“When an agency executes a congressional mandate against private parties, it exercises executive power – no ifs, ands, or quasis.” – Trump v. Slaughter
“We see no reason, however, why our central bank ought to be trapped in amber any more than any other aspect of our constitutional scheme.” – Trump v. Cook
Last week, on the second-to-last day of the Supreme Court’s 2025 term, Chief Justice John Roberts delivered two opinions that read like they had been written by two different people. In Trump v. Slaughter, Roberts overruled bedrock precedent and held unconstitutional the structure of much of the federal government. In Trump v. Cook, a case arising from President Donald Trump’s attempt to fire Federal Reserve Governor Lisa Cook, Roberts carved out an exception for the Federal Reserve System (the Fed) — “as currently structured and with its existing enforcement authorities.” Although the Chief wrote for majorities in both cases, only one other justice, Brett Kavanaugh, wholly endorsed his approach.
The three justices appointed by Presidents Obama and Biden (Kagan, Sotomayor, and Jackson) joined the Chief in Cook but dissented in Slaughter. On their view, the design of the Fed is constitutional for the same reason that the design of the Federal Trade Commission (FTC) is constitutional: because Congress has the power to create offices and set the terms by which they are held, and the President has to follow those laws.
Justice Clarence Thomas, by contrast, joined Chief Justice Roberts in Slaughter, striking down congressional limits on the President’s power to remove government officials, but derided Roberts for not seeing that ruling through to its logical conclusion. “The Court makes many policy arguments for an ‘independent’ banking agency,” he wrote in his dissent in Cook, “but those are ultimately arguments against the Constitution” (emphasis added).
Justices Samuel Alito and Neil Gorsuch were elliptical. They objected that the question of whether Fed officials were covered by the Court’s theory of presidential power, while “indeed important and sensitive,” was not properly before the justices and so should be left unaddressed (and Lisa Cook, at least temporarily, unemployed). As they explained, President Trump had not challenged the constitutionality of the statutory restrictions in the Federal Reserve Act and so the case concerning the removal of Lisa Cook could be decided in a hypothetical world in which Slaughter had not been decided.[1]
Justice Amy Coney Barrett, meanwhile, described the Chief’s reasoning about the exceptional nature of the Federal Reserve Act as “conclusory.” “Even assuming that the Court is right on the merits,” she wrote, “the issue warrants much more than a few paragraphs.” There are many unresolved issues. For example: “How can history support both a categorical rule and a carveout?” “Do all the Federal Reserve’s existing regulatory powers have the requisite connection to monetary policy? If not, are they grandfathered in?” “And is the Federal Reserve unique, or might history sanction other exceptions too?”
This essay begins to work through these questions. It argues that the Court’s carveout for the Federal Reserve fails on its own terms. With one hand, Roberts cuts down the Humphrey’s Executor “exception” to the categorical rule of Myers (grounded in functionalism) and declares: “Humphrey’s Executor is dead!” With the other hand, Roberts manufactures a new exception to this same rule (now grounded in “history and tradition”), proclaiming in effect: “Long Live Trump v. Cook!” The dissonance is jarring, and the history, faulty. The result is a judicial brick without straw.
From Functionalism to History and Tradition
Last May, in a case called Trump v. Wilcox, the Court indicated that it was seeking a way to save the Fed from the Unitary Executive. But the justices left open how exactly they would justify treating central bankers differently, writing only that the “Federal Reserve is a uniquely structured, quasi-private entity that follows in the distinct historical tradition of the First and Second Banks of the United States.”
