Symposium – Executive Overreach and the Rule of Law in Trump II

The Removal Question: A Timeline and Summary of the Legal Arguments

Aditya Bamzai & Peter M. Shane *

A Note for Readers: Adversarial collaborations are written by scholars who hold opposing views on their topic—together, they write one Essay to clarify points of agreement, precisely identify areas of disagreement, and engage with the evidence supporting each side. These Essays aim to elevate the quality of debate and clarify precisely where viewpoints diverge.

Few constitutional issues have been as significant to the first 100 days of the second Trump Administration as the question whether the President may fire the heads of agencies subject to statutory protection from removal. The underlying issue—whether Article II confers on the President a power to remove executive subordinates that Congress cannot take away—has a claim to being one of the oldest constitutional debates in American law. The members of the First Congress addressed the question at length in 1789, and the political branches repeatedly debated it well before the Supreme Court decided any cases on the topic. 1 See Part I. The Supreme Court cases that followed these political debates—Myers v. United States, 2 272 U.S. 52 (1926). Humphrey’s Executor v. United States, 3 295 U.S. 602 (1935). and Morrison v. Olson, 4 487 U.S. 654 (1988). to take a few—are well-known and significant precedents of American constitutional law. More recently, in Seila Law LLC v. Consumer Financial Protection Bureau 5 140 S. Ct. 2183 (2020). and Collins v. Yellen, 6 141 S. Ct. 1761 (2021). the Supreme Court has limited the reach of Humphrey’s Executor and Morrison and has reembraced part of the reasoning of Myers. 7 See, e.g., Seila Law, 140 S. Ct. at 2197-2200.

The debate over the wisdom of these steps has been ongoing and has reemerged with ferocity in the Trump Administration’s early days. In a letter dated February 12, 2025, the then-Acting Solicitor General advised Senators that the Department of Justice had “determined that certain for-cause removal provisions that apply to members of multi-member regulatory commissions are unconstitutional and that the Department will no longer defend their constitutionality.” 8 Letter from Sarah M. Harris, Acting Solic. Gen., to Richard J. Durbin, Ranking Member of the U.S. Senate Judiciary Comm., Re: Restrictions on the Removal of Certain Principal Officers of the United States 1 (Feb. 12, 2025). The letter argued that, “[a]s presently constituted, [various independent] commissions exercise substantial executive power, including through ‘promulgat[ing] binding rules’ and ‘unilaterally issu[ing] final decisions . . . in administrative adjudications.’” 9 Id. at 2 (quoting Seila Law, 140 S. Ct. at 2199-2200). And the letter said that, to the extent Humphrey’s Executor permits such removal restrictions, the Department of Justice “intends to urge the Supreme Court to overrule” the case. 10 Id. at 2. In the weeks that followed, President Trump dismissed a number of officials subject to removal limits of various kinds, leading to significant litigation. At the time of this Essay, lower courts have addressed the removal provisions governing a number of these officials and agencies. 11 These cases are: Dellinger v. Bessent, 768 F. Supp. 3d 33 (D.D.C. 2025) (addressing removal restriction on the Special Counsel of the Office of Special Counsel), stay granted pending appeal, 2025 WL 887518 (D.C. Cir. Mar. 10, 2025); Harris v. Bessent, No. 25-412, 2025 WL 679303 (D.D.C. Mar. 4, 2025) (addressing removal restriction on member of the Merit Systems Protection Board); Wilcox v. Trump, No. 25-334, 2025 WL 720914 (D.D.C. Mar. 6, 2025) (addressing removal restriction on member of the National Labor Relations Board); Grundmann v. Trump, 770 F. Supp. 3d 166 (D.D.C. 2025) (addressing removal restriction on member of Federal Labor Relations Authority); U.S. Instit. of Peace v. Jackson, No. 25-cv-804, 2025 WL 1428641 (D.D.C. May 19, 2025) (addressing structure of U.S. Institute of Peace); LeBlanc v. U.S. Priv. & C.L. Oversight Bd., No. 25-542, 2025 WL 1454010 (D.D.C. May 21, 2025) (addressing structure of the Privacy and Civil Liberties Oversight Board). Following the initial flurry of litigation, the Supreme Court stayed two separate district court decisions enjoining the President’s removal of members of the National Labor Relations Board (NLRB) and Merit Systems Protection Board (MSPB), without citing Humphrey’s Executor. The Court contended that the “Government is likely to show that both the NLRB and MSPB exercise considerable executive power.” 12 Trump v. Wilcox, 145 S. Ct. 1415, 1415 (2025).

We cannot hope to resolve all the issues raised by these cases in this Essay. Instead, our goal is more modest. Many of the cases addressing this topic have relied on the practices of the executive and legislative branches prior to the Supreme Court’s decisions in Myers and Humphrey’s Executor. 13 For recent examples of judges relying on portions of the record discussed in Part I, see Harris v. Bessent, Nos. 25-5037, 25-5057, 2025 WL 980278, at *2-7 (D.C. Cir. Mar. 28, 2025) (Walker, J., concurring); id. at *36-39 (Millett, J., dissenting); and Seila Law, 140 S. Ct. at 2226-33 (Kagan, J., concurring in the judgment with respect to severability and dissenting in part). In recent years, the scholarly literature addressing these practices has been voluminous, and we are occasionally on different sides of the debate. Yet, as of this moment, that literature lacks a short and simple timeline cohesively addressing the arc of the practices, which might be usable by scholar, student, and practitioner alike. In this adversarial collaboration, we hope to provide a brief narrative of that arc—and then to set forth the arguments that can be made based on that arc. 14 In distilling the narrative and setting forth our arguments in a truncated format, we are no doubt leaving important issues on the cutting room floor. For instance, we say nothing about the topic of qui tam litigation and Article II. See, e.g., Nitisha Baronia, Jared Lucky & Diego Zambrano, Private Enforcement at the Founding and Article II, 114 Cal. L. Rev. (forthcoming 2026) (on file with Authors); United States ex rel. Zafirov v. Fla. Med. Assocs., LLC, 751 F. Supp. 3d 1293 (M.D. Fla. 2024). Nevertheless, we hope that the brevity of this contribution provides benefits in other ways—namely, in providing a short introduction and overview of the major fault lines of the debate for those who would benefit from such a treatment. Part I sets forth the timeline from the First Congress through the Jeffersonian and Jacksonian dismissals, the Johnson impeachment, the incipient formation of the federal civil service, and the creation of the Interstate Commerce Commission and the first independent agencies. To allow readers to form their own assessments, Part I describes these issues in as non-argumentative of a fashion as is possible. Part II and Part III then turn to legal argumentation, first by sketching out an argument by one of us (Bamzai) in favor of a robust removal power, then by turning to an argument by the other (Shane) for a limited removal power.

I. Removal: A Short and Sweet Timeline

Except in the clauses authorizing the impeachment and conviction (and removal) of federal officers, 15 See U.S. Const. art. I, § 3, cl. 7 (“Judgment in Cases of Impeachment shall not extend further than to removal from Office”); id. art. II, § 4 (“The President, Vice President and all civil Officers of the United States, shall be removed from Office on Impeachment for, and Conviction of, Treason, Bribery, or other high Crimes and Misdemeanors.”). the constitutional text does not speak directly to the manner of removing executive branch officers. Disputes have thus focused on the inferences that might be drawn from more general constitutional grants of authority to the President and to Congress. On the one hand, Article II of the Constitution vests the “executive Power” in the President, 16 Id. art. II, § 1. requires that he “take Care that the Laws be faithfully executed,” 17 Id. § 3. and authorizes him to appoint all principal officers with the Senate’s advice and consent. 18 Id. § 2, cl. 2. On the other hand, Article I gives Congress the power “[t]o make all Laws which shall be necessary and proper for carrying into Execution . . . all other Powers vested by this Constitution in the Government of the United States, or in any Department . . . thereof” 19 Id. art. I, § 8, cl. 18. —which presumably includes at least some authority to organize and control the bureaucracy.

