A key congressional officer, the Comptroller General monitors government spending, rules on its legality, and polices impoundments, among other things. President Trump, however, clashed repeatedly with the last Comptroller General, and the position has been vacant since December, filled on an acting basis by a senior staff member. Meanwhile, the Trump administration’s “appropriations presidentialism”—its push for stronger executive control over spending—has prompted reformers to wonder if the office might need some sort of revitalization.

It seems to me that reforms along these lines could indeed be beneficial (see here for some proposals), but in addition to their practical pros and cons, they would raise a number of fraught constitutional questions that require careful attention.

Background

To begin with some history, comptroller positions with authority to approve spending were a feature of federal administration from the start. As Mark Thomas recounts in an excellent paper on the subject—written as a student no less—early comptrollers nevertheless did little to discipline expenditures. In the early republic (as I discuss in a draft article of my own), executive officials often illegally shifted funds between accounts, employed retained balances for unauthorized purposes, and incurred obligations without supporting funds. Though comptrollers in each agency needed to countersign warrants for treasury withdrawals and thus could have imposed a check on some abuses, in practice they showed little independence. Thomas characterizes them as “deferential to the point of enabling malversation.”

The situation improved in some respects when the Dockery Act of 1894 consolidated comptroller functions into a single office and rendered the new Comptroller’s decisions binding on the executive branch. But the real watershed was the landmark federal budget reform of 1921. Whereas individual agencies previously submitted independent spending requests to Congress, the 1921 legislation required presidents to present a unified executive branch budget each year. At the same time, it established a new executive-branch Budget Bureau (the precursor to today’s Office of Management and Budget) and a new “Comptroller General” to monitor spending for Congress. As I discuss in my paper, this change and other reforms in the following decades appear to have established strong norms of executive appropriations compliance by the mid-twentieth century.

Today, the Comptroller General remains the central authority on federal appropriations law. Yet the position’s powers used to be more concrete. As Thomas documents, early Comptrollers General not only issued opinions on spending questions, as they do today, but also issued warrants to authorize treasury withdrawals, audited and settled agency accounts, immunized officials from personal liability for expenditures, and certified litigation judgments and settlements for payment from the Judgment Fund. In 1974, the Impoundment Control Act empowered them to sue when the executive branch refused to spend funds that Congress declined to rescind, and the “Gramm-Rudman-Hollings” deficit control legislation in 1985 required them to impose pro rata spending cuts through “sequestration” when federal deficits exceeded certain caps.

These powers have atrophied. In some cases, the change occurred administratively. For example, regulations adopted beginning in 1950 waived the authority to countersign treasury warrants, and comprehensive auditing of the executive branch quickly proved impracticable. But the biggest problem proved to be the agency’s design. Appointed to fifteen year terms, Comptrollers General are removable during their terms only for cause and only through a joint resolution, meaning a resolution approved by both houses of Congress and the President (or else passed by two-thirds margins over the President’s veto).

In 1986, in Bowsher v. Synar, the Supreme Court held (correctly, it seems to me) that Congress’s role in removing Comptrollers General precluded them from exercising executive functions. Bowsher thus invalidated the Comptroller General’s sequestration authority, but Thomas observes that its reasoning should equally invalidate any other executive powers of the office. Indeed, for that reason, Congress in the 1990s moved the authority to certify payments from the Judgment Fund to the Treasury Department and the authority to settle private claims against agencies to the Office of Management and Budget.

To the extent the executive branch is once again undermining Congress’s proper authority over spending, this history naturally prompts the question whether changes to the Comptroller General could help restore better balance. Again, I believe the answer is yes, but any reform would need to account for Bowsher and other Article II requirements such as the Appointments Clause.

Option 1:  Congressional Appointment

One proposed change, advocated in a recent letter to Congress and an R Street report, would adopt a “less is more” strategy and provide for appointment of the Comptroller General by Congress, without any participation by the president. The Architect of the Capitol and the Director of the Congressional Budget Office are already selected this way, and the House recently passed legislation to cut the President out of appointments for the Librarian of Congress and the Director of the Government Publishing Office. Such arrangements are lawful under Bowsher and the Appointments Clause insofar as the positions’ duties are neither executive in character nor so significant as to make the positions “officers of the United States.”

For congressional appointment to be valid in this case, the Comptroller General might need to lose a few remaining powers, such as its role in adjudicating bid protests for government contracts. In addition, its appropriations law rulings could not carry concrete legal effects. As Thomas explains, however, that is already effectively true by virtue of Bowsher, whether or not all government officials realize it. By the same token, although cutting presidents out of the appointment process might reduce the position’s prestige and influence inside the executive branch, its opinions could continue to carry substantial persuasive force, just as they do now, especially if Congress backed them up through its own actions.