Now we know that the Court is going with “the Historical Practice Exception”—a route initially suggested by Professor Daniel Tarullo in his law review article, The Federal Reserve and the Constitution. In so doing, the Court walked away from some of the most dubious aspects of the Wilcox dicta. For example, in Cook, Chief Justice Roberts dropped the term “quasi-private,” perhaps because he and many of his colleagues have long scorned the concepts of “quasi-legislative” and “quasi-judicial” power that undergirded the Court’s reasoning in Humphrey’s Executor. (Indeed, Roberts ridiculed these terms again in Slaughter, proclaiming that the President’s power to remove executive officials is illimitable, “no ifs, ands, or quasis about it.”)[2]
Instead, Cook doubles down on “history and tradition.” The Federal Reserve, the Court explains, “follows in [the] lineage” of the First and Second Banks of the United States. How? In Slaughter, Roberts gives us one sentence: both the Bank and the Board “influence[d] monetary policy,” while not being “subject to plenary Presidential control.” And this is sort of true—at least, if by “monetary policy” we mean the price and quantity of money in circulation. (If by “policy,” we mean “public policy,” then it is not true—the Bank had no influence over the government’s monetary policy, over which Congress retained almost total control.)
In Cook, Roberts gives us a bit more. He recognizes that the Board is “more powerful than its predecessors, managing a vastly more complex economy in a vastly more complex world,” while nonetheless insisting that “[w]hat matters is that the Federal Reserve remains ‘consistent with the principles that underpin’ the First and Second Banks—namely, that monetary policy should not be subject to political interference” (emphasis added). This last bit about political interference in monetary policy is also true in many respects. In the judgment of Roberts (and Kavanaugh), “the Federal Reserve maintains the ‘balance struck by the founding generation’ under ‘modern circumstances.’” While this final point is debatable, it is not obviously wrong.
In addition to these sentences, the Court adds citations to case law. The language about “principles” quoted above comes directly from U.S. v. Rahimi.[3] (So does the “trapped in amber” language quoted initially.) Rahimi concerns the scope of Second Amendment rights and when they can be limited by the government. The test involves identifying some restrictions in the eighteenth century and analogizing them to restrictions today. Apparently, the idea in Cook is that the design of the Bank restricted the President’s removal power in a way that is sufficiently related to the restrictions on the President’s removal power found in the Federal Reserve Act.
But the opinion does not grapple with any of the problems raised by invoking Rahimi. Consider a few. First, if history and tradition is the appropriate way to identify constitutionally permissible limits on the President’s “right” to fire government officials, why did the Court fail to conduct such an analysis for the Federal Trade Commission, the agency at issue in Slaughter? Second, what is the appropriate level of generality at which to compare the Banks to the Board and why? Third, and relatedly, if the Bank is a sufficient analog for insulating administration from partisan politics in the case of “monetary policy,” shouldn’t it also be good enough to underwrite the structure of the FTC? In other words, why should monetary policy be any different from competition policy, capital markets policy, or telecommunications policy? Because Alexander Hamilton wrote about the benefits of separating the power to expand the money supply from the government and its fiscal requirements? Isn’t the relevant Rahimi “principle” that some public policy goals require some measure of non-partisan administration and that Congress gets to decide (subject to Presidential veto) what those goals are?
A Closer Look at the History
The trouble deepens once you engage further with the actual history of the Bank of the United States. For example, the Court fails to address the reasons why the Bank might not be an appropriate analog for purposes of identifying exceptions to the (purported) demands of Article II. The most obvious problem, which Justice Thomas emphasizes in his dissent, is that the Bank is not an abridgement, in any way, of the President’s right to fire government officials. It is not inconsistent with Slaughter’s interpretation of Article II. After all, the Bank is a private enterprise like JPMorgan Chase or Amazon (depending on your level of generality). It was incorporated by the government, but it was not owned or controlled by the government. The Bank had non-governmental shareholders, many of whom were not even U.S. citizens. (See here and here.)
In other words, even assuming that it is proper to apply Rahimi to the structural Constitution, the Banks do not even enter a formalist analysis. Using them to justify restrictions on the President’s power to remove members of the Fed’s Board would be like justifying a restriction on the ownership of rifles today by reference to a law restricting the ownership of poison in 1790. It’s apples and giraffes.