Against this backdrop, the scope of presidential removal power has long been a subject of constitutional debate—for far longer, as we explain below, than 1926, when the Supreme Court weighed in on the issue in a significant manner in Myers v. United States. 20 This Part borrows from, but makes significant additions to, the historical summary contained in Jerry L. Mashaw, Peter M. Shane, Aditya Bamzai, Emily S. Bremer, Margaret B. Kwoka & Nicholas R. Parrillo, Administrative Law: The American Public Law System 178-81 (2025).

A. The First and Second Congresses

1. The Creation of the Departments

The question of where, if at all, the Constitution lodges a removal power emerged in legislative debates during the First Congress. Debating a bill to establish the Department of Foreign Affairs, different members of Congress expressed different positions on the meaning of the Constitution. Among the positions taken were (1) the Constitution granted the President plenary removal power, (2) Presidents could remove officers only to the extent authorized by Congress, or (3) Presidents could remove officers—at least Senate-confirmed officers—only with the Senate’s consent to removal. 21 See Aditya Bamzai & Saikrishna Bangalore Prakash, The Executive Power of Removal, 136 Harv. L. Rev. 1756, 1774 (2023) (noting, in addition, that some theorized that officers could be removed only by impeachment and conviction). During the debates over the Constitution’s ratification, Alexander Hamilton had advocated the last of these positions in The Federalist No. 77, 22 The Federalist No. 77, at 459 (Alexander Hamilton) (Clinton Rossiter ed., 1961). but he appeared to change his opinion in favor of plenary removal power at some point thereafter. 23 Letter from William Smith (S.C.) to Edward Rutledge (June 21, 1789), reprinted in 16 The Documentary History of the First Federal Congress of the United States of America: Correspondence 832-33 (Charlene Bangs Bickford, Kenneth R. Bowling, Helen E. Veit & William Charles DiGiacomantonio eds., 2017) (representing that Hamilton informed a representative that he “was now convinced that the [President] alone [should] have the power of removal at pleasure”); see also Alexander Hamilton, Pacificus No. 1 (June 29, 1793), reprinted in 15 The Papers of Alexander Hamilton, June 1793January 1794, at 39-40 (Harold C. Syrett ed., 1969) (declaring that with certain “exceptions [(the appointment, treaty, and war powers)] the Executive Power of the [United States] is completely lodged in the President;” that “[t]his mode of construing the Constitution has indeed been recognized by Congress in formal acts, upon full consideration and debate;” and that “[t]he power of removal from office is an important instance”).

In what came to be known as “the Decision of 1789,” 24 For an early (possibly the first) use of the term “the Decision of 1789” to describe the First Congress’s actions, see Senator Daniel Webster’s remarks, 10 Reg. Deb. 382 (1834). the House of Representatives struck language in the statute creating the Department of Foreign Affairs that would have made the Secretary explicitly “removable from office by the president of the United States.” 25 The Congressional Register (June 22, 1789), reprinted in 11 The Documentary History of the First Federal Congress of the United States of America: Debates in the House of Representatives 1028 (Charlene Bangs Bickford, Kenneth R. Bowling & Helen E. Veit eds., 1992); Gazette of the United States, 24 June 1789, reprinted in id. at 1036. The House substituted statutory language that presupposed that the President could remove heads of departments, although the text did not specify the source of that power or address the possibility of Senate involvement. 26 Id. at 1026-27; The Congressional Register (June 22, 1789), reprinted in 11 Documentary History, supra note 25, at 1028; Monday, June 22, 1789, reprinted in 3 The Documentary History of the First Federal Congress of the United States of America: House of Representatives Journal 91-92 (Linda Grant De Pauw, Charlene Bangs Bickford & LaVonne Siegel Hauptman eds., 1977).

Scholars and jurists draw different inferences from the Decision of 1789. Some, such as Chief Justice Taft in Myers, regard the statutory change and legislative debate as evincing Congress’s view that the Constitution gives the President a robust removal power. Others regard the debate as ambiguous. 27 See Bamzai & Prakash, The Executive Power of Removal, supra note 21, at 1773-77, 1793-1802; Jed H. Shugerman, The Indecisions of 1789: Inconstant Originalism and Strategic Ambiguity, 171 U. Penn. L. Rev. 753, 760 (2023). One aspect of the question is undeniable—it was a close debate, with a tie vote in the Senate that had to be broken by Vice President John Adams, who adhered to the perspective that the executive power Vesting Clause conferred removal authority on the President. 28 See Foreign Affairs Bill, H.R. 8 (1789), reprinted in 4 The Documentary History of the First Federal Congress of the United States of America: Legislative Histories 696, 697 n.4 (Charlene Bangs Bickford & Helen E. Veit eds., 1986).

2. The Mint and Sinking Funds

In two statutes enacted in 1790 and 1792, Congress established boards to oversee the U.S. Mint and the “sinking fund.” By statute, the board overseeing the U.S. Mint included the Chief Justice and four executive officers subject to presidential removal. 29 See Act of Apr. 2, 1792, ch. 16, § 18, 1 Stat. 246, 250 (providing by statute that the board would include the Chief Justice, the Secretary of State, the Secretary and the Comptroller of the Treasury, and the Attorney General). The board met once a year to assay coins. 30 See id. The Sinking Fund Commission was composed of the Chief Justice, the Vice President, and three executive officers. 31 Act of Aug. 12, 1790, ch. 47, § 2, 1 Stat. 186, 186 (providing by statute that the board would include the Chief Justice, the Vice President, the Secretaries of State and the Treasury, and the Attorney General). It had the authority to purchase U.S. securities on the open market in the wake of the Revolutionary War. 32 See id.

The inclusion of the Chief Justice (and with one of the boards, the Vice President, who was at the time elected separately from the President) is notable. These two officials could not be removed by the President—at least from their posts as Chief Justice and Vice President. Accordingly, some have argued that the statutory schemes for these boards conferred a measure of policy independence from the President, establishing early predecessors to the independent agencies that Congress later created. 33 For the Supreme Court’s treatment, see Collins v. Yellen, 141 S. Ct 1761, 1785 n.19 (2021), which contends that three members of the Sinking Fund Commission “were part of the President’s Cabinet and therefore removable at will.” For dueling perspectives on this issue, see Christine Kexel Chabot, Is the Federal Reserve Constitutional? An Originalist Argument for Independent Agencies, 96 Notre Dame L. Rev. 1, 4 (2020); and Aditya Bamzai, Tenure of Office and the Treasury: The Constitution and Control over National Financial Policy, 1787 to 1867, 87 Geo. Wash. L. Rev. 1299, 1334-40 (2019).

