Concept
Bowsher and Congressional Removal
Removal of officers
Bowsher and Congressional Removal
Article II gives Congress one express way to end an executive officer's tenure: the House impeaches and the Senate convicts for treason, bribery, or other high crimes and misdemeanors.1 Bowsher v. Synar held that this is the only way. Congress cannot reserve for itself the power to remove an officer charged with the execution of the laws except by impeachment.2 The Court has flatly rejected congressional efforts to take any larger role in removal: Congress may not “draw to itself,” other than through impeachment, the power to remove or the right to participate in the exercise of that power.3 Myers v. United States and Humphrey's Executor v. United States asked how far Congress may restrict the President's own power to remove. Bowsher asked whether Congress may hold a share of that power itself, and the answer was no. That answer draws the clearest line in the removal cases, and it explains why some statutes that limit removal survive while others fail.
The statute in Bowsher made the Comptroller General the officer whose judgment set federal spending cuts in motion. Under the Balanced Budget and Emergency Deficit Control Act of 1985, known as Gramm-Rudman-Hollings, a deficit estimate above the statutory target triggered a process of automatic reductions. The Office of Management and Budget and the Congressional Budget Office calculated the cuts required and sent a joint report to the Comptroller General. He weighed those estimates with independent judgment and reported his conclusions to the President. The President then had to put the reductions into effect unless Congress met the deficit goal another way within a set time.4 The cuts thus followed from the Comptroller General's conclusions, so the terms of his tenure became the center of the constitutional challenge.
The challenge focused on removal. The President nominates the Comptroller General from a list of three names recommended by the Speaker of the House and the President pro tempore of the Senate, and the Senate confirms the nominee. Yet Congress may remove the officer not only by impeachment but also by joint resolution “at any time” on any of five grounds: permanent disability, inefficiency, neglect of duty, malfeasance, or a felony or conduct involving moral turpitude.5 The President could veto such a resolution, but a two-thirds vote of both Houses could override the veto, so the Court read the provision as authorizing removal by Congress alone.6 The removal rule came from the Budget and Accounting Act of 1921, not from the deficit statute. The constitutional problem arose when Congress joined that older rule to a newer grant of executive duties.7
The Court rested its holding on text, history, and precedent. The text comes first. Once an officer has been appointed and confirmed, the Constitution provides for removal by Congress only upon impeachment by the House and conviction by the Senate, and impeachment may rest only on treason, bribery, or other high crimes and misdemeanors.8 History agrees. In the First Congress's debate over the Department of Foreign Affairs, which this chapter has already traced, James Madison urged that Congress have no part in removing executive officers other than by impeachment, and the Court recounted that his view prevailed.9 The Court gave that “Decision of 1789” weight as contemporaneous evidence of the Constitution's meaning, because many Members of the First Congress had taken part in framing it.10 Precedent is the third support. The statute in Myers required Senate approval before the President could remove a postmaster, and Chief Justice Taft held that Congress may not “draw to itself, or to either branch of it, the power to remove or the right to participate in the exercise of that power.”11 Humphrey's Executor then upheld a for-cause limit on the President's removal of Federal Trade Commissioners and confined Myers to “purely executive officers,” because the FTC acted quasi-legislatively and quasi-judicially, free from executive control.12 Humphrey's Executor and Wiener v. United States addressed the separate question of Congress's power to limit the President's removal of a commissioner, and the Court read them as reaffirming the bar on congressional participation.13
From that base the Court reached its holding, and the reasoning explains why the rule is categorical. An officer fears and obeys whoever can remove him: “it is only the authority that can remove him, and not the authority that appointed him, that he must fear and, in the performance of his functions, obey.”14 Humphrey's Executor made the same observation from the other side. One who holds office only during the pleasure of another cannot be depended upon to maintain independence against that person's will.15 An officer whom Congress can remove therefore has a standing reason to execute the laws as Congress wishes. Placing execution in an officer answerable only to Congress would, in practical terms, reserve to Congress control over the execution of the laws, and Congress has no power to execute them.16 The Court tied the point to INS v. Chadha, which struck down a one-House legislative veto because Congress can carry out a policy decision in only one way, by bicameral passage followed by presentment to the President.17 An officer controlled by Congress would work as a standing veto of the same kind. Congress could remove, or threaten to remove, the officer for executing the laws in any fashion it found unsatisfactory, and no bill would have to clear either House or reach the President. Chadha forbids control of that sort over execution, so the Court treated congressional removal power over an executing officer as equally impermissible.