The Court also plays fast and loose with the meaning of words. For example, it uses the term “central bank” to describe eighteenth century corporations that were not “central banks.” “Our Nation’s first de facto central bank predates even our Constitution,” the Chief writes on the first page of his opinion in Cook. But he offers no citation for this claim. The Bank of North America (BNA), to which he is referring, was the nation’s first chartered bank (it was chartered by the Continental Congress). But it could hardly be described as a central bank, de jure or otherwise. A central bank is a bank for banks that oversees the banking system. At the time the Bank of North America was established, there were no other chartered banks in the country at all. And even after that changed, the BNA did not play a meaningful role in banking the banks that were added.[4]
The Bank of the United States, meanwhile, which is frequently described in the secondary literature as a proto central bank,[5] is properly analogized not to the Federal Reserve’s Board of Governors but to its twelve Federal Reserve Banks. They are the bank-part of the country’s current central banking system. And they are private—outside of the government—just like the Bank of the United States. The Board of Governors, by contrast, is not a bank. It has no balance sheet. It cannot lend. It cannot buy assets. It is a public, multi-member commission and a bank regulator. It regulates the Federal Reserve Banks, federally chartered national banks, state chartered banks, and companies that own and control banks (bank holding companies).
Obscuring this distinction, the Cook opinion also conflates the legal meaning of the word “regulate” with its colloquial meaning. For example, to rebut Justice Thomas’s point that the Banks “served no regulatory function” (Roberts’ words), Roberts cites Eric Lomazoff’s Reconstructing the National Bank Controversy (2018), which explains that the Banks “serve[d] as the ‘great regulating wheel’ of the early American financial system.” (Lomazoff is himself quoting the Bank’s first president, Thomas Willing.)
But Lomazoff isn’t using the word in the same way as Justice Thomas. And while Lomazoff’s is probably the best analysis of the first Bank’s role in the country’s monetary system, a fuller examination of his history only raises more problems for the Court’s application of Rahimi. For example, Lomazoff points out that most scholars “who have treated the subject [of how the Bank evolved into an instrument of both fiscal and monetary policy] explicitly identify the Bank’s monetary role as an unexpected post-1791 development” (p. 52 citing examples). And on Lomazoff’s account, the Bank’s transformation into a “regulator” was not intended and was greatly contested (pp. 53-55). More important still, what Lomazoff and Willing are talking about is not legal regulation. The Bank “regulated” the money supply through its commercial activity in the market. Their description of “a great regulating wheel” is even in the form of a metaphor. Indeed, the Bank did not exercise any governmental authority—it had no such authority. Put another way, the Bank regulated the money supply in much the same way that Apple regulates the applications firms can sell in its App Store or the Union Pacific Railroad regulated the flow of trains from St. Louis to California. Congress established the Board of Governors, by contrast, to regulate the money supply in an entirely different way, the way the Interstate Commerce Commission regulated train traffic: with the force of law.
The Chief’s use of Alexander Hamilton’s Report on a National Bank is one of the opinion’s most thoughtful uses of history, nicely explaining eighteenth century concerns with government management of monetary expansion, but even then, two important points are overlooked.
First, Hamilton was very concerned about accountability, just like the Court purports to be in Slaughter. But Hamilton thought that accountability to investors was the way to go. Only people with their own capital and livelihoods at stake, he argued, would have the proper incentives not to overexpand the money supply. And importantly, today’s monetary system continues in this tradition. Monetary expansion, in many respects, remains in the hands of investor-owned banks. What has changed is the addition of government regulation of this activity, in particular via the Federal Reserve System.
Second, and relatedly, Hamilton was concerned with government oversight. He thought that the Secretary of the Treasury should supervise the Bank. “[T]he government owes to itself and to the community … to reserve to itself a right of ascertaining, as often as may be necessary, the state of the Bank,” he explained. “[T]he Government should possess the means of ascertaining, whenever it thinks fit, that so delicate a trust [as is made by the government in the Bank] is executed with fidelity and care.” Although the powers that Congress granted the Secretary in 1791 were quite limited—a whole lot narrower than the Board’s—they are arguably the founding-era analog for the Board’s powers, and they belonged to an officer who served at the president’s pleasure. In other words, dig a bit deeper and the history and tradition of the Bank suggests that the Board falls under Slaughter’s categorical rule, not outside of it.