3. The First and Second Banks

In 1791, Congress enacted a statute to charter the First Bank of the United States, whose bills by statute would be “receivable in all payments to the United States.” 34 Act of Feb. 25, 1791, ch. 10, § 10, 1 Stat. 191, 196. Congress established that shareholders of the First Bank would elect its twenty-five directors, who would in turn choose the Bank President. 35 Id. §§ 4, 7, 1 Stat. at 192-93. The United States was statutorily barred from subscribing to more than one-fifth of the Bank stock, which meant that private shareholders could control the Bank’s operations. 36 See id. §§ 9, 11, 1 Stat. at 196. To be sure, the statute authorized the Department of the Treasury to supervise the Bank by demanding reports and inspecting records, which ensured that the federal government exercised a measure of control over the Bank’s operations. Id. § 7, art. XVI, 1 Stat. at 195.

Twenty years later, Congress decided not to renew the First Bank’s charter. 37 Max M. Edling, A Hercules in the Cradle: War, Money, and the American State, 1783-1867, at 121 (2014). But following financial difficulties in the wake of the War of 1812, Congress chartered a Second Bank of the United States in 1816. 38 See id. at 142; Bamzai, Tenure of Office and the Treasury, supra note 33, at 1343. In the new charter, Congress authorized the President to appoint five of the Bank’s twenty-five directors with the Senate’s advice and consent. 39 Act of Apr. 10, 1816, ch. 44, § 8, 3 Stat. 266, 269.

The President’s lack of control over all the shareholders of the First and Second Banks meant that these entities operated with a degree of independence from presidential directives. For that reason, some commentators—such as Justice Kagan—have described the Second Bank as “the first truly independent agency in the republic’s history.” 40 Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2231 (2020) (Kagan, J., concurring in the judgment with respect to severability and dissenting in part) (quoting Lawrence Lessig & Cass R. Sunstein, The President and the Administration, 94 Colum. L. Rev. 1, 30 (1994)). Others view the First and Second Banks as private institutions performing non-sovereign functions. 41 See, e.g., Bamzai, Tenure of Office and the Treasury, supra note 33, at 1346-57. Still others have taken the position that the Second Bank was a private corporation unlawfully exercising the sovereign functions of the federal government—a view famously articulated by President Andrew Jackson, whose veto of the recharter of the Second Bank led to its dissolution. 42 See id. at 1355-57; Veto Message from President Jackson Regarding the Bank of the United States (July 10, 1832), reprinted in 3 A Compilation of the Messages and Papers of the Presidents 1139, 1149 (James D. Richardson ed., 1897). Jackson’s statement did not profess any objection to the bank based on its relative insulation from presidential direction, even though it acknowledged that “[t]he bank is professedly established as an agent of the executive branch of the Government.” Id. at 1152. Some thus view Jackson’s veto message as implicit acknowledgement of the lawfulness of insulation of the Bank’s functions from presidential direction; others argue that Jackson proceeded from the premise that the Bank was a private entity to which Congress could not delegate sovereign functions. See sources cited supra notes 40-41.

B. The Jefferson Administration

The question of removal arose with ferocity during the first party turnover in the presidency, the transition from the administration of Federalist John Adams to Democratic-Republican Thomas Jefferson. At this time, Jefferson removed many Federalist holdovers in the executive administration. 43 Carl E. Prince, The Passing of the Aristocracy: Jefferson’s Removal of the Federalists, 1801-1805, 57 J. Am. Hist. 563, 565 (1970) (“Removals in one form or another for purely political reasons constituted the backbone of [Jefferson’s] effort to break the Federalists’ power.”).

Most famously, Jefferson and his Secretary of State James Madison elected not to deliver the commissions of certain justices of the peace of the District of Columbia, thereby preventing some Federalists (including a certain William Marbury) from taking office. In Marbury v. Madison, 44 5 U.S. (1 Cranch) 137 (1803). Chief Justice John Marshall addressed whether Madison could lawfully withhold delivery of Marbury’s commission. In the course of his analysis, Marshall wrote that because “the law creating the office [of justice of the peace], gave the officer a right to hold for five years, independent of the executive, the appointment was not revocable.” 45 Id. at 162. That dictum implied that Jefferson lacked the power formally to remove Marbury, but did not explain why such a limit on removal would be constitutional. 46 Because Chief Justice Marshall ultimately decided the Court lacked jurisdiction to address Marbury’s complaint, id. at 173-76, he rendered his assertion as to Marbury’s tenure technically dictum. Years later, the Court addressed and arguably cabined the meaning of this dictum from Marbury, 47 See Parsons v. United States, 167 U.S. 324, 335-36 (1897) (observing that Marbury concerned an appointee seeking “to retain possession of an office created by Congress in and for the District of Columbia” and its dictum was “not necessarily applicable to the case of an officer appointed to an office outside of such District”). but it has continued to play a significant role in debates over the removal power. 48 See, e.g., Humphrey’s Executor v. United States, 295 U.S. 602, 631 (1935). For dueling perspectives on Marbury’s dictum, see Bamzai & Prakash, Executive Power of Removal, supra note 21, at 1802-18; and Jane Manners & Lev Menand, The Three Permissions: Presidential Removal and the Statutory Limits of Agency Independence, 121 Colum. L. Rev. 1, 25-26 (2021).

C. Jackson and the National Bank

The question of removal arose again, with equal ferocity, when Andrew Jackson dismissed the Secretary of the Treasury for failing to initiate a funds transfer to weaken the power of the Second Bank of the United States. After Andrew Jackson vetoed a bill to recharter the Second Bank in 1832, the President and the Bank found themselves at electoral loggerheads during Jackson’s reelection campaign. 49 Robert V. Remini, Andrew Jackson and the Bank War: A Study in the Growth of Presidential Power 89-108 (1967). After winning the 1832 election, Jackson decided to weaken the Bank even before the expiration of its charter by removing federal deposits and transferring federal funds to state banks. 50 See id. at 109-53. But Jackson’s Secretary of the Treasury, William Duane, refused to remove the deposits. Jackson fired him. 51 See id. at 122-24.

Jackson’s removal of Duane set off a series of legislative debates over whether Jackson had the authority to remove the Secretary of the Treasury for failing to remove deposits from the Bank. 52 See id. at 137-53. The Bank’s statute vested the decision over transfers in the Secretary, and Duane interpreted the Act as authorizing the removal of deposits only if the Secretary determined, in his independent judgment, that deposits were unsafe in the national bank. The removal controversy ultimately elicited House resolutions supporting the President, but also a still-unique Senate resolution censuring the president and challenging his removal power. Nonetheless, near the end of his second term, Jackson’s Senate allies successfully managed a vote that expunged the censure, effectively vindicating his position. 53 Harold H. Bruff, Untrodden Ground: How Presidents Interpret the Constitution 95-104 (2015).

D. The Impeachment of Andrew Johnson

The question of removal arose with arguably even more ferocity when Congress impeached and almost removed President Andrew Johnson for seeking to replace the Secretary of War, Edwin Stanton, during Reconstruction. Then-Vice President Johnson had taken office following the assassination of Abraham Lincoln. 54 William H. Rehnquist, Grand Inquests: The Historic Impeachments of Justice Samuel Chase and President Andrew Johnson 143 (1992). Johnson’s taking of office precipitated a serious confrontation between the elected branches over Reconstruction. 55 See id. at 143. The Republican-dominated Congress was determined not to recognize the governments of secessionist states except under conditions stricter than even Lincoln had recommended. But Andrew Johnson sought a version of Reconstruction only partial and far more permissive.