Comptroller General rulings could even have certain indirect legal effects. For example, executive officials who defied an understanding of spending limits spelled out in a Comptroller General opinion might be on notice that any similar action in the future would qualify as “willful” and thus risk criminal penalties under the Anti-Deficiency Act. Similarly, conflicting Comptroller General opinions might weaken reliance defenses in future litigation when executive officials follow doubtful internal legal guidance from the executive branch.

As the heads of the Government Accountability Office, which investigates executive agencies and issues reports recommending changes, Comptrollers General do also hold the power to compel production of evidence. In addition, as noted earlier, they may sue over impoundments. These powers are not executive in character; they are congressional powers delegated to the office. Nevertheless, there could be an argument that they constitute “significant authority” under federal law and thus render the Comptroller General an “officer of the United States” who must be appointed in accordance with the Appointments Clause (meaning by the President with Senate advice and consent).

Though I hope to explore this question further, it seems to me that this argument should fail. Allowing Congress to delegate non-lawmaking legislative powers to officers appointed by the House or Senate (or both) seems consistent with past practice and could help Congress defend its institutional interests against an increasingly powerful executive branch.

Thus, on the whole, rendering the Comptroller General a more squarely congressional officer could carry benefits and would seem to satisfy Article II.

Option 2:  A Comptroller Commission

Moving in the opposite direction, Thomas advocates rejecting the Comptroller General’s current advisory character and instead restoring more concrete powers. Specifically, he proposes creating a multimember “Comptroller Commission” with authority to issue appropriations law decisions, direct the disbursal of funds, and enjoin unlawful impoundments. In this arrangement, the comptrollers’ actions would be legally binding, and in consequence they would need to be presidentially appointed and at least to some extent presidentially controlled.

Some measure of legal independence might be possible for such a commission. Although the Supreme Court held in Trump v. Slaughter that presidents can normally remove the executive agency heads at will, Thomas argues that the Comptroller Commission could be analogous to the Federal Reserve Board of Governors, a body whose independence the Supreme Court upheld last June in Trump v. Cook. The argument for independence here strikes me as a stretch, but who knows? Cook’s exception to Slaughter is incoherent and poorly unexplained, so it is hard to say.

As an alternative, Thomas proposes making removal of commissioners costly by requiring their approval for spending. Under this design, commission vacancies would effectively ensure a government shutdown and resulting political blowback for the President. That arrangement seems constitutional but it would create powerful incentives for presidents to control the commission and thus blunt its utility.

Option 3:  An Inferior Officer

A different reform along these lines might rely on an inferior officer rather than a new commission. In other words, Congress might go back to its approach from the early republic and establish comptroller positions, perhaps in the Treasury Department, with authority to police certain forms of spending non-compliance, either across the board or in particular domains. To render their decisions effective without resort to litigation, such comptrollers might be given authority to suspend other officials’ pay or impose personal liability on them unless those officers released required spending or recovered illegal payments.

Under Bowsher and other controlling decisions, such a position would be an office and would need to fall under the President’s chain of command. Nevertheless, under the Supreme Court’s case law to date, such a position could be tenure-protected so long as the agency head could countermand the officials’ decisions. Alternatively, the arrangement should be constitutional so long as the Treasury Secretary could remove comptrollers at will, even if the Secretary could not personally discharge their functions. Although such arrangements would enable determined presidents to control the comptrollers’ actions, the need to fire them or countermand their rulings to get the President’s way could raise the political costs of doing so. In consequence, norms of legal credibility and seriousness might develop around the position.

Option 4:  An Automated Fisc

Finally, Rohan Grey has advanced a much more wide-ranging proposal to “digitize the fisc,” in effect replacing (or complimenting) the Comptroller General with an automated system of fiscal control over executive agencies. Under his proposal, Congress would maintain a “bicameral digital database” recording “public funds appropriated, drawn, held, spent, and collected across the federal government.” Rather than receive funds through the executive branch’s Office of Management and Budget, as they do now, agencies would “‘draw’ on [a] credit line [recorded in this database] up to statutorily determined qualitative and quantitative limits by using a ‘Public Credit Card,’ issued and administered by Congress.”

To the extent this arrangement automated the legal effects of spending legislation, it might avoid conferring any executive authority on legislative officials in violation of Bowsher and other cases. Depending on the precise design, however, the administration of the digital database and associated credit lines could constitute a form of executive action, just as sequestration did under Gramm-Rudman-Hollings. To the extent executive functions are involved, presidents would need some means of control over officers performing those functions. Implementing this innovative proposal, if Congress had the will to do so, would thus require careful attention to potential Article II pitfalls.

*     *     *

The Comptroller General has helped shore up Congress’s power in the past and might do so again. To be effective, however, any reform will need to get the law right along with the institutional design.

Zachary S. Price is a Professor at the University of California College of Law, San Francisco (formerly UC Hastings).