Starting in the late 1880s with the Interstate Commerce Commission (ICC), Congress established “independent agencies” outside the executive branch. By one count, there are about 40 such agencies in existence today. (Seven more, including the ICC, were created and later abolished.) Familiar examples include the Federal Communications Commission, the Federal Reserve System, and the Federal Trade Commission.

Unlike executive branch officials, many independent agency officials are protected from removal by the President except for “inefficiency, neglect of duty, or malfeasance in office.” Disagreements with the President’s goals and policies are not enough to justify dismissing protected agency officials. In addition, Congress has given 19 of these agencies the authority to make regulations without oversight from the President or other officials of the executive branch.

For nearly a century, Congress, presidents, and the courts have struggled over the constitutional legitimacy of independent agencies. Since taking office for the second time on January 20, 2025, President Trump has appeared determined to bring this controversy to a head. He has fired agency officials without alleging that they are inefficient, neglectful of their duties, or guilty of malfeasance, and he has signed an executive order directing independent agencies with regulatory authority to submit proposed regulations for review by the Office of Management and Budget (OMB). The legal disputes triggered by these moves will almost certainly end up in the Supreme Court, as the Trump Administration intends, and the Court’s decisions could dramatically expand the scope of presidential powers. This outcome would exacerbate the existing imbalance in our constitutional system, where a weak Congress has yielded many of its powers to the President. It would also reduce agencies’ ability to offer informed and independent judgments at a time when partisan polarization narrowing the scope for such arguments in legislative debates. In short, if the Court finds for the president, governance would suffer, and presidents could exercise power with dangerously few checks and balances.

The State of the Law

The modern era of arguments about independent agencies and executive power began nearly a century ago with the case of Myers v. United States (1926). A law passed in 1876 had protected several categories of postmasters (a position appointed by the President with the advice and consent of the Senate) from dismissal without the consent of the Senate. Nevertheless, President Wilson fired Frank Myers, a postmaster, prior to the expiration of his guaranteed term of office. Myers sued, seeking backpay. Writing for the majority of the Supreme Court, Chief Justice (and former President) William Howard Taft held that the law guaranteeing Myers’s tenure in office was an unconstitutional infringement on the President’s authority to remove officials within the executive branch who, like postmasters, were appointed by the President with the Senate’s advice and consent—an authority the Court inferred from the Appointments Clause and a detailed review of historical practice. Going beyond what was needed to resolve the case before him, the Chief Justice opined that the President’s constitutional power of removal extended, as well, to officials who had responsibilities of a “quasi-judicial character.” As we’ll see, his opinion evoked powerful dissents from the justices in the minority.

A decade later, in Humphrey’s Executor v. United States (1935), the Court limited the scope of the Myers decision. In 1933, William Humphrey, a Senate-confirmed member of the Federal Trade Commission (FTC) for a seven-year term expiring in 1938, received a letter from President Franklin Roosevelt requesting his resignation on the grounds that he did not agree with and would not carry out the President’s policies. When Humphrey refused to resign, FDR fired him. Humphrey sued for backpay and, after his death, his estate continued to pursue his claim.

In a unanimous decision, the Court found for Humphrey. The Myers decision, said the Court, applied only to “executive officer[s] restricted to the performance of executive function,” such as postmaster. By contrast, the FTC was created to exercise “quasi-legislative” and “quasi-judicial” powers to assist the legislative and judicial branches of government. In the Court’s view, Congress had the constitutional authority to create such independent agencies and to protect their commissioners against removal prior to the end of their terms of office except for “cause.” Disagreement with the President’s policies was not a legitimate basis for removal. This protection was the only way to ensure that the FTC could exercise independent judgment, which would be impossible if its commissioners served at the President’s pleasure.

For decades, Humphrey’s Executor served as both charter and shield for independent agencies. In 1958, for example, the Supreme Court ruled unanimously in Wiener v. United States that President Dwight Eisenhower did not have the constitutional authority to remove a member of the War Claims Commission except for good cause. Much like FDR, President Eisenhower had insisted that the “national interest” required him to fill the commission with “personnel of my own selection.” Citing Humphrey’s Executor, the Court told him that he lacked the authority to do so. In an opinion that tartly criticized Taft’s reasoning in Myers, Justice Felix Frankfurter insisted that the right of Congress to establish independent agencies depended on the “nature of the function” that Congress vested in them. Congress had the right to secure the independence of the War Claims Commission because its task was of “intrinsic judicial character.”