Concerned that Johnson would change the composition of the Lincoln Administration to favor Southern interests, Congress enacted the Tenure of Office Act of 1867 over Johnson’s veto. 56 See Act of Mar. 2, 1867, ch. 154, 14 Stat. 430, 432 (repealed 1887). The Act provided that any civil officer appointed with the Senate’s advice and consent would serve until a successor was appointed by the President, also with the Senate’s advice and consent. It went on to specify that the Secretaries of State, of the Treasury, of War, of the Navy, and of the Interior, as well as the Postmaster General and the Attorney General would hold their respective offices “for and during the term of the President by whom they may have been appointed, and for one month thereafter, subject to removal by and with the advice and consent of the Senate.” 57 Id. § 1, 14 Stat. at 430. Lest there be any doubt as to Congress’s intention to hamstring Johnson specifically, the Act’s final paragraph provided that “every person who shall violate any of the provisions of this section shall be deemed guilty of a high misdemeanor.” 58 Id. § 9, 14 Stat. at 432.

With his control over the executive branch threatened, Johnson removed the Secretary of War, Edwin Stanton, from office in February 1868. 59 Rehnquist, supra note 54, at 212-15. The House impeached Johnson three days later, in part for violating the Tenure of Office Act. 60 See id. at 217. The Senate proceedings included extensive debates over the scope of the President’s constitutional removal authority and the lawfulness of the Tenure of Office Act. 61 See, e.g., Bamzai, Tenure of Office and the Treasury, supra note 33, at 1380-83.

Johnson escaped conviction by a single senator’s vote, and the episode established a controversial precedent in the history of the President’s removal power. 62 See Cong. Globe Supp., 40th Cong., 2d Sess. 36 (May 16, 1868); id. at 412-15 (May 26, 1868). For an accounting of the impeachment and trial, see David P. Currie, The Reconstruction Congress, 75 U. Chi. L. Rev. 383, 437-52 (2008). When Chief Justice Taft wrote Myers, he remarked that in later years “the injury and invalidity of the Tenure of Office Act in its radical innovation were immediately recognized by the executive and objected to” and that the Act “in so far as it attempted to prevent the President from removing executive officers who had been appointed by him by and with the advice and consent of the Senate, was invalid.” 63 Myers v. United States, 272 U.S. 52, 167, 176 (1926). For Justice Brandeis’s contrary perspective, see id. at 253-59 (Brandeis, J., dissenting).

The 1867 Act remained on the books until its repeal in 1887. 64 Act of Mar. 3, 1887, ch. 353, 24 Stat. 500. That development hardly terminated, however, the ongoing political debate about the scope of presidential removal power.

E. The Civil Service and Independent Agencies

At approximately the same time as the repeal of the Tenure of Office Act, Congress took the initial steps to create what became the civil service system, as well as to create the first of what became known as the “independent” agencies.

1. The Civil Service

Early Presidents had taken different approaches to removing prior administrations’ lower-level executive branch appointees. Some removed few, effectively leading to an expectation of a lifetime appointment among members of the federal bureaucratic service. Others—most notably, Andrew Jackson in 1829—removed many prior appointees, in a purge that some saw as the introduction of a needed democratic reform and others saw as the introduction of the “spoils system.” 65 For a classic treatment, see Carl Russell Fish, The Civil Service and the Patronage 21821 (1905).

In 1883, Congress enacted the Pendleton (or Civil Service) Act, 66 Act of Jan. 16, 1883, ch. 27, 22 Stat. 403. establishing for the first time a competitive civil service. The Pendleton Act required individuals in the “classified” service to be selected through a merit-based, rather than patronage, system. The statute limited removal solely in a single section, which barred the discharge of any officer or employee “for giving or withholding or neglecting to make any contribution of money or other valuable thing for any political purpose.” 67 Id. § 13, 22 Stat. at 407. Almost in parallel, in 1886 the Supreme Court held that Congress could impose limits on removal when it vested authority to appoint an inferior officer in the head of a department. 68 See United States v. Perkins, 116 U.S. 483, 485 (1886).

Approximately a decade later, invoking authority delegated to the President by the Pendleton Act, President William McKinley enhanced the removal protections for those in the civil service. He did so by amending the civil service rules to provide that removals would not be made from civil service positions “except for just cause and upon written charges filed with the head of the Department, or other appointing officer, and of which the accused shall have full notice and an opportunity to make defense.” 69 Exec. Order No. 101 (July 27, 1897). But McKinley’s action—taken by executive order alone—did not change the underlying statutory framework. In 1912, Congress enacted a similar removal protection into statute, 70 Lloyd-La Follette Act of 1912, ch. 389, § 6, 37 Stat. 540, 555. leading to the creation of statutory removal provisions that apply to those lower-level executive branch officials described as serving within the civil service.

2. Independent Agencies

In 1887, Congress created what is generally considered the first modern independent regulatory commission, the Interstate Commerce Commission (ICC). 71 Act of Feb. 4, 1887, ch. 104, § 11, 24 Stat. 379, 383. The statute creating the ICC provided that the President could remove the five commission members for “inefficiency, neglect of duty, or malfeasance in office.” 72 Id. Shortly after it established the ICC, Congress created a second board subject to a similar limit on the President’s removal authority, the Board of General Appraisers. 73 Customs Administrative Act of 1890, ch. 407, § 12, 26 Stat. 131, 136; see generally Aditya Bamzai, Taft, Frankfurter, and the First Presidential For-Cause Removal, 52 U. Rich. L. Rev. 691, 712-19 (2018) (discussing the creation of the Board of General Appraisers).

The expansion in the number of agencies whose heads are subject to such limits on presidential removal began decades later during President Woodrow Wilson’s Administration. For example, in the early days of Wilson’s presidency Congress created the Federal Reserve Board, 74 Federal Reserve Act, ch. 6, 38 Stat. 251 (1913). The status of the Board of Governors of the Federal Reserve, and whether the Federal Reserve’s structure warrants a distinct constitutional analysis, is a matter of significant debate. Compare Trump v. Wilcox, 145 S. Ct. 1415, 1415 (2025) (contending that the Court’s treatment of the removal provisions regarding the NLRB and MSPB does not “necessarily implicate the constitutionality of for-cause removal protections for members of the Federal Reserve’s Board of Governors or other members of the Federal Open Market Committee,” because “[t]he 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”), with id. at 1421 (Kagan, J., dissenting) (describing the majority’s reasoning on the Federal Reserve as coming “out of the blue” because “the Federal Reserve’s independence rests on the same constitutional and analytic foundations as” other independent agencies, “which is to say it rests largely on Humphrey’s”). and the Federal Trade Commission. 75 Act of Sept. 25, 1914, ch. 311, 38 Stat. 717. Later years saw the creation of a variety of additional agencies with comparable limitations on removal. 76 For a canonical study, see Robert E. Cushman, The Independent Regulatory Commissions (1941).

* * *

We have sketched the political debates that occurred over removal from the founding-era through Reconstruction to the foundations of the civil service and the independent agencies. These historical developments have provided fodder for the legal debates that have ensued.

