakash, “The Executive Power of Removal,” Harvard Law Review 136: 1756-1844 (2023) (arguing that the “executive power” historically was understood to include the power to remove officials).

[61] See A.C. Pritchard and Todd J. Zywicki, “Finding the Constitution: An Economic Analysis of Tradition’s Role in Constitutional Interpretation,” North Carolina Law Review 77: 409-521 (1999).

[62] See Maxwell L. Stearns, Todd J. Zywicki, and Thomas J. Micelli, Law and Economics: Private and Public 578-98 (St. Paul, MN: West Academic Publishing, 2018). Smith famously also argued that although we appeal to sellers’ self-interest rather than their benevolence for our bread, those merchants will collude rather than compete if given the opportunity and ability to do so. See Adam Smith, An Inquiry into the Nature and Causes of the Wealth of Nations, Book 1, Ch. 10 (1776).

[63] See Stearns, Zywicki, and Micelli, n. 71. Of course, this applies to the courts as well—for example, an increase in the power of the federal government will also increase the potential influence of federal judges over issues they care about.

[64] Michael Greve has argued for analogous reasons deriving from the incentives of state officials that competitive federalism has a tendency to collapse into “cartel federalism” in which state governments can collude to bring in the federal government to dampen this competitive process. See Greve, n. 33.