Making DOGE Constitutional

The “Department of Government Efficiency” (DOGE) has been hard to pin down. In the wake of last year’s election, Elon Musk and his erstwhile partner, Vivek Ramaswamy, gestured toward some of the effort’s contradictory impulses in a Wall Street Journal op-ed. “We will serve as outside volunteers, not federal officials or employees,” they insisted. But “unlike government commissions or advisory committees, we won’t just write reports or cut ribbons. We’ll cut costs.”

It was hard to see exactly how outside volunteers would cut costs. And evidently, the DOGE team could not resolve the question because by the time the administration got going, Ramaswamy had parted ways with the project, and Musk was made a “Special Government Employee” after all.

Yet even as a White House office, if not quite the “department” its name suggests, DOGE remains a kind of foreign object in our constitutional system, purporting to exercise on the president’s behalf a power of the purse that does not belong to the president. As it has taken shape, the effort involves lodging representatives (often Silicon Valley engineers or managers on leave) within various government agencies and empowering them to find wasteful, fraudulent, or progressive-freighted spending to eliminate. There are other facets of DOGE’s work, such as culling the civil-service and modernizing government computer systems. But digging for loose change beneath the couch cushions has taken center stage.

Musk has claimed these efforts could save up to a trillion dollars a year, although the cuts proposed so far would hardly approach such a level. But the administration has not been clear about just how reductions in spending are supposed to take effect. Listing targets for cuts and actually reducing spending are not the same thing. And to the extent that DOGE proposes to trim appropriated money, the administration faces legal obstacles to making cuts without permission from Congress.

The question of just how to deal with those obstacles reaches to the core of the White House’s agenda of constitutional confrontation. Some of President Trump’s advisors are eager for a fight over the president’s freedom to impound—that is, to willfully refrain from spending appropriated money. They insist that impoundment is an inherent executive power and argue that Congress exceeded its constitutional bounds when it legislated limits on the practice half a century ago. That argument is part of a broader critique of the constitutional order advanced by a number of Trump’s aides who want to radically expand executive power. They appear to view DOGE’s work as providing ideal fodder for a battle that will leave the president bestriding our government like a colossus.

But this view is dangerously misguided. By beginning from the premise that the American constitutional system is already too far gone to save, it threatens to further deform that system. By advancing a view of the separation of powers that the Supreme Court is (rightly) likely to reject, it risks leaving the presidency weaker rather than stronger. And by ignoring the character of DOGE’s proposed spending reductions, it overlooks an obvious opportunity to strengthen both Congress and the presidency and enact durable (albeit modest) spending cuts.

That opportunity presents itself in the very law that the administration wishes to attack: The Congressional Budget and Impoundment Control Act of 1974. That law created the modern federal budget process, with its formal budget resolutions, reconciliation bills, and dozen annual appropriations measures. And while it restricted the president’s freedom to impound, the statute also created a legal mechanism by which the two elected branches could cooperate on targeted spending reductions.

That mechanism was a response to presidential abuses. Impoundment, at any scale, was always a disputed power. But modern presidents, especially starting in the mid-20th century, began to deploy it more aggressively by asserting not an inherent but a statutory authority to withhold appropriated dollars. The Antideficiency Act, enacted in 1905 and significantly amended in 1950, was intended to make sure that the president could not spend money that Congress had not appropriated. But the law included language that seemed to permit the president to also refrain from spending money that had been appropriated by putting such money into reserve funds and thereby essentially returning it to the Treasury.

Such reserve funds, the law stated, could be established “to provide for contingencies, or to effect savings whenever savings are made possible through changes in requirements, greater efficiency of operations, or other developments subsequent to the date on which such appropriation was made available.” That last clause is, of course, very vague. “Other developments” can mean pretty much anything. And in the decades after the Second World War, presidents increasingly came to use that vagueness as a license to impound at will.

Congress grew resentful of that practice. And in the Impoundment Control Act (which ended up being folded into the larger Budget Act), legislators finally closed the loophole with a provision that replaced the above quoted text with: “reserves may be established solely to provide for contingencies, or to effect savings whenever savings are made possible by or through changes in requirements or greater efficiency of operations.” No more “other developments,” and so no more cuts to spending that Congress had mandated but the president just didn’t like.  

The ICA thus sharply restricted the potential for impoundment. However, it also created a new legal pathway for targeted spending reductions. The law established a process called “rescission,” by which the president can propose specific reductions in current-year spending to Congress, and Congress can take up the proposal as a privileged and expedited bill, which cannot be filibustered and therefore requires only simple majorities in both houses to pass.

This expansion of the potential for current-year spending cuts is, in practical terms, more significant than the ICA’s constraint on the impoundment power. The law increased the potential for targeted spending reductions. It just required such reductions to be taken up cooperatively by the two elected branches rather than pursued by the president alone.

And the rescission mechanism is perfectly suited to the circumstance that Republicans now confront: They have narrow majorities in both houses and a presidentially appointed team pointing out potential modest, focused cuts in spending. Inciting a fight over impoundment in this circumstance makes no sense. It would undermine Congress’s spending authority with the tacit acquiescence of a congressional majority, yet also lead to a needless constitutional conflict in the courts that the administration would likely lose. The Republican president and Congress would thus both be left weaker.

Rescission offers an obvious alternative. It could draw on DOGE’s work to strengthen both Congress and the president and allow Republicans to highlight their spending reductions. Rescission bills can be small and frequent. Rather than setting up a single, cataclysmic confrontation, the president could send Congress modest rescissions bill every week if he so desired. Each bill would be considered on its own merits and could be amended by Congress giving legislators more of a role, more control over the details, and more ownership of the outcome. Some Democrats might well find it difficult to resist joining in the rescission of unpopular spending. But even if they didn’t, Republicans could pass rescission bills on a party-line vote if necessary, despite their narrow majorities.

Such small and frequent rescissions could also satisfy the administration’s more cynical political aims: DOGE’s proposed cuts are relatively small, and Republicans (at least as reflected by the ambitions of the tax legislation they’re pursuing) are planning to increase the deficit on net in the coming years, not to reduce it. However, by creating a recurring spectacle of budget cutting, they could nonetheless claim to be budget hawks all they wanted.

Ultimately, however, the promise of rescission is not cynical. It offers a way to help Congress reassert its authority over spending while allowing the administration to advance spending cuts. It could turn DOGE’s haphazard improvisations into durable constitutional action.

If the administration truly wants to leverage DOGE’s work toward substantive and political gains, the path of rescission outlined by the Impoundment Control Act would be the natural choice. To opt for a battle over impoundment instead would be to choose fruitless, performative confrontation over substantive policy progress.

Yuval Levin is the director of social, cultural, and constitutional studies at the American Enterprise Institute and the editor of National Affairs. His new book, American Covenant: How the Constitution Unified Our Nation - and Could Again, has just been published.