Dissent

For ninety years, the constitutional foundation of the modern American government has rested on a settled understanding: Congress may, in creating multimember regulatory commissions, protect the members of those commissions from removal except for specified causes. That settled understanding was first articulated in Humphrey's Executor v. United States, 295 U. S. 602 (1935) — unanimously, by a Court that included all four sitting Justices who had joined the majority in Myers v. United States, 272 U. S. 52 (1926). It was reaffirmed, unanimously, in Wiener v. United States, 357 U. S. 349 (1958). It was reaffirmed, unanimously as to the single-layer holding, in Free Enterprise Fund v. Public Company Accounting Oversight Board, 561 U. S. 477 (2010). And it was reaffirmed, by a majority that explicitly preserved it, in Seila Law LLC v. Consumer Financial Protection Bureau, 591 U. S. 197 (2020).

The Court does not formally overrule Humphrey's Executor today. But by holding that the modern FTC falls outside that decision's scope, the Court empties Humphrey's Executor of its content. The decision survives, on the Court's account, only as authority for the proposition that the 1935 FTC — a body whose institutional descendants do not exist — could be insulated from at-will removal. That is overruling by attrition. The dissent on which we write would not take that step.

The Court's reasoning rests on three principal moves. First, it treats the modern FTC's executive powers as putting it outside the reach of Humphrey's Executor's holding. Second, it treats the line of cases from Free Enterprise Fund through Collins v. Yellen, 594 U. S. 220 (2021), as having so confined Humphrey's Executor that single-layer for-cause protection for a multimember commission can no longer be sustained when the commission exercises significant executive authority. Third, it treats the doctrine of stare decisis as supplying no obstacle, because what it does today is, in its view, not an overruling at all. Each of these moves is mistaken.

We address them in turn. We then turn to the founding-era record the Court underweighs, to the institutional consequences the Court does not face, and to the question — which the Court leaves unresolved — of where the principle articulated today comes to rest.

I

The Court characterizes Humphrey's Executor as resting on the premise that the 1935 FTC "exercised no part of the executive power vested by the Constitution in the President." 295 U. S., at 628; see ante, at 13. If that were the whole of Humphrey's Executor's holding, the case for confining it would be stronger. But it is not. The Court in 1935 said something more careful, and more durable, than the Court today permits itself to acknowledge.

What Humphrey's Executor held was that Congress, in creating "quasi-legislative or quasi-judicial agencies," may "require them to act in discharge of their duties independently of executive control," and that this authority "includes, as an appropriate incident, power to fix the period during which they shall continue in office, and to forbid their removal except for cause in the meantime." 295 U. S., at 629. The Court was not announcing a categorical rule about which agencies do or do not exercise executive power. It was identifying a class of agencies — multimember bodies of experts, balanced along partisan lines, exercising mixed regulatory and adjudicatory functions — that Congress could constitutionally insulate from at-will removal. The constitutional ground for that insulation was the recognition that "one who holds his office only during the pleasure of another, cannot be depended upon to maintain an attitude of independence against the latter's will." Id., at 629.

That holding has been reaffirmed at every turn. In Wiener, the Court extended the same principle to the War Claims Commission even in the absence of an express statutory restriction, because "the philosophy of Humphrey's Executor, in its explicit language as well as its implications, precludes" at-will removal of officers Congress intended to be "completely free" from executive interference in their adjudicatory duties. 357 U. S., at 356. In Free Enterprise Fund, the Court invalidated the Sarbanes-Oxley Act's dual-removal protection — but it did so on the basis that "[t]his Court has determined, however, that this authority is not without limit," 561 U. S., at 483, and it expressly preserved single-layer for-cause protection. The case before the Court was, the Court emphasized, "a new situation not yet encountered." Id., at 483. Seila Law described Humphrey's Executor as "permit[ting] Congress to give for-cause removal protections to a multimember body of experts." 591 U. S., at 216.

The Court reads these reaffirmations as the kindling for what it does today. We read them as the foundation that it is undoing.