II. In Defense of Robust Presidential Removal Authority (Bamzai)

Everyone can agree that the federal Constitution creates a “unitary executive” in one sense. At the constitutional convention, the question arose whether to give the “executive power” to a single individual who would fill the office of the President or to have several executives or an executive council. 77 1 The Records of the Federal Convention of 1787, at 21 (Max Farrand ed., 1911) (Virginia Plan); id. at 244 (New Jersey Plan). See, e.g., id. at 65-66 (remarks of James Wilson) (arguing that “a single magistrate” would “giv[e] most energy despatch and responsibility to the office” and that “Unity in the Executive . . . would be the best safeguard against tyranny”); id. at 66 (remarks of Edmund Randolph) (opposing “a unity in the Executive magistracy” and questioning why “the great requisites for the Executive department, vigor, despatch & responsibility could not be found in three men, as well as in one man”). The latter notion was by no means outlandish, because many states had (and currently have) plural executives, 78 In other words, by lodging some appointments and removals with state legislatures, many state constitutions effectively vested a portion of the “executive power” in a plural body. For discussion, see, for example, Aditya Bamzai & Saikrishna Prakash, How to Think About the Removal Power, 110 Va. L. Rev. Online 159, 176-84 (2024). See also The Federalist No. 47, at 304-05 (James Madison) (Clinton Rossiter ed., 1961) (noting that the Massachusetts Constitution of 1780 resulted in a “partial mixture of powers,” most notably because “a number of the officers of government are annually appointed by the legislative department” and because “the appointment to offices, particularly executive offices, is in its nature an executive function”). and some even viewed the Congress under the Articles of the Confederation as a plural executive authority. 79 Jack N. Rakove, The Beginnings of National Politics: An Interpretive History of the Continental Congress 175 (1979) (observing that Thomas Burke described the Continental Congress as “a deliberating Executive assembly,” and Thomas Jefferson described it as “both legislative and executive”). The idea of a single executive prevailed at the Convention and made its way into our Constitution.

So at least nominally, in form, there is a single executive atop the executive branch at the federal level. The question, then, becomes whether the Constitution means for there to be a single executive in both substance as well as in form and, if so, what the contours of that substance might be. In our modern jurisprudence, the question of substance has depended on whether the President has a power to remove subordinate executive branch officials under the Constitution.

To take a somewhat fanciful example, imagine that our current Congress were to cleave off a portion of the executive branch—take your pick, but let’s say the State Department—and render its head not answerable to the President, but rather to the Senate Majority Leader. Would that be constitutional? In my view, no. And in this regard, I find myself in good company, because Justice Kagan (in her Seila Law dissent) indicates that she agreed that, in such areas, the President “need[ed] the capacity to fire such advisers for disagreeing with his policy calls.” 80 Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2235 n.9 (2020) (Kagan, J., concurring in the judgment with respect to severability and dissenting in part); see id. at 2232 n.6 (contending that “Congress may not impose removal restrictions preventing the President from carrying out his own constitutionally assigned functions in areas like war or foreign affairs”). That position indicates that there are some implied limits on how Congress might structure executive agencies at the federal level and a basis for a removal authority that Congress cannot take away by statute.

Once one agrees that there are such implied limits, the question turns to locating their source and defining their contours. Consider, first, the question of basis in the constitutional text. The first clause of Article II vests “[t]he executive Power . . . in a President of the United States.” Article II then goes on to confer a number of specific powers and impose duties on the President, including the authority to appoint officers with the advice and consent of the Senate, the obligation to “take care that the laws be faithfully executed,” and the authority to obtain the opinions of principal officers. The most plausible interpretation of these provisions is that the clause vesting “executive power” confers some authority to manage the bureaucracy that Congress cannot divest—specifically a removal authority—in addition to the authority provided by the clauses that follow. If it does confer such authority, then my position (and incidentally, Justice Kagan’s on foreign affairs advisers) is readily explainable. 81 Perhaps Justice Kagan would derive her limited removal authority from other sources in the Constitution, such as the Commander-in-Chief Clause or some sort of presidential foreign-affairs power. The difficulty with such an argument is that the Commander-in-Chief Clause was not understood to confer such authority; rather, the Executive Vesting Clause was the ground for the historical debate on such questions. See Bamzai & Prakash, supra note 21, at 1824-30.

One might wonder why the Framers listed specific executive functions, such as the Appointments Clause and the Opinions Clause, if the Vesting Clause itself implied a removal authority. But there are several potential reasons for the structure of Article II. First, some clauses, like the Appointments Clause, can be understood to carve out a portion of the “executive power” (i.e., appointments) and vest it in the Senate (i.e., the advice-and-consent function). Second, some clauses, like the Opinions Clause, may be (as Hamilton put it) “a mere redundancy in the plan, as the right for which it provides would result of itself from the office.” 82 The Federalist No. 74, at 447 (Alexander Hamilton) (Clinton Rossiter ed., 1961). I therefore do not believe that the inclusion of such a minor authority as the Opinions Clause in Article II negates implying a robust removal power from the Vesting Clause. Having said that, I find myself in agreement with Professor Shane’s additional point that the President is not “to perform [the] duties” that Congress allocates to executive branch subordinates “personally,” but rather through the agencies that Congress has authorized. See infra text accompanying note 104. But to my mind, this premise does not necessarily cut against a robust removal power. The President must act through his duly appointed subordinates, whom he has a power to remove.

Turn, next, to the contours of the principle. As explained in Subpart I.A above, the very first debate that Congress engaged in over the meaning of the Constitution was about who controlled the executive branch and how. During that debate, then-Representative James Madison asked and answered his own rhetorical question on the matter: “Is the power of displacing an executive power? I conceive that if any power whatsoever is in its nature executive it is the power of appointing, overseeing, and controlling those who execute the laws.” 83 James Madison, Removal Power of the President, 16 June, 1789, reprinted in 12 The Papers of James Madison: Congressional Series 228 (Charles F. Hobson & Robert A. Rutland eds., 1979); see also The Daily Advertiser (18 June, 1789), reprinted in 11 Documentary History, supra note 25, at 846. And he declared that he could not “subscribe” to the argument that “the power of displacing from office is subject to legislative discretion” because Congress had the authority to “limit or modify as it thinks proper.” 84 The Congressional Register (June `7, 1789), reprinted in 11 Documentary History, at 921-22. These twin propositions—the first about the meaning of the executive power Vesting Clause, and the second about the implications from the Necessary and Proper Clause—formed the dominant understanding of the First Congress’s handiwork.

Consider two further aspects of that debate. First, although the debate concerned the Secretary of Foreign Affairs, Congress used similar language regarding removal in the statute creating the Department of the Treasury. 85 Act of Sept. 2, 1789, ch. 12, § 7, 1 Stat. 65, 67 (providing that the Treasury “Assistant” would have custody of departmental papers whenever the President removed the Secretary of the Treasury). That is much stronger evidence that the First Congress considered the same basic approach to removal to be applicable to the Secretary of the Treasury than the fact that they might have also legislated with greater specificity regarding the Treasury Secretary’s functions. No less importantly, it is difficult to differentiate modern executive departments based on their supposed foreign-affairs or national-security functions. After all, the Secretary of the Treasury oversees the Undersecretary for Terrorism and Financial Intelligence and the Office of Foreign Assets Control, among others. The salient point is that all these entities exercise the “executive power.”

Second, the statutes enacted by the First Congress necessarily rejected Alexander Hamilton’s famous suggestion in The Federalist No. 77, on which Professor Shane relies, that the Senate would be necessary to remove an officer, just as it was to appoint one. 86 See infra text accompanying notes 106-108. I take Professor Shane’s point but have several reactions. First, Hamilton (equally famously) changed his mind on this question, as Professor Shane acknowledges. 87 See supra note 23 and accompanying text (addressing Hamilton’s change of heart). Second, the implication from The Federalist No. 77 is not that Congress has the authority to lodge removal authority where it sees fit, but rather that the Constitution lodged removal authority jointly in the President and Senate. That position, as I see it, is conceptually closer to the view that the Constitution lodges an executive power of removal (albeit jointly, in the President and Senate) than that the Constitution is silent on the matter. Third, I do not understand Professor Shane or Justice Kagan to embrace Hamilton’s position, for it suggests that Senate consent would be required even to remove officers such as the Secretaries of State and Defense.