There matters stood for the next two decades. But in the wake of Vietnam and Watergate, Congress began pushing back against what many commentators of the day called the “imperial presidency.” Oversight hearings were launched against alleged abuses of presidential power, and laws were passed to trim these powers, international and domestic. During the Nixon, Ford, and Reagan presidencies, senior administration officials pushed back against these restrictions. Allegations of excessive presidential power were countered by laments for what conservative officials called the “fettered presidency,” the title of an influential volume of essays published by the American Enterprise Institute in 1989. In the same year, Justice Antonin Scalia’s famous dissent in Morrison v. Olson argued that independent agencies violated the bedrock constitutional principle of separation of powers. (More on this dissent later.)

When George W. Bush took office in 2001, he brought with him strategically placed lawyers who had developed Scalia’s critique into a theory of the “unitary executive,” one of whose tenets was that Myers had been correctly decided and that subsequent decisions had wrongly restricted the President’s powers over independent agencies. Within a decade, in cases such as Free Enterprise Fund v. Public Company Accounting Oversight Board (2010), the influence of this theory became apparent. And in the pivotal case of Seila Law v. Consumer Financial Protection Bureau (2020), the theory had come to function as the Court majority’s bedrock assumption.

Writing for the majority in Seila, Chief Justice Roberts asserted that Myers had correctly stated the general principle that the Constitution gives the President broad removal powers and recognizes (at most) narrow exceptions. He questioned the Court’s depiction of the FTC in Humphrey’s Executor as legislative and judicial rather than executive, insisting that “[t]he Court’s conclusion that the FTC did not exercise executive power has not withstood the test of time.”

The Court held that the Consumer Financial Protection Bureau (CFPB) was unconstitutional on the narrow ground that it was headed by a single director for a fixed term who could be removed by the President only for cause, not at will. Still, the Chief Justice based his opinion on “our constitutional structure,” with which the CFPB was “incompatible.” As a general matter, he argued, the Constitution avoids concentrating power in the hands of a single individual—except for the presidency. The founders’ strategy was straightforward, Roberts wrote: “divide power everywhere except for the Presidency, and render the President directly accountable to the people through regular elections.” Any measure that divides executive power is presumptively illegitimate and must discharge a heavy burden of justification. “The President’s removal power,” said Roberts, “is the rule, not the exception.” Against this backdrop, he framed the two main holdings to the contrary as exceptions allowing Congress to establish for-cause removals for certain narrow classes of officer: in Humphrey’s Executor, for those serving on multi-member commissions that are engaged in quasi-legislative and quasi-judicial but not executive functions; in Morrison v. Olson, for certain inferior officers occupying offices of limited tenure and exercising limited authority and jurisdiction.

To many observers, the Chief Justice’s language suggested that these exceptions were now hanging by a thread. “We do not revisit Humphrey’s Executor or any other precedent today,” he observed [emphasis added]. But in the immortal words of Scarlett O’Hara, tomorrow is another day.

On March 1, 2025, U.S. District Court Judge Amy Berman Jackson ruled that President Trump’s firing of Hampton Dellinger, the head of the Office of Special Counsel (OSC), was illegal. Among the many reasons offered in a 67-page decision, she noted that the majority in Seila had explicitly distinguished between the CFPB that it was addressing and the OSC, on the grounds that the latter exercised narrower and less clearly executive powers. The Trump Administration was asking the Court to expand the President’s removal powers beyond Seila, a step for which she saw no compelling rationale. The Administration challenged Judge Berman’s decision in the U. S. Court of Appeals for the District of Columbia and, through the acting solicitor general, announced its intention to challenge the constitutionality of Humphrey’s Executor. After a three-judge panel of the appeals court paused the district court’s ruling, Dellinger withdrew his suit.