II

The Court's central textual move is to seize on a sentence in footnote 2 of Seila Law — "The Court's conclusion that the FTC did not exercise executive power has not withstood the test of time," 591 U. S., at 216 n. 2 — and to treat that sentence as having stripped Humphrey's Executor of its constitutional reasoning while leaving its disposition in place. The footnote, the Court tells us, "identifies a factual premise on which Humphrey's Executor's holding rested." Ante, at 17. Strip the premise, and "there is nothing left to distinguish Humphrey's Executor from Myers." Ibid.

That reading mistakes both the footnote and the decision. Seila Law was a decision about a particular institutional novelty: an "independent agency that wields significant executive power and is run by a single individual who cannot be removed by the President unless certain statutory criteria are met." 591 U. S., at 204 (emphasis added). The Court emphasized that the CFPB's structure had "no foothold in history or tradition," id., at 220, and that its "single-Director configuration" was "incompatible with the structure of the Constitution, which—with the sole exception of the Presidency—scrupulously avoids concentrating power in the hands of any single individual," id., at 222 (emphasis added). The doctrinal work in Seila Law was done by the singularity of the CFPB, not by the supposed evaporation of Humphrey's Executor's factual premise.

This is not a textual quibble. The two readings of Seila Law lead to opposite results in the case before us today. On the Court's reading, Seila Law is the engine of Humphrey's Executor's destruction: once we recognize that the modern FTC exercises executive power, we cannot sustain its removal protections, because Humphrey's Executor depended on the contrary premise. On the alternative reading — which we think is the better one — Seila Law is what it said it was: a decision about a "new configuration," 591 U. S., at 217, which the Court declined to bring within Humphrey's Executor but did not use as occasion to abandon the multimember tradition Humphrey's Executor protected.

Collins v. Yellen did not change that picture. The agency at issue in Collins was, like the CFPB, a single-director agency. The Court there described its holding as "a straightforward application of our reasoning in Seila Law." 594 U. S., at 251. The "breadth not dispositive" language on which the Court relies today, ante, at 17, was offered to reject the argument that a single-director agency with limited authority might escape Seila Law's reach. It was not offered to displace the multimember tradition. To read it that way is to construe a sentence about how unimportant breadth is to Seila Law's rule as a sentence about how unimportant breadth is to Humphrey's Executor's rule. The two cases describe different rules.

When Seila Law was decided five years ago, four Justices wrote that the majority's reasoning, if extended, would unsettle the institutional architecture of the modern administrative state. See 591 U. S., at 240 (KAGAN, J., concurring in judgment with respect to severability and dissenting in part). The Court today extends that reasoning, and produces the unsettling those Justices predicted. It does so not by overruling Humphrey's Executor but by interpreting its survival as so narrow that nothing now in operation falls within it.

III

We turn to the broader doctrinal claim that justifies the Court's holding: that the President's authority to remove officers who exercise the executive power is "conclusive and preclusive," and that Congress accordingly cannot regulate its exercise. The phrase is borrowed from Trump v. United States, 603 U. S. 593, 608–609 (2024). The Court treats it as supplying the rule of decision. Ante, at 10–11.

That treatment expands "conclusive and preclusive" beyond the foundation on which the phrase rests.

Trump v. United States described the President's authority as "conclusive and preclusive" in particular respects — most plainly with respect to the conduct of criminal investigations and prosecutions, 603 U. S., at 619–621, and the "supervision and direction" of officers exercising those functions. But the Court was careful to acknowledge what its holding did not do. "[N]ot all of the President's official acts fall within his 'conclusive and preclusive' authority." Id., at 651 (BARRETT, J., concurring in part). "Congress has concurrent authority over many government functions, and it may sometimes use that authority to regulate the President's official conduct." Id., at 652. Justice Jackson's tripartite framework in Youngstown Sheet & Tube Co. v. Sawyer, 343 U. S. 579 (1952), on which Trump v. United States relied, expressly identifies areas in which presidential authority is "at its lowest ebb" — "when the President takes measures incompatible with the express or implied will of Congress." Id., at 637 (Jackson, J., concurring). And the canonical example Justice Jackson supplied for that lowest-ebb category was "President Roosevelt's effort to remove a Federal Trade Commissioner." Id., at 638 n. 4.