Appellants answered that the Comptroller General performed the office's duties independently and was not subservient to Congress, and the Court rejected that answer on the statute's own terms. The critical factor was removability: the Comptroller General could be removed only at Congress's initiative.18 The legislative history confirmed the design. One member wanted the officer brought under the sole control of Congress, so that Congress could remove him without the long process of impeachment. Another described the plan as giving the legislative branch control of the audit “not through the power of appointment, but through the power of removal.”19 The statutory grounds for removal were also broad. Inefficiency, neglect of duty, and malfeasance, as Congress would interpret them, could support removal for many perceived transgressions of the legislative will. The Constitutional Convention had refused to permit impeachment for “maladministration” after Madison objected that so vague a term would equal tenure during the pleasure of the Senate.20 Congress had also long treated the office as its own, declaring in the Reorganization Acts of 1945 and 1949 that the Comptroller General and the General Accounting Office are part of the legislative branch.21 Finally, the Court refused to wait for Congress to use the power. The officer's wish to avoid removal by pleasing Congress creates the subservience in the present, and the separation of powers cannot turn on a judicial estimate of whether an officer is on good terms with Congress.22
Bowsher also fixed a distinction on which later cases depend. Statutes that set judicially enforceable substantive limits on a removal decision made by the President or another executive official can be permissible. Humphrey's Executor accepted such a limit when Congress, creating agencies meant to act independently of executive control, forbade removal of the commissioners except for cause.23 Statutes that give Congress, or an arm of Congress, power to make or block a particular removal decision are always impermissible.24 The difference lies in who holds the decision. A for-cause standard binds an executive remover who still decides whether cause exists, and a court can enforce the standard, so Congress has written a rule and left execution where it was. A congressional role in the decision hands the final say to the branch that makes the laws, and that transfer is the harm Bowsher identified. The usual exam mistake is to see a limit on removal and conclude that Bowsher applies. The question to ask is whether Congress holds a seat in the decision itself.
Morrison v. Olson, decided two years later, supplies the other pole. The independent counsel there could be removed only by the personal action of the Attorney General and only for good cause, with no requirement of congressional approval and with the decision subject to judicial review.25 Because the statute put the removal power squarely in the hands of the Executive Branch, the Court said the case, unlike Bowsher and Myers, involved no attempt by Congress to gain a role in removal beyond impeachment and conviction.26
The rule has limits that follow from its reason. Bowsher's concern is Congress accreting to itself the power to control the functions of another branch, so the bar applies when the officer performs executive or judicial functions.27 Congress may dismiss its own committee counsel, investigators, and staff, because they assist Congress in its legislative duties and no usurpation of another branch's authority is possible.28 That limit also shows why the Comptroller General case came out as it did. Congress treated the officer as its own yet assigned him executive duties, and Congress cannot give an officer under its control a power it lacks. Mistretta v. United States later added that Bowsher does not mean one branch may never exercise any removal power over members of another. It upheld Congress's vesting in the President of the power to remove for good cause a judge serving on a nonadjudicatory commission within the judicial branch.29 The statutes that create independent agencies typically provide for removal by the President for specified causes or say nothing about removal, and none gives Congress a removal role, so Bowsher leaves them untouched.30
Bowsher is a leading example of formal separation-of-powers analysis. The majority classified the Comptroller General's activities as executive, legislative, or judicial and then asked whether the officer sat under the control of a different branch, which Congress's removal power answered.31 Justice Stevens, concurring in the judgment, rejected that approach as too dependent on labeling the Comptroller General's functions executive, observing that governmental power cannot always be given one of three labels. He regarded the Comptroller General as an agent of Congress, and he relied on Chadha for the point that Congress cannot do indirectly what it cannot do directly.32 Justice White dissented, calling the majority's view “distressingly formalistic.”33 His objections were textual and practical. The statute allowed removal only for specified cause, to be found by Congress after a hearing and presumably subject to judicial review, so the officer was no mere congressional agent.34 A joint resolution must pass both Houses and go to the President, so on his reading removal satisfied the bicameralism and presentment that Chadha demands.35 In practice, he argued, the Comptroller General was one of the most independent officers in the federal establishment, since none of the six who had held the office had been threatened with removal.36
The majority met the point about the President's role by noting that two-thirds of both Houses could override a veto, and it met the practical point by holding that structural protections cannot depend on how an office happens to be operating.37 Justice White also stressed what the majority left alone. It declined to hold that executive powers may be exercised only by officers the President can remove at will, which leaves the Humphrey's line in place.38 The next section turns to Morrison, where a more functional method reaches a different result for a removal power held inside the Executive Branch.