Dig deeper still, and the history of the Bank starts to collapse the entire premise of Slaughter itself. After all, when the First Congress created the Treasury Department, it intentionally declined to label it an “executive department”—unlike the Department of War and the Department of State—a fact that goes unmentioned in both Cook and Slaughter.[6] And none other than James Madison argued on the floor of Congress–on June 29, 1789–that aspects of the Treasury’s role in the government were not “executive” in nature but “judicial”:
Several arguments were adduced to show the Executive Magistrate had constitutionally a right to remove subordinate officers at pleasure. Among others it was urged, with some force, that these officers were merely to assist him in the performance of duties, which, from the nature of man, he could not execute without them, although he had an unquestionable right to do them if he were able; but I question very much whether he can or ought to have any interference in the settling and adjusting the legal claims of individuals against the United States. The necessary examination and decision in such cases partake too much of the judicial capacity to be blended with the executive. I do not say the office is either executive or judicial; I think it rather distinct from both, though it partakes of each, and therefore some modification, accommodated to those circumstances, ought to take place. I would, therefore, make the officer responsible to every part of the Government.
We are left with the distinct impression that Cook is “policy as history” and what is motivating Chief Justice Roberts and Justice Kavanaugh is a concern with the consequences of extending the Slaughter Rule to the Federal Reserve. “Even temporary uncertainty about the status of the Federal Reserve could spark political upheaval,” Justice Kavanaugh wrote in his concurrence, “including confusion about whether the President could immediately remove multiple Governors at will, as well as turmoil in the U.S. and world economies.” Or as the Chief puts it, “We see no reason to … sow doubt as to the status of one of our Nation’s (and the world’s) most important financial institutions.”
What’s Next?
Where does this leave us? Probably not with a stable doctrinal equilibrium that could rival the Humphrey’s Executor Era (1935-2020) or the Marbury Era before it (1803-1899).[7] Although the rest of the Cook opinion correctly interprets the words “for cause” in the Federal Reserve Act and applies the relevant precedent (see here)—President Trump is likely to pursue removal of Lisa Cook once again. In other words, Cook II beckons. Meanwhile, whenever the Federal Reserve next promulgates a rule that isn’t directly related to overnight interest rate policy, regulated parties may well bring suit on the ground that the relevant statutory authority is not sufficiently connected to monetary policy under Cook and is therefore improperly delegated to an agency not subject to the appropriate control by the President (cf. Bowsher v. Synar). And this could force some tough issues to the fore (such as whether the Federal Reserve can continue to regulate financial market utilities).
Although Roberts and Kavanaugh have indicated how they will vote in such cases (see note 6), the Court’s composition will eventually change. When that happens, will new justices try to narrow the Cook exception? Could the peculiar 5-4 majority in favor of the current vision give way to a 5-4 majority against it? Might the overarching Slaughter theory, in other words, work itself “pure”? Might the Court eventually reject the power of Congress to insulate monetary policy from partisan politics? Perhaps the purification proceeds in stages: first the Federal Reserve loses some or all of its enforcement authorities; then maybe it loses some or all of its rulemaking powers; and at last the “for cause” limit drops out, “unworkable” and “ill-conceived.” We’ve seen this show before. See Loper Bright v. Raimondo, 603 U.S. __ (2024) (Kagan J., dissenting) (the way it works is you “[s]top applying a decision where one should; throw some gratuitous criticisms into a couple of opinions; issue a few separate writings questioning the decision’s premises; give the whole process a few years . . . and voila!—you have a justification for overruling the decision”) (cleaned up).
But there is also another possibility. The theory might work itself more impure. Cook could be expanded. It could be read to protect not just independent monetary policy but all those government functions that require some measure of nonpartisan administration. This might start with new majorities deciding that the Cook exception can cover predominantly adjudicatory bodies. They might then extend Cook to the Federal Communications Commission or the Federal Energy Regulatory Commission. One day perhaps the exception could even come to swallow the rule, and Cook could be invoked to reverse the holding in Slaughter itself. Could not a future majority hold that, notwithstanding the President’s illimitable power to remove executive officers per Slaughter, the Federal Trade Commission is, in fact, “consistent with the principles that underpin” the independence of the Bank of the United States? The justices, after all, have not had a chance to consider such an argument. Besides, the Court has already changed its mind once on this very issue. Perhaps the third time’s the charm.