The abrupt termination of this case did not end the battle over the Administration’s challenge to congressional restraints on the President’s removal power, however. After taking office, Trump removed Gwynne Wilcox from the National Labor Relations Board without giving a reason, although the law allows members of this board to be removed only for misconduct. In her decision blocking his action, U.S. District Judge Beryl Howell stated that the President “seems intent on pushing the bounds of his office and exercising his power in a manner violative of clear statutory law to test how much the courts will accept the notion of a presidency that is supreme.”

The Trump Administration knows that the ice on which it is stepping may not be thick enough to bear the weight of the policy of the changes it proposes. For example, Trump’s executive order subordinating independent agencies to the President specifically exempts the Board of Governors of the Federal Reserve System and the Federal Open Market Committee (FOMC) “in its conduct of monetary policy.” But this exemption seems to concede the principle that in some cases, Congress is within its constitutional rights to establish a body whose members are insulated from presidential removal and delegate to this body some of the most consequential choices that the federal government can make. If so, what is the legal principle that distinguishes the FOMC from other formally similar administrative entities?

This is the backdrop against which the cases sparked by President Trump’s removal of independent agencies’ leaders will reach the Supreme Court. Claims and counterclaims will multiply. History and constitutional principles provide the basis for understanding the merits of competing arguments, but they are unlikely to offer unequivocal guidance.

What History Teaches

The fundamental questions at issue in this struggle go back to the beginning of the republic. Article II, section 2 of the Constitution spells out how the authority to appoint officials is allocated between the President and Congress, but the Constitution is silent on the question of who has the power to remove these officials. The dispute over this issue emerged in 1789, during the first Congress convened under the Constitution. Because many of the representatives in this Congress had been delegates to the Constitutional Convention of 1787, judges and historians have given great weight to their views. But the historical record is anything but straightforward.

The participants in this debate disagreed with one another, and often with themselves. James Madison initially believed that the Constitution’s failure to specify the removal power had left the matter for Congress to determine, only to change his mind during the debate and conclude that the logic of the Constitution gave the President this power. In Federalist 77, Alexander Hamilton had argued that in cases when the Senate’s consent was required to make an appointment, it would also be required for removal. During the debate, however, Hamilton (now a member of President Washington’s Cabinet) had let it be known to his friends in Congress that he had changed his mind and supported the President’s power to remove executive branch officials without Senate consent. During the debate, four radically different theories emerged, and the interaction among them was mind-bogglingly complex. Congress eventually decided that the President enjoyed unfettered removal power over executive branch officials, but the disputants’ disagreement over the basis for this decision has resonated down to the present.

In Myers, Chief Justice Taft reviewed at length the records of the Constitutional Convention and the first Congress, concluding that they spoke clearly in favor of the President’s unfettered power to remove executive branch officials. Writing in dissent, Justice James McReynolds examined the same history and concluded that it provided no basis for the President’s alleged unfettered removal power. Another dissenter, Louis D. Brandeis, observed that in the foundational case of Marbury v. Madison, Chief Justice John Marshall “assumed, as the basis of decision, that the President, acting alone, is powerless to remove an inferior civil officer appointed for a fixed term with the consent of the Senate; and that case was long regarded as so deciding.”

Writing three decades later for a unanimous Court in Wiener v. United States (1958), Justice Felix Frankfurter dismissed Taft’s historiography as largely irrelevant to the issue before the Court. Writing in dissent six decades later, Justice Elena Kagan again reviewed the historical record in detail and ended up endorsing a leading scholar’s conclusion that the implications of the historical record are “highly ambiguous and prone to overreading.” A year later, law professor Saikrishna Prakash, a conservative, once more reviewed the record, disagreeing with Kagan and concluding that the debate during the Founding Era had yielded a clear majority agreement on the constitutional principles underlying an expansive presidential removal power and that “Chief Justice Taft was right all along.”

Arguing First Principles

Given the ambiguity of the historical record, it is not surprising that proponents of presidential control over independent agencies often base their arguments on what they consider to be basic constitutional principles, brought together under the theory of the unitary executive.  The core claim is that the Constitution assigns responsibilities to the President that cannot be discharged unless the President has wide-ranging constitutional powers to control the executive branch.

