In a recent paper, Professor John C. Harrison remarks that “[p]erhaps the most famous example of [presidential] removal for policy purpose was President Jackson’s replacement of Secretary of the Treasury William Duane with Roger Taney.” John C. Harrison, The Unitary Executive Without Inherent Presidential Removal Power 43 n.171 (2022). This might understate matters because, depending on what one means by “policy purpose,” Jackson’s removal of Duane might be the only example of such removal in American history.
Professor Prakash identifies Andrew Jackson, not coincidentally, as the originator of the concept of the popular mandate, which Prakash blames for the pernicious “reconception of the office” from a chief executive focused on faithful execution of the law to the claimed embodiment of the popular will. See The Living Presidency at 75, 77-78.
Surprisingly, though, Prakash only alludes to the episode at the heart of this reconception. He notes that Jackson justified his decision to withdraw funds from the Bank of the United States by reference to the mandate allegedly given by the people in 1832, when Jackson was reelected after campaigning against the Bank. See id. at 75. But the constitutional clash between Jackson and his congressional critics (led by the Senate’s Great Triumvirate of Daniel Webster, Henry Clay, and John C. Calhoun) did not center on the fact money had been withdrawn from the Bank, but how it had been withdrawn.
Congress had delegated the authority to remove funds from the Bank to the secretary of the treasury, not to the president. Accordingly, Jackson asked Secretary of the Treasury Duane to remove the funds, a “request” he clearly expected to be honored. But Duane hesitated. He maintained, reasonably enough, that the law provided that only the treasury secretary could remove the funds and therefore it was his responsibility to make the decision. Strauss, 75 Geo. Wash. L. Rev. at 706; Jon Meacham, American Lion 257, 268 (2008). Jackson countered that the president could direct him to remove the deposits, thereby taking it on the president’s responsibility. Meacham, American Lion at 268. Duane thought this theory would reduce him (and presumably all heads of department) “to a mere cipher in the administration.” Id. at 258.
As a matter of unitary executive theory, the situation raises several issues. Taft might ask whether the statute “peculiarly and specifically committed to the discretion” of the treasury secretary the issue at hand. Arguably, the statute here perhaps implicitly required the decision to be made based on financial prudence and other factors within the treasury secretary’s expertise, but the case against presidential interference would have been stronger had the statute expressly set forth what factors should govern the decision to withdraw funds. Put another way, if the statute allowed the secretary to withdraw funds for any reason, it would be more constitutionally problematic for Congress to attempt to insulate this decision from presidential control.
On the other hand, Prakash presumably would contend that the president must be able to direct the treasury secretary in the exercise of his lawful discretion, regardless of congressional intent or the nature of the decision. Professor Harrison, who contends that the unitary executive requires the president to control all “policy decisions,” a term he does not define but which apparently would have included the decision to withdraw funds from the Bank, seems to have the same view. The current Supreme Court appears to agree, as the secretary would be exercising executive power which belongs to the president. Unless, of course, it falls under a “special arrangement sanctioned by history” such as “that monetary policy should not be subject to political interference.” See Trump v. Cook, 609 U.S. __, slip op. at 23-24 (2026) (insert shrug emoji).
All of these perspectives, however, assume that the secretary is being directed to exercise his discretion in a lawful manner. Harrison states that if it were not lawful to withdraw the funds from the Bank, as Duane evidently believed, then Jackson’s direction to Duane would itself be illegal. Prakash possibly believes the same, but he also contends that “the Constitution authorizes the President to superintend and control the legal interpretations of his mere instruments.” Too Unitary, 135 Yale L. J. F. at 553. It would appear, therefore, that Duane would be bound to accept Jackson’s determination that the statute allowed withdrawal of funds from the Bank. And even if Duane were permitted to adhere to his own interpretation, it is not clear what remedy would be available to allow him to resist Jackon’s illegal order.
In any event, Jackson neither knew nor cared about these distinctions, and he responded to Duane’s recalcitrance by firing him and recess appointing Taney to take his place and carry out the order. This is what set off the firestorm in Congress. No one disputed that Jackson had the formal power to fire Duane, but many (including Webster, Clay and Calhoun) argued that it was improper to exercise that power as a means of compelling obedience to the president’s order to withdraw funds from the Bank. Furthermore, they warned against the consequences of an inherent presidential removal power, particularly when combined with Jackson’s claim of a popular mandate and his expansive view of his authority to direct executive officers.
Webster, for example, condemned “an absolute and perfectly irresponsible removing power [which] tends to turn the whole body of public officers into partisans, dependents, favorites, sycophants, and manworshippers.” Myers, 272 U.S. at 179 (McReynolds, J., dissenting). Clay denounced Jackson’s claim of an inherent removal power as tending toward monarchy and despotism. Id. Jackson’s actions, he contended, would lead to “a total change in the pure Republican character of the Government, and to the concentration of all power in the hands of one man.” Meacham, American Lion at 275. Calhoun likewise protested Jackson’s actions as displacing Congress and consolidating the powers of the government in the presidency. Id. at 276-77.
As we have discussed before, Justice Joseph Story expressed similar reservations about the removal power, and he warned against extending it to inferior officers. He wrote in his classic constitutional treatise that “if this unlimited power of removal does exist, it may be, in the hands of a bold and designing man, of high ambitions, and feeble principles, an instrument of the worst oppression, and most vindictive vengeance.” 3 Joseph Story, Commentaries on the Constitution § 1533, at 390 (1833). He cautioned that “such a prerogative in the executive was in its own nature monarchical and arbitrary,” and it “would convert all officers of the country into the mere tools and creatures of the president.” Id.
