The Unbearable Lightness of the Unitary Executive Theory
Through the unlawful firing of National Labor Relations Board member Gwynne Wilcox and a February 18 executive order, entitled “Ensuring Accountability for All Agencies,” President Donald J. Trump is trying to eliminate the independence of independent regulatory agencies. These are the multi-member agencies, such as the National Labor Relations Board, the Federal Trade Commission (FTC), and the Federal Communications Commission, whose members are protected by statute against being fired without good cause. The relevant statutes typically provide also that members will serve specific terms and that neither major party can hold more than a bare majority of seats.
The modern independent agency design, which the U.S. Congress first adopted for the Interstate Commerce Commission in 1887, is intended to foster deliberation, provide a monitoring system for each party through the capacity for dissent, reduce the prospects for agency “capture” by special interests, and promote impartiality in the administration of statutes.
In contrast, President Trump’s campaign against agency independence is part of his undisguised effort to effectuate a radical dismantling of our checks-and-balances system without actually amending the U.S. Constitution. It goes hand in hand with his efforts to reshape the civil service, praetorian-ize the military, and staff the Administration from top to bottom based on personal fealty, rather than qualifications and proven character.
If President Trump succeeds in neutering the capacity for independent judgment by either individual professionals or agency heads, he will have been enabled by a theory of the constitutional presidency that the Roberts Court has embraced but which ought to be embarrassing in its speciousness. This constitutional reading, now widely known as the “unitary executive theory,” has all but undercut the U.S. Supreme Court’s unanimous 1935 opinion in a case called Humphrey’s Executor v. United States. In that case, the Court upheld the constitutionality of the FTC and held invalid President Franklin D. Roosevelt’s dismissal of an FTC commissioner, William Humphrey, without good cause and in violation of the FTC Act. The Trump Administration has explicitly called for overruling Humphrey’s Executor.
The theory behind Humphrey’s Executor is straightforward. It starts with the recognition that the executive branch of government draws on two streams of legal authority. Some of what it does involves carrying out powers vested directly in the President by the Constitution. Treaty-making and fulfilling the President’s commander-in-chief role are prominent examples. But most of what the executive establishment does—nearly all of what it does in domestic affairs—draws on authority that Congress has given to the executive branch by creating administrative agencies and assigning them missions, such as protecting the environment or enforcing civil rights. The core of independent agencies’ work in this respect involves both rulemaking, which the Humphrey’s Executor Court called “quasi-legislative,” and administrative adjudication, which it called “quasi-judicial.”
What the Court held in Humphrey’s Executor is that if an agency is of the latter kind—that is, the agency’s job description involves a mixture of quasi-legislative and quasi-judicial functions that are not within the President’s explicit Article II powers—then it is up to Congress to determine whether this kind of agency’s heads serve at the President’s pleasure. If such an agency’s role is essentially “to carry into effect legislative policies embodied in statute,” then Congress may protect its members against discharge except for good cause.
Against this entirely commonsense understanding, the unitary executive theory insists on a vision of the Constitution that muddles the text, is weakly grounded in history, and ignores how executive power can easily metastasize into autocracy—a possibility I argued years ago as a hypothetical, but which is now being played out in real time.
The unitary executive theory rests on two foundational premises. The first is that the President, constitutionally speaking, is a one-person executive branch. The President, in the Court’s words, is “the only person who alone composes a branch of government.” The second is that, in vesting “the executive power” in “a president,” the Constitution gave the President the entirety of the government’s executive power— not “some of the executive power, but all of the executive power,” in the words of the late Justice Antonin Scalia, who even italicized the words “some” and “all.”
Beyond these two premises, advocates of the unitary executive theory may differ as to the scope of the precise authorities that Article II confers. But all advocates of the theory share a view that the Supreme Court embraced in a 5-4 decision rendered in 2020, Seila Law v. Consumer Financial Protection Bureau. The idea advanced there was that all executive branch personnel are “subject to the ongoing supervision and control of the elected President.” It follows, according to this theory, that the President must be able, directly or indirectly, to fire anyone in the executive branch. Seila Law extended the President’s removal authority to the principal officer in charge of any single-headed executive agency. The Court held out the possibility, however, that multi-member bodies might remain as an exception to this rule. President Trump wants to overturn the exception, so that he can fire independent board or commission members even without “good cause.”