According to Article II, section 3, the President has the duty to “take care that the laws be faithfully executed.” To this end (among others), “The executive power shall be vested in a President of the United States” (Article II, section 1). Backers of the unitary executive theory point to the difference between this vesting clause and the parallel clause in Article I, which vests “all powers herein granted” in the Congress. There are, they conclude, inherent executive powers beyond those specified in Article II, including control over the executive branch. When Congress creates islands of power insulated from presidential control, it limits the President’s ability to fulfill his “take care” duties, in violation of the Constitution.

Justice Scalia’s vigorous dissent in Morrison v. Olson offers the best-known articulation of the unitary executive thesis. Scalia begins by quoting from the Massachusetts Constitution of 1780:

In the government of this Commonwealth, the legislative department shall never exercise the executive and judicial powers, or either of them; The executive shall never exercise the legislative and judicial powers, or either of them; the judicial shall never exercise the legislative or executive powers, or either of them; to the end it may be a government of laws, and not of men.

Scalia asserts that the Framers of the Constitution embraced the principle of separation of powers—understood as the drafters of the Massachusetts Constitution did—as the “central guarantee of a just government,” and he summons James Madison as a witness to this effect. In Federalist 47, Scalia notes, Madison wrote of separation of powers that “No political truth is…of greater intrinsic value, or is stamped with the authority of more enlightened patrons of liberty.” On this basis, Scalia proceeds to his key argument: When Article II, section 1 of the Constitution states that “The executive power shall be vested in a President of the United States,” this does not mean “some of the executive power, but all of the executive power.” If executive power is divided in any way, the President loses the power to defend his office against the encroachments of the other branches, risking a fundamental disruption of the balance among the branches that the Framers believed was the best guarantee of liberty. For this reason, matters of degree do not matter. “[I]t is ultimately irrelevant,” Scalia declares, “how much [a] statute reduces Presidential control” over the executive branch. Efforts to balance the dilution of presidential control against other goals produce a slippery slope of arbitrary all-things-considered judicial judgments for which no clear rule or principle can be offered.

In response, it might be argued that Scalia fundamentally misread Madison, who wrote Federalist 47 to defend the Constitution against the charge that it had commingled executive, legislative, and judicial powers in violation of the separation principle. Madison argues that the Constitution’s critics did not properly understand this principle. It does not mean that the branches ought to have no “partial agency in, or control over, the acts of each other,” but rather that liberty is subverted where the whole power of one department is exercised by the same hands which possess the whole power of another department.” In short, Madison affirms what Scalia denies, that separation of powers is a matter of degree.

To drive home this point, Madison proceeds to examine the constitutions of the states that came together to form the union. While all of them affirm the principle of separation of powers, he observes, some in “unqualified terms,” there is no state that has kept the branches of government “absolutely separate and distinct.” He quotes the section of the Massachusetts Constitution with which Scalia began his dissent, showing with a wealth of examples that despite the absolutism of the Massachusetts statement, its institutions embody a “partial mixture of powers.” In Federalist 48, finally, Madison wrote that the separation of powers didn’t require that the three branches of government be “wholly unconnected with each other.” On the contrary, they needed to be “blended” so as to give each “constitutional control over the others.”

If we take Madison as our guide, Scalia erred by taking an abstract proposition as the basis for constructing and construing the Constitution we actually have.

Consider the “take care” clause, from which some might infer that faithfully executing the law is the sole province of the executive branch. It is not, of course. Congressional oversight can identify lapses in implementing the law, and the judiciary determines whether the executive branch has overstepped legal boundaries. In a similar vein, one might think that the ability to appoint top-level executive officials such as the heads of departments is at the heart of executive power. But the Constitution subjects the President’s nominations of these officials to senatorial consent. The inescapable conclusion: Separation of powers is not the same as separation of functions. In our constitutional design, as Madison emphasized, we have the former but not the latter.

A final complication: The language of the vesting clause—“the executive power”—is sonorous but not self-defining. In an 1835 address on the appointing and removal power, Senator Daniel Webster posed the fundamental question:

It is true, that the Constitution declares that the executive power shall be vested in the President; but the first question which then arises is, What is executive power? What is the degree, and what are the limitations? Executive power is not a thing so well known, and so accurately defined, as that the written constitution of a limited government can be supposed to have conferred it in the lump.