The logic of these eminent constitutional thinkers is unmistakable. Congress by law creates offices, and it assigns to those offices certain powers and duties. When the president nominates someone to fill an office, the Senate evaluates the nominee to determine whether she has the capability and character to faithfully interpret and execute the law as it pertains to her office. All of this might be of little moment, however, if once confirmed the officer is in fact and in law a mere tool or creature of the president, a “cipher” whose role is simply to serve as the instrument of the president’s will.
Fortunately, the warnings of Story and the Great Triumvirate did not come to be fully realized at that time. This may have reflected in part the substantial political cost that Jackson suffered from pursuing his war against the Bank of the United States, including a censure by the Senate, which declared that Jackson “in the late executive proceeding in relation to the public revenue, has assumed upon himself authority and power not conferred by the Constitution and laws, but in derogation of both.” The Senate’s reaction demonstrated that Congress did not acquiesce in the exercise of a directive power regarding duties specifically delegated to an officer. Furthermore, while in theory some executive branch lawyers (such as Taney) may have maintained that the president could direct an executive officer in the lawful exercise of his discretion, that still left the president without any authority to order the officer to act illegally. At the very least, the president would be in a politically precarious position if a subordinate contended he was without legal authority to comply with the president’s direction. This perhaps explains why there were few if any instances where Jackson’s successors attempted to claim overtly a power to direct executive officers in the exercise of an authority expressly committed to them by statute.
Even where presidential direction has generally been understood as permissible, strong norms could ensure that presidents exercised their power with great circumspection. For example, historically presidents have had, but rarely exercised, the power to direct federal prosecutors with regard to specific criminal or civil cases. This formal power (widely if not universally acknowledged) co-existed with norms of prosecutorial independence that developed over time. See Bruce Green & Rebecca Raiphe, Can the President Control the Department of Justice?, 70 Ala. L. Rev. 2, 69 (2018) (“The tradition of prosecutorial independence has evolved from the inchoate reality of prosecution in the early republic into a way to protect against partisan corruption and, finally, as a check on presidential power.”).
The tension between formal constitutional power and norms of independence was at the heart of another famous episode in which a president used the removal power to coerce an executive officer to take a particular action (though arguably not one involving “policy discretion”). In the spring of 1973, President Richard Nixon nominated Elliot Richardson to serve as attorney general, and during his confirmation hearings Richardson promised the Senate that if confirmed he would appoint Archibald Cox as special prosecutor and guarantee him complete independence to investigate the Watergate scandal.
Subsequently, Nixon attempted to get Cox to drop a grand jury subpoena for White House tapes relevant to the Watergate investigation. When Cox refused, Nixon ordered first Richardson, then Deputy Attorney General William Ruckelshaus, to fire Cox; both refused and resigned. Finally, Solicitor General Robert Bork, as acting attorney general, agreed to fire Cox. This incident, known as the “Saturday Night Massacre,” ignited a series of events which led ultimately to Nixon’s resignation.
As in Jackson’s fight against the Bank, there was no question that the president had the power to remove the attorney general and deputy attorney general. The issue was whether the president could properly use this power to coerce executive officers to take actions within their authority, not the president’s. In the case of the Saturday Night Massacre, those actions included both firing Cox and withdrawing the grand jury subpoena. It was debatable whether Nixon had the authority, for example, to direct Cox in the exercise of his prosecutorial discretion with regard to the subpoena. See Green & Roiphe, 70 Ala. L. Rev. at 20 n. 90. But even if Nixon technically had this power, Congress and the public viewed its exercise as violative of longstanding norms, particularly in an investigation where the president’s personal interests were implicated.
It should be noted that the legal position staked out by Prakash goes beyond what either Jackson or Nixon claimed. According to Prakash, Jackson could have simply bypassed Duane and directly ordered the withdrawal of federal funds from the Bank of the United States. Nixon likewise could have directly ordered the withdrawal of the grand jury subpoena and/or fired Cox personally.
It might be contended that there is little difference among (1) a president who can remove at will, but cannot direct a specific decision; (2) a president who can direct a specific decision but cannot take the action himself; and (3) a president who can take an action directly if an officer refuses to obey. But the farther along this continuum one goes, the more difficult it is to sustain a norm of independence. See Strauss, 75 Geo. Wash. L. Rev. at 712 (explaining how the psychology of office changes if an officer takes the president’s views as “a command that she has a legal as well as political obligation to honor, and for whose justifications she thus has no particular responsibility.”).
Historically there has been substantial uncertainty regarding the president’s authority to direct executive officers in the exercise of their lawful discretion. Presidents undoubtedly have had the ability to wield substantial influence over the decisions of officers they appoint and can remove at will, but they have faced substantial political and legal risks if they sought to direct an officer who was adamantly opposed to taking the action in question. These risks were heightened if the officer contended that the action was unlawful. And before Humprey’s Executor was undermined in the last two decades, and ultimately overruled this year, Congress retained the ability to substantially insulate many governmental functions from presidential interference.
The most important consequence of the unitary executive is not to require at will presidential removal, but to change the psychology of executive office from being an agent of the law to being an agent of the president. We will look at the dynamics of that change in the next post.