Trump’s War on the Administrative State Faces New Resistance

President Donald Trump talks about the Golden Dome missile defense system at the White House in Washington, D.C., May 20, 2025.
President Donald Trump talks about the Golden Dome missile defense system at the White House in Washington, D.C., May 20, 2025. (Chip Somodevilla/Getty Images)

The Humphrey’s decision is not dead yet.

Sign in here to read more.

The Humphrey’s decision is not dead yet.

I n an unusual order issued last Thursday, the Supreme Court signaled that it is poised to pare back significantly, if not outright reverse, its 1935 decision in Humphrey’s Executor v. United States, as President Trump is urging it to do. Humphrey’s approved administrative agencies deemed “independent” of presidential control even though they exercise some executive powers — independence that includes “for good cause” restrictions on a president’s authority to fire the bureaucrats who run such agencies. The Court’s most formidable supporter of the administrative state, Justice Elena Kagan, countered Thursday’s order with a spirited dissent. The next day, Judge Beryl Howell, the president’s longtime nemesis on the federal district court in Washington, D.C., was emphatic that this 90-year-old plinth of progressive governance is not dead yet.

An Unusual Emergency Order

The Supreme Court’s order casting doubt on its Humphrey’s decision was unusual because it was issued from the emergency docket. Emergency rulings tend to be tightly controlled by precedent, not occasions for overturning precedent. A party seeking a stay, as the president was in Trump v. Wilcox, must show a likelihood of success on the merits when the case is fully litigated. Almost invariably, such showings go hand in hand with fidelity to precedent.


The case involves Trump’s firing of Gwynne Wilcox and Cathy Harris, members, respectively, of the National Labor Relations Board (NLRB) and Merit Systems Protection Board (MSPB). I’ve written about the case before (see, e.g., here.) The rulings of Democrat-dominated lower courts, including the Obama-appointed Judge Howell, had enabled these Biden-appointed bureaucrats to remain in their jobs.




Congress created the NLRB and MSPB as “independent” agencies and restricted the president’s removal authority to good cause — essentially, misfeasance or malfeasance in office. The Justice Department concedes — indeed, it flaunts — that the president had no such cause; he fired Wilcox and Harris at will on the theory that the NLRB and MSPB wield executive power and that it is unconstitutional for Congress to impinge on the chief executive’s power to remove officials who do so.

Humphrey’s is the precedent that cuts the other way. As Justice Kagan framed it in her dissent (joined by the Court’s two other progressives, Justices Sonia Sotomayor and Ketanji Brown Jackson), “Humphrey’s undergirds a significant feature of American governance: bipartisan administrative bodies carrying out expertise-based functions with a measure of independence from presidential control.”


That it is a significant feature cannot be gainsaid. Whether it is a constitutional feature is the question.

The Humphrey’s Debate

As I’ve explained previously, the Court has cut back on Humphrey’s, and two of its originalists, Justices Clarence Thomas and Neil Gorsuch, have called for its reversal. But of course, they must call for its reversal because it is still binding precedent. It may well be that Humphrey’s was wrongly decided and that, properly applied, separation-of-powers principles would reject (a) the establishment of administrative agencies that agglomerate executive, legislative, and judicial power, and (b) limits on the president’s authority to remove officials who exercise any executive power. But to get there, Humphrey’s has to be cast aside. That hasn’t happened . . . yet. Sensing which way the wind is blowing, though, Justice Kagan calculates that it will happen by next term.

For what it’s worth, I think Humphrey’s is a bad decision. I’m more offended by the fact that administrative agencies exercise a combination of governmental powers that the Framers were adamant about keeping discrete than about “for cause” limitations on presidential-removal power. The latter are objectionable, too, but the former eviscerates the core of Madison’s separation-of-powers theory. Even more than the Bill of Rights, separation of powers is the Constitution’s bulwark of liberty.


I also don’t accept the progressive premise, well argued by Justice Kagan, that there are “certain spheres of government” in which “knowledgeable people from both parties” can be trusted to make decisions “likely to advance the long-term public good,” and therefore that said “knowledgeable people” should be insulated from politics — i.e., from such measures as removal without cause by someone who is actually elected, the president. Is there a limiting principle that defines what “spheres of government” should be independent of separation of powers and ordinary democratic accountability? I don’t see one.