The Court does not engage that footnote. It is not a small omission. Youngstown's framework treats the FTC removal at issue in Humphrey's Executor as the paradigmatic case of valid statutory limitation on presidential authority. The framework has been treated as canonical by Courts of every ideological complexion for seventy years. To uphold the for-cause restriction in 15 U. S. C. § 41 is not, on Jackson's framework, to expand presidential power's "lowest ebb" — it is to recognize that ebb where the framework places it. To strike down the restriction, conversely, is to read Trump v. United States as having displaced Jackson's framework not just for the criminal-investigation context that decision addressed, but for the entire structure of administrative governance.

We do not think Trump v. United States did any such thing.

IV

The historical record the Court underweighs deserves direct engagement.

The Court treats the Decision of 1789 as supplying the authoritative founding-era statement on the removal power. Ante, at 27. We do not contest that the Decision of 1789 is relevant, or that it bears on what the Vesting Clause is best read to mean. But the same First Congress that made the Decision of 1789 also made decisions of 1790 and 1792 that bear on the same question — and that the Court treats as if they did not exist.

In 1790, the First Congress created the Sinking Fund Commission. Act of Aug. 12, 1790, ch. 47, 1 Stat. 186. Alexander Hamilton, who proposed the Commission, regarded it as essential to the credit of the Nation. The Commission was composed of the Vice President, the Chief Justice, the Secretary of State, the Secretary of the Treasury, and the Attorney General. Hamilton, Report Relative to a Provision for the Support of Public Credit (Jan. 9, 1790). It exercised executive authority over the management of the public debt, including the authority to authorize "open market purchases" of government securities — authority Chief Justice Jay used, on at least one occasion, to overrule the policy preferences of the Secretary of State. See Chabot, Is the Federal Reserve Constitutional? An Originalist Argument for Independent Agencies, 96 Notre Dame L. Rev. 1, 43–46 (2020).

The Court responds that three of the Commission's five members — the Treasury Secretary, the Secretary of State, and the Attorney General — were "indisputably removable at will." Collins, 594 U. S., at 253 n. 19. That is true. But two were not. The Vice President holds office under a separate constitutional warrant; the Chief Justice holds an Article III office. Neither could be removed by the President from the Sinking Fund Commission while continuing to hold his underlying office. See Bamzai & Prakash, The Executive Power of Removal, 136 Harv. L. Rev. 1756, 1842–1843 (2023). The First Congress thus knowingly designed a commission a majority of whose members could, in some configurations of votes, exercise executive authority over a matter Hamilton himself described as essential to the national interest, in ways that the President could not undo by removal.

The Revolutionary War Debt Commission of 1790 is a more direct example still. Act of Aug. 5, 1790, ch. 38, § 1, 1 Stat. 178. The three commissioners served fixed terms. President Washington's commissions to them omitted the usual language indicating that the commissioners served "during the pleasure" of the President. See Chabot, The Interstitial Executive: A View from the Founding 20 (Oct. 28, 2025). Hamilton himself described them as "distinct and Independent Officers, charged with a special and delicate Trust." Letter from Alexander Hamilton to Benjamin Hawkins (Mar. 12, 1794).

These are not isolated curiosities. They are choices made by the same First Congress whose Decision of 1789 the Court treats as dispositive. The most natural reading of the Decision of 1789 — that the Vesting Clause vests the entire executive power in the President, including the power to remove officers exercising it — is incompatible with these choices unless one of two qualifications applies: either (a) the Sinking Fund and Revolutionary War Debt Commissioners were not exercising "executive power" within the meaning of the Vesting Clause, or (b) the Vesting Clause is more accommodating of congressional regulation than the Court today permits.

The Court takes the first path implicitly: by treating the Sinking Fund Commissioners as not really doing executive work, or as functioning in Cabinet-supervised configurations that the President could control through other means. We think the second path more candid, and more faithful to the historical record. The Vesting Clause states a general rule. Like most general rules in our constitutional law, it admits of exceptions traced to historical practice and to the structural purposes the Constitution was framed to serve.

V

We turn to the limiting-principle problem.