– – – – – – – – – – – – –
[1] Although it is possible that Alito and Gorsuch will eventually endorse a carveout from the Slaughter Rule along the lines proposed by Roberts and Kavanaugh, as a practical matter their vision of independence would be independence in name only, as they would have had the Court grant the President’s stay application in Cook, which would have allowed removal during the pendency of the litigation.
[2] The Court essentially ignores the fact that these terms long predate Humphrey’s Executor and reflect how the founding generation thought about government power. See, e.g., here.
[3] The Court, mystifyingly, also cites Justice Stephen Breyer’s majority opinion in NLRB v. Noel Canning, 134 S. Ct. 2550 (2014) as well as Justice Antonin Scalia’s concurrence. But Roberts does not acknowledge that these opinions employ very different historical practice theories and ones that are different from what the Court applied in Rahimi. See Ash Ahmed, Historical Practice Theories, 127 Colum. L. Rev. (forthcoming 2027). Scalia’s historical practice theory applies only if the text and structure of the Constitution are “ambiguous,” 134 S. Ct. at 2594, something that Kavanaugh and Roberts flatly deny is the case in Slaughter, Slip. Op. at 4, 8. Meanwhile, Breyer’s historical practice theory (a “gloss” theory) looks to practice over time, id. at 2573, and, if it applied to the Federal Reserve Act, it would certainly demand a contrary result in Slaughter.
[4] This phrase “de facto central bank” is used to describe the Bank of North America primarily in two places: Wikipedia (“the country’s first de facto central bank”), and a website about the history of the American Bankers’ Association maintained by the ABA, the country’s primary bank lobbyist (“Congress charters the Bank of North America … [a] de facto central bank”). The ABA cites no sources. Wikipedia cites page 87 of Jerry Markham’s Financial History of the United States (2002), which says that “[i]n addition to being the first state bank to be officially chartered, this institution was America’s initial attempt at creating a central bank that would become a counterpart to the Bank of England.” Markham, who himself is being a bit loose with language here, does not claim that the Bank of North America ever achieved the status of “de facto” central bank. To the contrary, Markham explains how the Bank actively competed with the Bank of New York (which was established in 1784) for business in New York City and quickly fell out of favor with many government officials (pp. 87–88). The Bank of North America never closed its doors; it is today part of Wells Fargo.
[5] As Bray Hammond explains, it is clear that the Bank was “not simply another commercial bank.” Bray Hammond, Banks and Politics in America from the Revolution to the Civil War 114 (1957). Still, neither Hammond nor any of the other leading historians of American banking would say Congress chartered the Bank to be a “central bank.” As the economic historian Richard Timberlake explained: “The First Bank was not intended to be a central bank; it was not to control the quantity of money. Nor was it to act as a centralized depository, an office of discount for commercial banks, or a lender of last resort. … [T]he four banks in existence in 1791 did not need a central bank in any of its manifestations. In fact, each of these banks approached the model of a single isolated banking system.” The Origins of Central Banking in the United States 4 (1978).
[6] The history also shows that the First Congress created at least two independent commissions that are wholly inconsistent with Slaughter’s Article II theory—the Revolutionary War Debt Commission, whose members were granted secure tenures by Congress, and the Sinking Fund Commission, which included the Vice President and the Chief Justice of the United States. Both are unconvincingly dismissed by the majority in Slaughter and go unmentioned in Cook.
[7] Prior to Humphrey’s Executor, Marbury underwrote Congress’s power to limit the President’s grounds for removing term-tenured officers. 1899 arguably marks the end point of this doctrinal stability, as that year the President fired a member of the Board of General Appraisers without cause, notwithstanding a statutory provision limiting removal to cases of “inefficiency, neglect of duty, or malfeasance in office.” This termination was upheld on (dubious) statutory grounds, but the constitutional claim was eventually vindicated in Myers v. United States in 1926.