Furthermore, experience teaches that, notwithstanding what the “long-term public good” may be, “knowledgeable people” are just as prone as other officials to advance the partisan agenda of the president who appoints them. The ongoing litigation is a case in point: Trump seeks to pare down executive agencies and reduce their capacity to undermine the chief executive’s policy preferences. The Democratic appointees, though nominally “executive” and thus subordinate to the president, seek to thwart the Republican chief executive — although when a Democrat is in the White House, they walk in lockstep with the chief executive, even if it means violating statutory law.


Since these “knowledgeable people” not only operate in a democratic republic but are guided more by politics and power than technical know-how, they should be electorally accountable, like our system’s other policymakers. That is, when the public elects a president, that president should get to pick those who wield executive power, subject only to the constitutional requirement of Senate consent. Contrary to what progressives say, this is not removal “without cause”; it is removal for a very good cause: The president runs the executive branch, and government agencies should execute, rather than resist, his lawful policies.

All that said, progressive jurists have made a number of effective arguments in favor of preserving Humphrey’s and the administrative state. Kagan is in the vanguard, illustrating (for example, in her 2020 dissent in Seila Law v. Consumer Financial Protection Bureau ) that there is a significant history of assumptions of constitutionality, by Congress and the Supreme Court, regarding both (a) “independent,” “quasi-legislative,” and “quasi-judicial” administrative agencies, and (b) “for good cause” limitations on presidential removal authority. Judge Howell posited these same arguments in her above-linked opinion in Gwynne Wilcox’s case. Regardless of how one comes out on the soundness and persuasiveness of these assumptions, they make it a more uphill battle for originalists to advance doctrinaire separation-of-powers contentions. As Justices Thomas and Gorsuch have shown, those arguments are strong, but they’re not undeniable.

The Originalists’ Achilles’ Heel: Should Trump Run the Federal Reserve?

Putting aside the competing versions of history and principle, I believe the best policy argument Democratic-appointed judges have made regarding Humphrey’s involves the Federal Reserve.


The Fed is structured as an independent, multi-member commission. It makes mistakes, and it has too much policy discretion. I’m no economist, but I believe the best critiques of the Fed call for decision-making that is less interventionist, less subjective, and more rules-driven, based on proven formulas (see, e.g., analyses by David Bahnsen and John Taylor). Nevertheless, the regulators are serious experts who usually resist doing foolish things and often — as now — have tetchy relations with the White House. It is notable that, despite lots of bullying, no president has ever tried to fire a Fed governor, vice chairman, or chairman. And given the trainwreck the political branches have made of the economy in the last quarter century — an acknowledged national debt of over $36 trillion (with mountains more in unfunded liabilities), a profligate determination not to deal with unsustainable entitlements, and an impending crisis that will be even worse if interest rates rise, as is likely — few people want to eradicate the Fed’s independence in favor of the whims of demonstrably irresponsible pols.




Judges are well aware that today’s Justice Department is not at liberty to assume that the president’s positions could be problematic, even for argument’s sake. Consequently, they have been pressing government lawyers about what becomes of the Fed if they succeed in reversing Humphrey’s. Legally, the lawyers grasp that if Humphrey’s goes, Fed independence goes with it; politically, though, they know the public prefers the comparative stability of Fed independence to the president’s economic caprices.


The lawyers thus struggle mightily to sidestep the judges’ questions — essentially insisting that the Fed is somehow different from other independent agencies that exercise executive power, and that the president has not tried to remove Fed officials, so there’s no need to go there. Saying aloud that the president wants unfettered control over the Fed would roil markets, be deeply unpopular, and perhaps cause the Supreme Court’s conservative majority to flinch when the question of whether to uphold or overturn Humphrey’s is squarely presented.

The Supreme Court knows this too. Though their cases do not directly involve the Fed, Wilcox and Harris contended that if the Court allowed them to be removed from the NLRB and the MSPB, it would “necessarily implicate the constitutionality of for-cause removal protections for members of the Federal Reserve’s Board of Governors or members of the Federal Open Market Committee” (I’m quoting the majority’s recital of these contentions; the link is to the Fed’s website, explaining the FOMC). The Court’s majority took pains to disagree and distinguish the Fed from the NLRB and MSPB, but not very convincingly: “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.”