The Court's holding rests on the proposition that officers who exercise executive power must be removable at will, subject to two narrow and progressively confined exceptions. The Court does not say what those exceptions are — only what they include and do not include. The exception for adjudicative bodies of the kind addressed in Wiener survives, the Court says, but possibly not in the form Wiener announced. Ante, at 22. The exception (if any) for the Federal Reserve survives, the Court says, but only because the Federal Reserve has a unique quasi-private structure and a distinct historical pedigree. Ibid. The exception for inferior officers with limited duties — the Morrison exception — survives in some form, but its operative test has been disavowed. Ante, at 13. The civil service generally is unaffected by today's decision, the Court assures us, but the principle on which today's decision rests provides no obvious reason why civil servants exercising executive power should remain protected from at-will dismissal.

The Government, when pressed at oral argument, did not provide a clean answer to these questions. The Solicitor General candidly acknowledged that "[t]here are tough — there are maybe tough line-drawing questions" with respect to Article I courts. Tr. of Oral Arg. 20. He declined to commit to a position on the constitutional status of the Tax Court, the Court of Federal Claims, or the Court of Appeals for the Armed Forces. Id., at 15, 47–49. He invited the Court to "overrule Wiener as well." Id., at 16.

The Court today accepts that invitation in part and declines it in part — but it offers no principled criterion by which future Courts will know how to handle the cases not before us. The Federal Reserve is preserved by historical pedigree, but historical pedigree is precisely what Humphrey's Executor enjoyed. Wiener is provisionally preserved by reference to its "adjudicative" character, but the War Claims Commission was no more purely adjudicative than the modern FTC is purely executive. The civil service is preserved by what the Court describes as not being before us, but the Court's logic — that officers exercising executive power must be subject to at-will presidential removal — does not obviously stop at the principal-officer line.

The petitioner's theory has an answer to these questions, but the answer is unattractive. On the petitioner's theory, the Vesting Clause categorically requires at-will removal of all officers exercising executive power, and the historical exceptions for Article I courts, the Federal Reserve, and the like are either explainable on independent grounds or are themselves constitutionally suspect. The Court today is unwilling to embrace that theory — and rightly so, because its consequences are difficult to live with. But by declining to embrace the theory, the Court leaves itself without a doctrinal apparatus capable of explaining why the FTC falls on one side of a constitutional line and the Federal Reserve falls on the other.

VI

The Court's holding implicates more than the Federal Trade Commission. There are, on the most recent count, "some two-dozen multimember independent agencies." Seila Law, 591 U. S., at 230. They include the National Labor Relations Board, the Federal Election Commission, the Nuclear Regulatory Commission, the Securities and Exchange Commission, the Federal Communications Commission, the Federal Energy Regulatory Commission, the Consumer Product Safety Commission, the Federal Reserve Board, the Merit Systems Protection Board, the Federal Maritime Commission, the Federal Mine Safety and Health Review Commission, the Occupational Safety and Health Review Commission, the Postal Regulatory Commission, the Surface Transportation Board, the United States International Trade Commission, and others.

The Court tells us that the constitutional status of these agencies is not before us today. Ante, at 25. We accept that as a description of what the Court has decided. But the framework the Court announces is one whose natural extension reaches these agencies, and reaches them to similar effect. Each of them exercises executive power in the sense the Court now treats as decisive — each files enforcement actions, issues binding rules, or both. Each is structured as a multimember commission with for-cause removal protection. Each has been understood, since Humphrey's Executor, to be a permissible exercise of Congress's authority to structure the administrative state.

These agencies were not designed casually. They were designed by Congresses that determined, in the particular regulatory domains those agencies were charged with administering, that effective regulation required some measure of insulation from the political shifts of presidential succession. They were designed by Congresses that included Members of both political parties, in periods of unified and divided government, over more than a century. And they were signed into law by Presidents of both parties — fifteen of them, since Humphrey's Executor — who could have vetoed the legislation, declined to nominate Commissioners, or sought to repeal the relevant statutes. None did.

The argument that this longstanding institutional architecture nonetheless violates the Constitution is, as we have said, not unserious. But it is also not the only constitutional argument available. The Constitution allocates power among the branches; it also presupposes that the branches will work out, through the political process, the institutional arrangements by which national policy is administered. The political branches did that work over the course of the twentieth century. The Court today declares the result unconstitutional, in a domain where the political branches have repeatedly affirmed it, and where their affirmations have been understood — by this Court, until today — as constitutive of the constitutional settlement.

VII

A final word about stare decisis.