In my view, the dominant perspective was that the Decision of 1789 (and the better interpretation of the Constitution) acknowledged a presidential authority to remove that Congress cannot take away by statute. Several prominent commentators expressed this perspective. Consider the views of Chief Justice Marshall in his book on The Life of George Washington. He described the statutory changes made by the First Congress as intended “[t]o obviate any misunderstanding of the principle on which the question [of removal] had been decided.” 88 5 John Marshall, The Life of George Washington 199-200 (Philadelphia, C.P. Wayne 1807). In Chief Justice Marshall’s view, the amendment was made because the earlier statutory language might have suggested that “the power of removal by the president might hereafter appear to be exercised by virtue of a legislative grant only, and consequently be subjected to legislative instability.” 89 Id. at 200. But the matter was “fixed in the [C]onstitution.” 90 Id. at 199-200 (ascribing this view to Representative Egbert Benson, who sought the statutory amendment). That construction, according to Chief Justice Marshall, had “ever been considered as a full expression of the sense of the legislature on this important part of the American [C]onstitution.” 91 Id. at 200.

Chief Justice Marshall’s writing both reflects an important early commentary on the meaning of the Decision of 1789 and also bears on the meaning of his dictum in Marbury v. Madison. Professor Shane notes that Chief Justice Marshall “regarded William Marbury as protected against presidential removal by his statutory term of years.” 92 See infra text accompanying note 113. But even assuming the dictum in Marbury points in that direction, Marbury nowhere explained why that protection against removal was constitutional. Certainly, in his 1807 book, Chief Justice Marshall did not suggest that he disagreed with the constitutional understanding of the Decision of 1789. The most plausible interpretation of Marbury is that Chief Justice Marshall had a narrower principle in mind, perhaps believing that the statute was constitutional because the justices of the peace were judicial officers or territorial officers. 93 See Bamzai & Prakash, Executive Power of Removal, supra note 21, at 1803-05, 1810-13. I have set to the side the fact that President Jefferson viewed the justices of the peace as removable executive officers. See id. at 1806-10.

Or consider the views of Justice Joseph Story. In his Commentaries on the Constitution, Justice Story observed that in the Decision of 1789, “the final vote seems to have expressed the sense of the legislature, that the power of removal by the executive could not be abridged by the legislature.” 94 Joseph Story, II Commentaries on the Constitution of the United States § 1537 n.3, at 350 (Thomas M. Cooley ed., Boston, Little, Brown & Co. 1873) (1833); see also id. § 1542, at 354 (“After a most animated discussion, the vote finally taken in the house of representatives was affirmative of the power of removal in the President, without any co-operation of the senate.”).

Finally, consider the views of Chancellor James Kent. He mentioned that “the construction which was given to the constitution by congress . . . contemplated” that a department secretary was “removable from office by the president.” 95 1 James Kent, Commentaries on American Law 289 (New York, O. Halsted 1826). Chancellor Kent reasoned that “[t]his amounted to a legislative construction of the constitution, and it has ever since been acquiesced in and acted upon, as of decisive authority in the case.” 96 Id. (reasoning further that “[i]t applies equally to every other officer of government appointed by the president and senate, whose term of duration is not specially declared”).

To be sure, I do not mean to suggest that all major commentators had the same view on this issue; indeed, Justice Story himself appeared to be of two minds. 97 I read Justice Story’s Commentaries somewhat differently from Professor Shane. In sections 1539 and 1540, on which Professor Shane relies, Justice Story spells out arguments against a presidential removal power. See Story, supra note 94, §§ 1539, 1540, at 351-52. Justice Story then proceeds in section 1541 to lay out the argument “[o]n the other hand,” i.e., in favor of a unilateral presidential removal power. Id. § 1541, at 353. Finally, after declaring the Decision of 1789 settled, Justice Story proclaims that “[i]f there has been any aberration from the true constitutional exposition of the power of removal, (which the reader must decide for himself), it will be difficult, and perhaps impracticable, after forty years’ experience, to recall the practice to the correct theory.” Id. § 1544, at 356. In this manner, I understand Justice Story to be accepting the settlement of the Decision of 1789 and, at least in this text, leaving it to his “reader” to decide whether the First Congress was correct. In a sense, Justice Story was engaged in an “adversarial collaboration” with himself. But despite occasional controversy (especially during the presidency of Andrew Jackson), these views reflected the best interpretation of the text and early debate culminating in the Decision of 1789, as well as the prevailing practice until the Civil War era. Although Congress’s Civil-War-era enactment of the Tenure of Office Act departed from the Decision of 1789, as Justice Kagan has noted, that statute was a “historical ‘aberration’” that was “ultimate[ly] repudiate[ed]” in light of a “broad historical consensus that it went too far.” 98 Seila Law LLC v. CFPB, 140 S. Ct. 2183, 2232 n.6 (2020) (Kagan, J., concurring in the judgment with respect to severability and dissenting in part). For this reason, the various good-cause removal restrictions enacted after the Civil War to create independent agencies must be assessed against the backdrop of the pre-Civil War approach to the removal power. They cannot be understood to have displaced the traditional perspective on the allocation of such authority. That is the view expressed by Chief Justice Taft in Myers v. United States and Chief Justice Roberts in recent cases.

Let’s step back from the particulars and take stock of the broader jurisprudential debate. First, all can agree that this was a contested question around the time of the Constitution’s adoption. People debated removal vigorously. One might then ask what role early debate and dispute should play in settling the meaning of the Constitution. Indeed, some have suggested that early debate itself precludes giving the executive power Vesting Clause a settled meaning. I am disinclined to take that general jurisprudential view, because I believe certain statutes enacted by early congresses (such as some parts of the Alien and Sedition Acts) were unconstitutional, even though they received majority support in the eighteenth century. 99 See David P. Currie, The Constitution in Congress: The Federalist Period, 1789-1801, at 253-73 (1997). Thus, the mere fact that an issue was disputed does not necessarily mean that the Constitution lacks meaning on a particular question. History, in this fashion, can establish the contours of the debate, but judgment, too, must be used to decide the strength or weakness of a legal position.

Second, on occasion, some participants in the modern debate couple their perspective on these historical debates with a thumb on the scale in favor of less judicial involvement in deciding constitutional questions—what might be described as a Thayerian view of constitutional disputes generally. 100 James Bradley Thayer, The Origin and Scope of the American Doctrine of Constitutional Law, 7 Harv. L. Rev. 129 (1893). I take seriously that jurisprudential perspective, but its application in this instance is debatable. For example, recent and currently pending cases have witnessed injunctions and related remedies issued against the executive branch to keep certain officials in place, which will require constant monitoring and intervention into executive branch practices by any supervising judge. 101 See, e.g., Wilcox v. Trump, No. 25-334, 2025 WL 720914, at *18 (D.D.C. Mar. 6, 2025) (ordering, among other relief, that “plaintiff shall continue to serve as a member of the NLRB until her term expires,” that the defendants are enjoined “from removing plaintiff from her office without cause or in any way treating plaintiff as having been removed from office, from impeding in any way her ability to fulfill her duties as a member of the NLRB, and from denying or obstructing her authority or access to any benefits or resources of her office” and that the defendants are ordered to “provide plaintiff with access to the necessary government facilities and equipment so that she may carry out her duties during her term as a member of the NLRB”). Leaving to one side the important question whether those remedies are appropriate, the efficacy of removal restrictions on the question of who controls the government appears to depend on them. And if so, as we can see, the actual implementation of removal restrictions would require constant supervision of the executive branch by the third branch of the government.