Justice Kagan had some fun with that one. First, she observed that, despite the Court’s “out of the blue” defense of the institution, “the Federal Reserve’s independence rests on the same constitutional and analytic foundations as that of the NLRB, MSPB, FTC, FCC, and so on – which is to say it rests largely on Humphrey’s.”


Kagan also scoffed at the majority’s claim that its assertion about the First and Second Banks of the United States was supported by a footnote in the aforementioned Seila Law, one of the cases that cut back on Humphrey’s. In fact, that footnote (note 8) was skeptical about “the claim” that the Fed had “a special historical status”; the Seila majority merely “assumed” for argument’s sake that this was so, then moved on to curtail Humphrey’s on other grounds. Kagan was keen to this because she was the Seila Law dissenter who made the claim about the Fed’s special historical status. Somehow, she acidly observes, this “assumption made to humor a dissent” in Seila Law now “gets turned into some kind of holding” in the Court’s Trump v. Wilcox order.

Rather than straining to create “a bespoke Federal Reserve exception,” Kagan chided her colleagues: “If the idea is to reassure the markets, a simpler — and more judicial — approach would have been to deny the President’s motion for a stay on the continued authority of Humphrey’s.”


That’s not what the Court did. Instead, it signaled that the authority of Humphrey’s is about to be gutted. But again, it’s not gutted yet.

Judge Howell Weighs In: The U.S. Institute of Peace Case

On Friday, Judge Howell declined to stay pending appeal her ruling that Trump may not shut down the U.S. Institute of Peace (USIP) or fire its board members, who statutorily may be removed only for good cause (just as described above in connection with the NLRB and MSPB).

An Obama appointee, Judge Howell may be President Trump’s least favorite jurist, which — judging from the president’s Memorial Day proclamation — is saying something. We recently discussed her invalidation of the president’s executive order against the Perkins Coie law firm, a case in which the Justice Department unsuccessfully sought her recusal, citing her lengthy history of ruling against Trump, including when he was the subject of criminal investigations by special counsels.




While I think Judge Howell was right about the Perkins Coie executive order, I’m not impressed by her rationalization that restrictions on Trump’s removal of USIP board members are valid. She claims that the USIP is not an executive agency but, rather, an independent “think tank” that does not exercise executive power. Seriously? Not only does the president appoint most of the USIP’s board; two top cabinet officials, the secretaries of state and defense, are board members. And putting aside the “for cause” restrictions, Congress did vest removal power in the president in the USIP’s governing statute; this indicates that Congress saw the USIP as an executive component. As well it should have: The USIP is a “think tank” to assist the formulation and execution of foreign policy, which is a nigh plenary executive power, with only such exceptions as the Constitution stipulates (e.g., Congress’s power to regulate foreign commerce and the Senate’s powers to consent to treaties and confirm ambassadors) — and those are to be narrowly construed.

More interesting, though, is Howell’s discussion of the preceding day’s Trump v. Wilcox order.


The USIP board members do exercise executive power; ergo, under the reasoning of the Supreme Court’s Thursday order — and of Seila Law, on which the majority relied — the president should be able to remove its board members. Howell nevertheless gives that argument the back of her hand. She opines that Humphrey’s is still binding law unless and until the Supreme Court overrules it, and reasons that Humphrey’s upholds Congress’s authority to create agencies such as the USIP, which are not located in any particular branch. (Since I don’t see any “think tank” clause in the Constitution, I assume Howell believes the Constitution doesn’t materially limit Congress’s power to create pretty much any kind of entity it likes.)

Beyond that, in a tongue-in-cheek passage that I’m sure she grinned at while writing (and that must have made Justice Kagan beam when she read it), Judge Howell says that, rather than a font of executive power, USIP is just a “uniquely structured, quasi-private entity that follows in the distinct historical tradition of the First and Second Banks of the United States” — quoting the Supreme Court majority’s laborious attempt to preserve the Fed while it dismantles Humphrey’s.


I imagine Judge Howell will be reversed, just as Justice Kagan predicts her own efforts will prove futile. But there’s no denying their zeal to breathe new life into a dying New Deal–era precedent.

Exit mobile version