The Court says it does not overrule Humphrey's Executor. Ante, at 26. That formal disclaimer is true in the literal sense that the words "Humphrey's Executor is overruled" do not appear in today's opinion. But what the Court does today is to declare that Humphrey's Executor's holding has no application to any agency now in operation that respondent might point to as protected by it. The 1935 FTC, on whose specific characteristics the Court today rests the survival of Humphrey's Executor, does not exist anywhere. The current FTC, the Court tells us, falls outside Humphrey's Executor's reach. So do the other multimember commissions whose constitutional status will be litigated under the framework the Court announces today.

A precedent that protects nothing is not a precedent that survives. The doctrine of stare decisis is not a doctrine of nominal preservation. It is a doctrine that asks whether the Court's prior decisions remain the operative law governing the cases before us. Humphrey's Executor will not be that, after today. The Court's solicitude for its formal preservation, however well-intentioned, does not change what the Court has done.

We have, in recent Terms, articulated a stare decisis framework that asks four questions: whether the prior decision was egregiously wrong, whether it has proved unworkable, whether reliance interests counsel against overruling, and whether subsequent legal developments have undermined its foundations. See Dobbs v. Jackson Women's Health Organization, 597 U. S. 215, 268 (2022); Ramos v. Louisiana, 590 U. S. 83, 121–122 (2020) (KAVANAUGH, J., concurring in part). On each prong, Humphrey's Executor is on stronger ground than the precedents this Court has recently overruled.

It is not "egregiously wrong." The decision was unanimous in 1935. It has been reaffirmed, in whole or in part, by every Court that has considered it: Wiener (unanimous, 1958), Free Enterprise Fund (preserving its single-layer holding, 2010), Seila Law (declining to revisit it, 2020). The proposition that Congress may, in creating multimember regulatory commissions, protect their members from at-will removal is supported by the historical record of the founding era, by Justice Jackson's Youngstown framework, and by the institutional choices of Congresses and Presidents across more than a century.

It is not unworkable. The line Humphrey's Executor drew — between multimember regulatory commissions on the one hand and single-officer executive departments on the other — has produced, in the ninety years since 1935, almost no litigation. The cases this Court has decided under that line have arisen at its edges (Wiener, Morrison, Free Enterprise Fund, Seila Law, Collins), and in each the Court has provided guidance. The "unworkability" the Court today identifies — that judges have offered different formulations of Humphrey's Executor's scope, ante, at 17 — is itself a product of recent decisions confining Humphrey's Executor, not a feature of the precedent in its settled form.

Reliance interests are substantial. Congress has created some two dozen agencies in reliance on Humphrey's Executor's holding. Those agencies have, in turn, supplied the structure of regulatory governance in significant sectors of the economy, including securities markets, labor relations, energy regulation, consumer financial protection (in its pre-CFPB form), telecommunications, and the conduct of federal elections. The regulated public has organized its affairs around the operation of these agencies. Whatever else may be said about the merits of that institutional architecture, the case for disrupting it is far from "egregious."

Subsequent legal developments have not undermined Humphrey's Executor's foundations. They have confined the decision at its edges. Free Enterprise Fund held that two layers of for-cause protection are impermissible. Seila Law held that a single-director independent agency with significant executive power cannot be insulated from removal. Collins extended Seila Law's logic to a similar single-director agency. None of these decisions purported to disturb Humphrey's Executor's holding as applied to the traditional structure to which it spoke: a multimember commission exercising the mixed functions of regulatory governance.

Today's decision is the first to disturb that holding. The Court calls what it does today a confinement of Humphrey's Executor. We think it is an overruling — partial, indirect, and accomplished by interpretation, but an overruling all the same. We would not take that step.

* * *

The doctrine of separation of powers serves the protection of liberty. It does so by dividing power, and by ensuring that no branch acquires the kind of concentrated authority that the Framers had observed in the European monarchies and feared in their own design. The Court today reads that doctrine as requiring the consolidation of power in the President. It reads it as requiring that officers whose duties Congress designed to be exercised independently of political pressure must instead be subject to dismissal at the will of a single elected official. It reads it, in short, as requiring something Congress and Presidents have, for more than a century, declined to require of themselves.

We respectfully dissent.