In recent months, the Supreme Court has indicated that it will again take up the question of the president’s authority to dismiss a member of the executive branch without cause. In a recent Law & Liberty article, “Removal Power and the Original Presidency,” Ilan Wurman argued that the president “always has the right to remove” members of the executive branch who are appointed by the president to carry out his policies. Things are less clear, however, when the employee has an element of independence associated with his or her appointment. For example, when Congress establishes an independent agency to adopt and pursue policy in a particular area, the president’s removal power should not be unrestricted.
This was the unmentioned issue in Seila Law v. Consumer Financial Protection Bureau. There, the Supreme Court held that the head of any agency with a term of office of several years could be terminated by the president and replaced at any time, without cause. This is the question likely to be soon before the Court, and it is not specifically addressed in the Wurman article.
The CFPB was headed only by a single administrator, appointed for a term of years, but the Court made clear that it would also take up a case of a multi-headed body at a later time. The Court’s decision, however, clearly suggested that its decision for a multi-headed body would not be different. Meanwhile, President Trump—obviously anticipating the forthcoming power to do so—began to remove officials from several multimember bodies such as the Federal Trade Commission, appointed for terms of years, without any claim of deficient performance.
The Court’s position in Seila Law, written by Chief Justice Roberts, was the first of its kind, and the theory the Court was using produced an unusual outpouring of disagreement in the academic community, with many respected constitutional scholars arguing that the Court was wrong to hold that the president had the unbridled authority to dismiss the head of the CFPB or any other executive agency who had been appointed for a term of years specified by Congress.
Out of many academic specialists disagreeing with the Court’s decision, two in particular stood out.
One was Caleb Nelson of the University of Virginia Law School, to whom Ilan Wurman responded. Nelson, a former clerk for Justice Clarence Thomas, wrote:
It is true that Article II [of the Constitution] vests the executive power in the President. But Congress is in charge of creating offices within the executive branch, and the Constitution does not give the President unilateral power to dictate who will fill those offices or what their authorities and duties would be. … To my way of thinking, neither the Vesting Clause nor anything else in Article II compels the inference that after officers have been duly appointed … the President must be able to terminate the appointments … at will, or to dictate how such officers must use any discretion that the law attempted to give them.
Professor Nelson then continued:
If most of what the federal government does on a daily basis is “executive” and if the president must have full control over each and every exercise of “executive” power … then the president has an enormous amount of power—more power, I think, than any sensible person should want anyone to have, and more power than any member of the founding generation could have anticipated.
This is a vitally important point, since we are now living in a time when the president is expanding his power in an unprecedented way; it is important to define the limits that are likely to have been reasonably imposed by the Constitution.
Highly relevant to this question is the work of Boston University Law School Professor Jed H. Shugerman, who described in minute detail the debate in the first Congress over whether the president should have the ability to remove officers of the government at will. Shugerman’s 100-page paper is a minutely detailed description that covers virtually every vote and every relevant statement by a member of the first House and Senate on this issue, during a debate on the president’s removal power that proceeded for several months in the first Congress.
If the Supreme Court were now to adopt the removal idea as its own, it would be contravening the only known will of Congress on the subject—the weakest position for the Court.
According to Shugerman, James Madison pressed the House repeatedly to approve a removal power by the president, but over four months never received an affirmative vote for this idea in either the House or Senate.
The fact that the issue was seriously considered in Congress at one time in the past—and contemporaneously with the states’ approval of the Constitution itself—is significant on the question of the president’s removal power. Up to now, most constitutional scholars had not been exposed to such a detailed review of the debates in Congress on the removal question.
That doesn’t necessarily mean that presidential removal is not appropriate in any circumstance—especially where the officials to be removed would not or could not carry out the president’s specific directions or policies.
But it also means, even more significantly, that if the Supreme Court were now to adopt the removal idea as its own, it would be contravening the only known will of Congress on the subject—the weakest position for the Court—and, as the Court has probably thought, deciding an issue of purely constitutional interpretation. In this case, however, the Court would in effect be overturning an actual refusal of Congress to do what the Court is now doing.
Whether the members of the Court would feel comfortable proceeding with this is a serious question that the Court’s proponents of unrestricted removal should address.
After his detailed analysis of the debates in the first Congress, Professor Shugerman notes that in Free Enterprise Fund v. Public Company Accounting Oversight Board, and again in Seila Law, Chief Justice Roberts supported his position by relying on self-serving letters that Madison had written after he lost the votes in Congress.
Shugerman writes: “only nine of the fifty-three participating members of the House explicitly endorsed even the weaker version of the presidentialist interpretation of Article II: a presidential removal power without resolving whether Congress could set conditions. … Even fewer suggested that the President had a constitutional power to remove at pleasure.”
Ironically, as Shugerman notes in his paper, Madison’s efforts, although never coming close to approval in either House in the first Congress, only succeeded two centuries later—as the Roberts Court, Chief Justice Roberts himself, and other supporters of a presidential removal power—came to rely on Madison’s letters about the debates in Congress, with no reference to how the debates themselves actually turned out.
This is not the only time that the Court has failed to understand the proceedings in the late 1780s. At the Constitutional Convention, the members were so split on the question of whether to have a president who was an individual (instead of a group) and elected by the people instead of Congress, that this issue was the last major item decided before the convention adjourned. Many delegates did not want an elected president, for fear that he would use that support to become a monarch, an issue that is relevant even today. The issue was finally resolved by creating the Electoral College, which, in its original form (valid until 1804), did not have anything to do with the popular vote, but just the presidential preferences of some elites at the time.
Yet, in Seila Law, the Court said, “The Framers made the president the most democratic and politically accountable official in the government. Only the president (along with the vice president) is elected by the entire nation.” This was not true, for the reasons noted above, until 1804, when a new Electoral College was created to follow the popular vote in each state. And it’s not even literally true now: in 15 cases since 1844—including 2024—the College has elected a president who won the electoral vote, but who had not received a majority of the popular vote. Moreover, in the scramble at the convention to finish up in 1787, there was no discussion of whether the president would be “politically accountable,” let alone “democratically elected.” The Court’s history here is way off base.
Nevertheless, because of the Court’s error, Madison, who was defeated in Congress, might still win in the Supreme Court 238 years later.
This essay has been updated to correct a misquotation of Ilan Wurman’s “Removal Power and the Original Presidency.”