III. In Defense of Congressional Limits on Removal (Shane)

Contrary to unitary executive theory as urged by Professor Bamzai, the Constitution does not vest the President with absolute removal authority. With a few likely exceptions that may be derived by implication, the Constitution empowers Congress to design administrative structures, including removal protections, for the agencies it creates.

The premise of the Roberts Court’s unitary executive theory—that Article II creates a one-person branch of government 102 Trump v. Mazars USA, LLP, 140 S. Ct. 2019, 2034 (2020), quoted in Trump v. United States, 144 S. Ct. 2312, 2329 (2024) (The President is “the only person who alone composes a branch of government”). —is simply untrue. Section 2 explicitly anticipates the existence of “executive departments.” The Opinions Clause authorizes the President to “require the [o]pinion, in writing, of the principal [o]fficer in each of the executive [d]epartments, upon any [s]ubject relating to the [d]uties of their respective [o]ffices.” 103 U.S. Const. art. II, § 2, cl. 1. The “duties” to which the clause refers are duties assigned not by the President, but by Congress. (It would have been absurd to provide that the President could demand from department heads their opinions in writing about the duties the President assigns.) In this respect, agencies are first and foremost the agents not of the President, but of the legislative branch that created them. The relationship intended between the President and the departments is thus one of supervision, not control. He is to “take care that the [l]aws be faithfully executed” by others, not to perform those duties personally. 104 Id. § 3. As Attorney General William Wirt wrote in 1823: “If the laws . . . require a particular officer by name to perform a duty, not only is that officer bound to perform it, but no other officer can perform it without a violation of the law; and were the President to perform it, he would . . . be violating [the law] himself.” The President and Accounting Officers, 1 U.S. Op. Atty. Gen. 624, 625 (1823). Professor Bamzai and I agree on this point. See supra note 82. Properly understood, this supervisory role does not require at-will removal power; Presidents need only the power to remove persons who provide good cause by not faithfully executing the law.

What the Constitution does not do vis-a-vis executive departments is to specify their structure and the conditions of administrative office-holding. With the same clarity that Article II mentions “executive departments,” Article I leaves to Congress all decisions about how best to design the appropriate mechanisms, including administrative agencies, for implementing the law. Under the Necessary and Proper Clause, Congress may “make all [l]aws which shall be necessary and proper for carrying into [e]xecution . . . all other [p]owers vested by this Constitution in the [g]overnment of the United States, or in any [d]epartment or [o]fficer thereof.” 105 U.S. Const. art. I, § 8, cl. 18 (emphasis added). If Congress thinks some form of tenure protection appropriate for officers directing whatever agencies that the legislature creates to implement its Article I authorities, that is thus a judgment for Congress to make. Indeed, the text seems so compelling on this point that, unless a reader is pre-committed to finding a presidential removal power, the text most naturally points in the opposite direction.

Not only does the text not mention a presidential removal power, but the most famous pre-ratification guide for voters on the implications of the proposed Constitution denied that such a power existed. The Federalist, written to persuade voters in favor of ratification, advertised that removal would require Senate consent. Writing as Publius, Alexander Hamilton had described this as an advantage of the system. “A change of the Chief Magistrate,” he promised, “would not occasion so violent or so general a revolution in the officers of the government as might be expected, if he were the sole disposer of offices.” 106 The Federalist No. 77, at 459 (Alexander Hamilton) (Clinton Rossiter ed., 1961). Hamilton’s later change of opinion, mentioned by Professor Bamzai, does not make his pre-ratification assertion any less probative of how ratifiers would have understood what they were voting on. The Senate’s role “would contribute to the stability of the administration.” 107 Id. This was no inconsequential representation. According to Justice Joseph Story, Hamilton’s assurance “had a most material tendency to quiet the just alarms of the overwhelming influence and arbitrary exercise of [the removal] prerogative of the executive, which might prove fatal to the personal independence and freedom of opinion of public officers, as well as to the public liberties of the country.” 108 Story, supra note 94, § 1539, at 351.

The experience of Americans living under executive power as exercised under state constitutions would also have colored what they thought they were voting on. State constitutions adopted most closely in time to the federal Constitution—despite executive power vesting clauses and, in several cases, explicit separation of powers provisions not found in the federal document—frequently gave the legislature control over key administrators. States did not assume that vesting executive power in a single individual necessarily implied complete control over administration. 109 See Peter M. Shane, The Originalist Myth of the Unitary Executive, 19 U. Penn. J. Const. L. 323, 337-44 (2016).

For example, despite the Court’s current view that criminal prosecution is a “quintessentially executive function,” 110 Trump v. United States, 144 S. Ct. 2312, 2334 (2024) (quoting Morrison v. Olson, 487 U.S. 654, 706 (1988) (Scalia, J., dissenting)). there was no such 1780s consensus; prosecutors were considered as much judicial as executive officers. The same appears to be true of attorneys general. Of the constitutions under which the original thirteen states first operated after the Revolution, six mention an attorney general explicitly, and each of them refers to the attorney general in the same clause as it refers to state judges, implying that attorney general is a judicial office. 111 Peter M. Shane, Prosecutors at the Periphery, 94 Chi.-Kent L. Rev. 241, 252 & n.42 (2019). It is similarly revealing that the first federal statute creating the office of attorney general and providing for the first federal district attorneys is the Judiciary Act of 1789, not the organic act of any executive department. Nothing in this history supports an inference that the founding conception of executive power implied the power to remove prosecutors at will.

The historical episodes recounted in Part I of our discussion substantiate the absence of any consensus either in 1789 or during the ensuing century and more that the President enjoys an indefeasible removal power. The machinations regarding the so-called Decision of 1789 show that “the First Congress was undecided on any constitutional theory and retreated to ambiguity in order to compromise and move on to other urgent business.” 112 Shugerman, supra note 27, at 753. John Marshall regarded William Marbury as protected against presidential removal by his statutory term of years. 113 Marbury v. Madison, 5 U.S. (1 Cranch) 137, 162 (1803) (“[A]s the law creating the office, gave the officer a right to hold for five years, independent of the executive, the appointment was not revocable.”). Later Congresses thought themselves entitled via Tenure in Office Acts to require Senate consent for certain removals.

The creation of the first federal civil service system also belies any asserted consensus about presidential removal power. For starters, notwithstanding the Decision of 1789, Congress thought it necessary to state explicitly the President’s power to remove members of the Civil Service Commission. 114 See Act of Jan. 16, 1883, ch. 27, § 13, 22 Stat. 403, 407. Section 13 of the Act prohibited the removal of any civil servant, presumably even by the President, “for giving or withholding or neglecting to make any contribution of money or other valuable thing for any political purpose.” 115 Id. An amendment proposed in the Senate that would have codified a presidential power to remove civil service members without having to “assign cause for such action” was omitted from the final Act. 116 14 Cong. Rec. S282 (Dec. 14, 1882) (statement of Sen. Brown) (“Provided, That nothing in this act shall prevent the removal by the President, or any head of a department, of any officer or employee under him when he thinks the interest of the public service requires it, without being obliged to assign cause for such action.”).