Webster enumerated a long list of different definitions of executive power, each plausible, each with a distinguished pedigree, none clearly superior or self-evidently closer to the founders’ meaning. It is hard to see how conclusions in specific cases can be drawn from this abstract and contested concept. Scalia’s core thesis—that the vesting clause granted all executive power to the President—invites Webster’s retort: Very well, but when you use the term “executive power,” what do you mean, and what evidence supports your choice?

The Most Dangerous Branch

The debate over the President’s removal power may seem theoretical, but the real-world stakes are enormous. As Webster warned, “an absolute and perfectly irresponsible [presidential] removal power…tends to turn the whole body of public officers into partisans, dependents, favorites, sycophants, and man-worshippers.”

Judge Joseph Story, the author of perhaps the most influential commentary on the Constitution in U.S. history, went even further. If an unlimited power of removal is conceded, he wrote, “it may be made, in the hands of a bold and designing man of high ambition and feeble principles, into an instrument of the worst oppression and most vindictive vengeance.”

Ignoring these warnings would be risky in the best of times, but a change in our constitutional order that Madison did not anticipate makes them especially pertinent today. Scalia’s dissent quotes Madison’s analysis in Federalist 51: “In a republican government, the legislative authority necessarily predominates.” Its influence, therefore, must be weakened—by dividing it into different institutions, each with its own powers and modes of election. “As the weight of the legislative authority requires that it should be thus divided,” Madison continues, “the weakness of the executive may require…that it should be fortified.” This is why the Framers refused to divide executive power in any way, Scalia argues.

But what may have been true in Madison’s time is not so in ours. As countless scholars and pundits have shown, the ability of Congress to act effectively has declined dramatically in recent decades. Partisan polarization has all but paralyzed the legislative branch, and it has handed over many of its powers to the executive, including the authority to declare emergencies and to decide issues, such as the level and timing of tariffs, that were once at the center of congressional debate.

The judiciary, Hamilton declared, was the “least dangerous branch.” But today, the executive is the most dangerous branch.  This is one reason why the resolution of an apparently abstruse issue—the constitutional standing of administrative agencies—matters. Expanding the power of the President from today’s elevated level does not seem likely to restore a sustainable balance among the branches of our federal government.

We must also think about the ways in which presidents may use the expanded power that this Supreme Court could soon award them. No doubt expertise is under attack in today’s supercharged populist environment, and reasonable people can argue that experts have been given too much deference in recent policy decisions.

Still, presidents are always tempted to undermine expertise when it impedes their purposes. They often want lower interest rates during election years, whether these rates are conducive to the long-term well-being of the economy and the country or not. With unemployment rising a year before the 1972 election, Richard Nixon pressured Federal Reserve Board Chair Arthur Burns into lowering interest rates, a move that many economic historians believe contributed to the hyperinflation that dominated the rest of the decade.

There is, finally, the question whether any government in a highly developed economy and society can do without a wide range of administrative competence that functions to some extent independently. Both the legislative and executive branches seem overwhelmed by their current responsibilities. Would the affairs of state go better if administrative independence were trimmed or eliminated? As Alexander Hamilton, no foe of a strong executive, remarked in Federalist 68, “[T]he true test of a good government is its aptitude and tendency to produce a good administration.”

No doubt reforms of current practices are long overdue. But these reforms should be carried out thoughtfully and deliberately. For reasons of prudence, all branches of government should think twice before taking steps that could hamper the ability of independent agencies to contribute to the good administration Hamilton rightly prized.

These considerations apply in every presidency, but they are especially compelling today.  While most presidents chafe at limits to their authority, Donald Trump has displayed an unprecedented appetite for eliminating restraints on his actions.  In 2019, he declared that Article II of the Constitution gives him “the right to do whatever I want as President.”  Because there is no reason to believe that he has changed his mind, it would be especially damaging for the Court to remove the checks on executive power that prevent tyranny and safeguard liberty.