Given the practical and political implications of the unitary executive theory, it is astonishing to see how little it lines up with the Constitution. First, it cannot be true that Article II gives the President not “some of the executive power, but all of the executive power.” Section 2 of Article II explicitly requires the Senate to participate in the executive powers of treaty-making and appointing so-called principal officers. At most, Article II’s Vesting Clause gives the President whatever executive power is not otherwise constitutionally shared or regulated.
Likewise, the President is not the entirety of the executive branch. One of the President’s Article II authorities is to “require the opinion, in writing, of the principal officer in each of the executive departments, upon any subject relating to the duties of their respective offices.” The executive branch is thus foreseen as involving “executive departments.” With regard to those departments, the President has the duty to “take care that the laws be faithfully executed”—that is, executed by others. The text certainly reads as if “departments” are part of the executive branch.
The unitary executive theory retort is that the “departments” are just assistants to the President, not holders of executive power. But that is not what Article II says. The text just mentioned posits that departments have been assigned “duties”—duties presumably assigned by Congress. It would have been weird to spell out a presidential power to make department heads write out their opinions concerning duties that the President has assigned to them. Thus, although the Trump Administration and the Court refer to agencies as working “on behalf of” the President, they are, constitutionally speaking, working on behalf of Congress.
Finally, if one is looking for a comprehensive grant of power with regard to government operations, it will be found not in Article II, which governs the executive branch, but in Article I, which empowers the legislative branch. Article I authorizes Congress “to make all laws which shall be necessary and proper for carrying into execution … all … powers vested by this Constitution in the government of the United States, or in any department or officer thereof.” The most obvious reading of that language is that decisions as to how and by whom administrative functions shall be fulfilled—and whether administrators can be insulated from at-will removal—are subject to Congress’s determination as to what is “necessary and proper.” In short, read most straightforwardly, the constitutional text lines up perfectly with Humphrey’s Executor.
In the face of the text’s most obvious reading, advocates of the unitary executive theory have amassed what they argue is a historical record to substantiate that some clause in Article II—maybe the Vesting Clause, maybe the Take Care Clause—would have been understood in 1787 as embracing a comprehensive power of at-will presidential removability, even though removal is never mentioned. The historical literature is rich and complex, and it will persuade an open-minded reader that there were likely to be some people in 1787 who would have agreed with the unitary executive theory. But the historical record points much more strongly in the opposite direction, as shown in the work, for example, of Christine Chabot, Julian Davis Mortenson, and Jed Handelsman Shugerman, as well as in my own work. And it is hard to credit the idea that the presidential removal power was central to the founders’ interpretation of Article II when the issue went unmentioned at the Constitutional Convention, there is no indication that it played any role in ratification debates, and the First Congress sometimes placed administrative responsibilities in the hands of persons that the President could not fire.
Indeed, the case is so strong for the correctness of Humphrey’s Executor—a decision regarded in 1935 as a conservative reading of presidential power—that one naturally wonders why, beginning with the Reagan Administration, the unitary executive theory became a pet theory of the right. The most succinct and persuasive answer has been suggested by political scientists William Howell and Terry Moe. As they point out, every President wants to make a big mark and to move the country decisively in the directions promised in a presidential campaign. But such ambitions have different implications for progressive as opposed to conservative Presidents.
Once Congress started enacting modern progressive legislation, especially from the 1960s onward, progressive Presidents, usually Democratic, did not need to make bold constitutional arguments to accomplish their domestic goals. Their administrations could argue instead for generous readings of the statutory powers that Congress had already given them. Should progressive agencies overstep, a court might overturn their statutory readings. But such instances would be examples of ordinary legal error, not threats to the Constitution.
If, however, a President’s agenda is to hollow out government—to incapacitate agencies from implementing their legislative duties in a vigorous way—that President cannot rely on statutory power alone. Congress has not authorized the President to undo the administrative establishment it created. Thus, radically disruptive Presidents of a Trumpian sort have to argue that their “executive power” includes more than the duty to take care that the laws be faithfully executed. They have to argue that, as a one-person branch of government, the President possesses the authority to restructure government, to narrow the reach of law, and to resist efforts by the other branches to check presidential initiative.
Before the Trump Administration, however, the main debate, both judicial and academic, over the unitary executive theory was focused entirely on presidential removal power. The question at hand was just whether Presidents could be limited by statute to discharging principal administrative officers only on grounds of neglect of office, malfeasance, or an inability to do the job to which they had been appointed. But President Trump wants to take it further. His February18 executive order would require those agencies to clear their regulatory actions with the Office of Management and Budget, align their program expenditures with the President’s policy priorities, and meet performance standards and management objectives prescribed by the White House. Such compliance, of course, would be the opposite of independence.