The failure of our early history to cement any binding consensus on presidential removal power is also evident in the most famous of antebellum treatises on constitutional law. Justice Story endorsed Hamilton’s reading in The Federalist No. 77 as the correct understanding of the original text but conceded that forty years of government practice had likely settled the exclusion of the Senate from the process of removing principal officers. 117 Story, supra note 94, § 1538, at 351. He nonetheless urged that it remained within Congress’s power to insist upon Senate consent to the removal of inferior officers 118 Id. § 1544, at 356. —yet further evidence of how little the Decision of 1789 decided.

Professor Bamzai voices two concerns about acknowledging Congress’s power over officer tenure that I do not find worrisome. One is that Congress might exercise its authority in a way that would insulate a pivotal officer like the Secretary of State from presidential control. But recognizing a presumption of congressional authority over tenure need not reach so far. Any law that disables a branch of government from fulfilling its constitutional responsibilities, by definition, cannot be “necessary and proper.” Thus, it would not be “proper” for a Congress generally empowered to design administrative agencies to deny the President complete control over the few officers on whom Presidents have to rely for assistance in implementing executive powers specifically enumerated in Article II.

Something like that idea seems to have animated the First Congress, which recognized that the degree of flexibility to which the President would be entitled in supervising subordinate officers would relate directly to their function. Two of the first three new departments, Foreign Affairs and War, would be headed by officers who would necessarily be central in helping the President fulfill specific Article II roles, such as the negotiation of treaties or direction of the military as commander-in-chief. 119 See Act of July 27, 1789, ch. 4, 1 Stat. 28, 28-29 & n.(a); Act of Aug. 7, 1789, ch. 7, 1 Stat. 49, 49-50 & n.(b). By way of contrast the third new department, Treasury, would administer the new nation’s fiscal and monetary affairs 120 See Act of Sept. 2, 1789, ch. 12, §§ 2, 4, 1 Stat. 65, 65-66. —functions vested explicitly in Congress by Article I. 121 See U.S. Const. art. I, § 8. Congress realized that the President’s role with respect to the principal officers in the Foreign Affairs and War departments would be different. As Jerry Mashaw has written, “[w]hile Congress was content to direct the Secretaries of War and Foreign Affairs to do little more than carry out the instructions provided by the President,” the act establishing the Treasury Department was much more detailed. 122 Jerry L. Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law 40 (2012). Not only did Congress assign to the Secretary numerous specific functions, but it created other presidentially appointed officers within the Department, some of whom were given independent, quasi-judicial authority. 123 Id. Following this logic, the Constitution, properly interpreted, should be read to give the President plenary removal power over the modern-day Secretaries of State and of Defense. The same conclusion is arguable for the Director of National Intelligence and other key national security officials. But officers doing tasks assigned to the executive only by statute could be protected from at-will removal, if Congress so prescribed, without undermining any aspect of Article II.

I also disagree with Professor Bamzai’s concern that rejecting the existence of a presidential at-will removal power would “require constant monitoring and intervention into executive branch practices by any supervising judge.” 124 See supra text accompanying note 101. Rejecting a removal power beyond Congress’s control need not imply some more general judicially enforceable standard for how much presidential influence over a subordinate administrator would overstep the bounds of “faithful execution.” A rule of the latter kind, even if attractive in principle, would be very hard to administer. Distinguishing between officers who do or who do not assist in discharging an explicit Article II power, however, is straightforward.

But beyond the particulars of text and history, my argument against judicial conferral of a sweeping presidential removal power is also based on the proposition that exclusive reliance on a formalist approach to our founding history is unjustified as a normative matter. Even had the founding generation ratified a consensus understanding in favor of presidential removal power, it is doubtful that such a specific understanding should govern nearly two-and-a-half centuries later. 125 Peter M. Shane, Democracy’s Chief Executive: Interpreting the Constitution and Defining the Future of the Presidency 129 (2022). George Washington presided over a small, fledgling nation and a federal government roughly the size of a small cabinet department today. Donald Trump presides over the world’s most powerful nation and a civil service roughly the size of the entire United States in 1776. The risks attending unilateral presidential control over twenty-first century government dwarf the potential dangers of handing too much authority to George Washington.

Indeed, to be faithful in 2025 to America’s founding generation, it is most sensible to embrace most fundamentally their fear of arbitrary power and their pragmatic eclecticism in institutional design. 126 See Jerry L. Mashaw, Creating the Administrative Constitution: The Lost One Hundred Years of American Administrative Law 30-52 (2012) (detailing the pragmaticism and eclectic institutional design choices adopted by the First Congress). Justice Story foresaw the connection between presidential removal power and potential tyranny that is no less relevant today:

[I]t is utterly impossible not to feel, that, if this unlimited power of removal does exist, it may be made, in the hands of a bold and designing man, of high ambition and feeble principles, an instrument of the worst oppression and most vindictive vengeance . . . . It would convert all the officers of the country into the mere tools and creatures of the President. A dependence so servile on one individual would deter men of high and honorable minds from engaging in the public service. And if, contrary to expectation, such men should be brought into office, they would be reduced to the necessity of sacrificing every principle of independence to the will of the chief magistrate, or of exposing themselves to the disgrace of being removed from office . . . . 127 Story, supra note 94, § 1539, at 351-52. I leave it to the reader to judge whether Justice Story was alarmist or prescient.

In an earlier decade, I wrote that embracing unitary executive theory—including presidential removal power—might facilitate an organizational psychology of presidential entitlement within the executive branch that could prioritize loyalty to the President over fidelity to law. 128 Peter M. Shane, Madison’s Nightmare: How Executive Power Threatens American Democracy 142 (2009). I do not believe that possibility is merely hypothetical.

Conclusion

Having set forth competing arguments, we, of course, lack the authority to finally resolve the controversy eventually to be addressed by the Supreme Court. As we have noted, several lower courts have recently held that principal officers of multi-member boards cannot be lawfully dismissed in the face of statutory provisions limiting removal to particular causes. But in cases involving the NLRB and MSPB, the Supreme Court prevented the orders of the lower courts reinstating the dismissed officials from going into effect in a per curiam opinion issued over the dissent of three Justices. The Court expressed open-mindedness on whether either agency “falls within [a] recognized exception,” to what the Court now deems the presumptive rule that the President “may remove without cause executive officers who exercise [executive] power on his behalf”—an exception the Court describes as “narrow.” 129 Trump v. Wilcox, 1415 S. Ct. 1415, 1415 (2025). The Court’s opinion perhaps does not augur well for the fate or scope of Humphrey’s Executor in the immediate future. But if the historic struggles that we have described in this Essay are any precedent, disputes over the appropriate scope of the “executive power” will continue into the administrations of the future.

*Aditya Bamzai is the Martha Lubin Karsh and Bruce A. Karsh Bicentennial Professor at the University of Virginia School of Law. Peter M. Shane is a Distinguished Scholar in Residence and Adjunct Professor at the New York University School of Law, and the Jacob E. Davis and Jacob E. Davis II Chair Emeritus at the Ohio State University Moritz College of Law. This collaboration was prepared for a Symposium on “The First 100 Days of the Trump Administration” cohosted by the Stanford Neukom Center for the Rule of Law, Stanford Law Review, and Deborah L. Rhode Center on the Legal Profession. For helpful comments and encouragement, the authors thank Divya Bamzai, Diego Zambrano, and the editors of the Stanford Law Review Online.