The decision that now stands out most conspicuously as potential support for President Trump’s power grab is not any of the removal cases, but the Court’s 2024 decision on the scope of immunity for former Presidents from criminal prosecution for misconduct while in office. What is most startling about the decision, even to those of us who disagree with it, is not the claim that Presidents have some such immunity or that the Constitution vests directly in the President certain comprehensive authorities. What is most startling is the majority’s view that the President’s power of administrative supervision—an authority that must exist to some extent by virtue of the President being the chief executive—is actually within the President’s arsenal of “exclusive and preclusive powers.” These are the powers that neither Congress nor the judiciary may limit. The Court managed to infer from the President’s obligation to take care that the laws be faithfully executed a seemingly comprehensive power to direct all “those who wield executive power on his behalf.” It is thus through an amazing display of interpretive acrobatics that the Court extracted from the President’s obligation of fidelity to law an all-but-absolute power to ignore it.
As radical as this seems, it is easy enough to see how President Trump is connecting the dots between the one-person executive branch idea and his unprecedented claims of entitlement to control every aspect of administrative government. If the President alone is the executive branch, then any delegation of authority by Congress to an administrative agency begins to look advisory. Congress may want the executive branch to do “something.” It may prefer for that “something” to be the task of a particular agency. But if the President is a one-person branch of government, then, constitutionally, it arguably follows that the President is entitled as possessor of all the executive power to take over an agency’s mission or even assign it to another agency. Indeed, at least some unitary executive theory champions explicitly argue something like this. If that is true, then what any administrator does depends not just on what authority Congress has delegated to the executive. It depends on the President’s willingness to leave that delegation in place and not take over personally. As pithily explained by the sociologist Kim Lane Scheppele, an expert on authoritarianism, “under the unitary executive theory, agencies no longer trace their primary constitutional authority to congressional delegation of its legislative powers but instead to presidential delegation of his executive power.”
This is why President Trump thinks he can close agencies he does not like and put their authorities elsewhere. This is why President Trump thinks he can tell agencies not to spend congressionally appropriated dollars. This is why President Trump thinks he can convert all of the executive branch into an army of lickspittles. If Congress cannot regulate his supervisory power and if his conduct, no matter how corrupt, can never be the target of prosecution, why not?
Of course, the Court may not go that far. But President Trump’s apparent hope that the Roberts Court will give him the control he craves over every corner of the administrative state is not without cause. If a conservative majority is willing to be as creative on his behalf as the immunity opinion suggests, why not push it further?
Yet the case for putting independent agencies under the President’s thumb, even for the Roberts Court, is not a slam dunk. It may be that the Court does not want to undermine the independence of the Board of Governors of the Federal Reserve System; such a decision could destabilize both domestic and global markets. Holding that the United States is unable to have an independent agency controlling the money supply would be an extreme move. Because it is difficult to see how Humphrey’s Executor could be overruled without invalidating the Federal Reserve’s independence, the Federal Reserve may prove the most decisive reason for keeping Humphrey’s Executor alive. President Trump’s executive order tries to navigate this difficulty by controlling only the Federal Reserve’s “supervision and regulation of financial institutions,” but not “its conduct of monetary policy.” But that will not solve the removability problem. Members of the Federal Reserve cannot be half-fired, half-empowered.
It is also noteworthy that, as a judge on the U.S. Court of Appeals for the D.C. Circuit, Justice Brett Kavanaugh offered a strong argument for treating multimember agencies differently from single-headed agencies. Indeed, the account of independent agency virtues with which I started this essay is actually drawn from his opinion. One cannot yet know whether he regards his arguments as sufficient when push comes to shove to draw a constitutional line between single-headed and multi-member agencies for removal purposes. But if he should take that view and at least one other conservative Justice unites with the liberals, Humphrey’s Executor could survive.
The Roberts Court has already shown itself to be the most executive-indulgent Court since World War II. It has enabled President Trump’s dreams of control; only time will tell how far it will go. We know where President Trump is placing his bets. Those who would prefer a more pluralist democracy can hope only that his gambles come to naught.
This essay was first posted in The Regulatory Review, a publication of the Penn Program on Regulation.