Congress Executive Branch

A Primer on the Impoundment Control Act

Zachary Price
Tuesday, January 28, 2025, 1:50 PM

Refusals to undertake required spending are already a major theme of Trump’s second term, but the law limits such executive action.

US Congress on Capitol Hill, Washington DC (https://commons.wikimedia.org/wiki/File:US_Congress_02.jpg)

Published by The Lawfare Institute
in Cooperation With
Brookings

During his 2024 campaign, President Trump indicated that he planned to “restore executive branch impoundment authority to cut waste, stop inflation, and crush the Deep State.” Russell Vought, Trump’s pick to lead the Office of Management and Budget (OMB), the agency that oversees spending for the executive branch, repeated this promise at his Senate confirmation hearing on Jan. 15. But what is “impoundment authority,” and what would restoring it mean?

 An impoundment is an executive refusal to spend funds appropriated by Congress. Although U.S. presidents historically impounded funds with some regularity, Congress curtailed this practice by statute after President Nixon abused it. As amended over time, the Impoundment Control Act of 1974 (ICA) now limits the executive branch’s authority to decline to spend or commit to spending funds that Congress has appropriated.

At the same time, other statutes do the opposite: While the ICA forbids officials from refusing to use funds in an appropriation, other laws bar them from spending, or even committing to spend, money without one. In particular, the Anti-Deficiency Act, a law enacted in the 19th century and strengthened over time, makes such action unlawful—and sometimes even criminal. 

Current law thus often catches the executive branch in a vise: Presidents can neither spend money without an appropriation nor refuse to spend funds once Congress has provided them. From both directions, Congress has reinforced its “power of the purse”—its authority to control the use of federal money.

Given this constraint’s potency, it may not be surprising that President Trump would want a way to escape it. To that end, officials floated proposals to narrow the ICA’s interpretation during the first Trump administration, and in the meantime one of the same officials has suggested that the ICA is altogether unconstitutional. Although the statute presents some interpretive puzzles, the constitutional arguments against it are weak and should not gain traction.

Historical Background

Beginning at least with Thomas Jefferson, U.S. presidents have asserted an authority to impound federal funds. In 1803, for example, Jefferson declined to spend $50,000 that Congress had appropriated to build 15 gunboats on the Mississippi. The military threat that prompted the appropriation had disappeared, Jefferson explained, so he sought to conserve funds for the Treasury.

Jefferson’s action was not actually contrary to the letter of the applicable appropriations law: The statute had authorized the expenditure without requiring it. By contrast, in some later examples, presidents impounded funds despite statutory language that seemed to make the spending mandatory, or at least not clearly optional. Overall, however, a general understanding seems to have developed that presidents were often free to economize when they could in carrying out Congress’s objectives.

Sometimes such presidential actions produced considerable friction with Congress, but these disputes were worked out informally, typically by one branch or the other acquiescing to the other’s preferences. Such conflicts finally became acute during Nixon’s presidency. Whereas earlier presidents had often acted with tacit congressional support and reserved the most significant impoundments for military and foreign affairs spending, Nixon impounded large sums from social programs that congressional Democrats favored and he did not. These conflicts, along with the larger scandals of the Nixon administration, led Congress to finally enact statutory limits on impoundment.

Statutory Limits on Impoundments

The product of these Nixon-era controversies, the Impoundment Control Act of 1974, limits impoundments by sorting them into two categories and imposing sharp limits on each.

Limits on “Deferrals”

 The first form of impoundment addressed by the statute is a “deferral,” meaning a delay in the “obligation” or commitment of appropriated funds. Under the Anti-Deficiency Act, the executive branch generally must “apportion” funds for agency activities: It must spread out the funding over the fiscal year so that no shortfall arises. If the administration puts off the spending, however, it makes a “deferral” under the ICA and must report its action to Congress. The House and Senate may then consider legislation to override the deferral using special expedited procedures.

Though earlier versions of the statute allowed a broader range of deferrals, the ICA today allows deferrals only “to provide for contingencies,” “to achieve savings made possible by or through changes in requirements or greater efficiency of operations,” or “as specifically provided by law.” The upshot is that, absent specific statutory authority, executive officials are not supposed to delay spending based on disagreement with the policy underlying it; they can instead make deferrals only to address practical obstacles or to employ funds more efficiently. As explained below, however, the scope of any authority to delay spending for “programmatic” rather than “policy” reasons has emerged as a recurrent point of controversy.

Deferrals also may not extend beyond the end of the fiscal year. Instead, executive officials must release any funds that expire at the end of a fiscal year in time for officials to prudently employ them to achieve Congress’s purposes. The executive branch cannot thwart Congress’s goals by deferring spending until the very last second so as to prevent it from being spent at all. 

Limits on “Rescissions”

If the executive branch wants to cancel spending altogether, it must make the other type of impoundment, namely, a “rescission.” The ICA does not allow the executive branch to make this second form of impoundment at all on its own authority. The statute only allows the president to propose rescissions to Congress, which again may consider such proposals using expedited procedures.

The statute does allow presidents to put spending on hold while Congress considers a rescission proposal, but these holds may last for no longer than 45 days of “continuous session” in Congress (meaning days when Congress is at work, including weekends and other breaks of no more than three days). After that period, the ICA requires release of the funding.

The Overall Framework

This statutory framework may seem confining, and indeed it is. It is designed, after all, to put Congress in the driver’s seat when it comes to federal spending. At the same time, however, the appropriations laws themselves often provide some executive flexibility.

For one thing, appropriations laws sometimes define spending objectives broadly, a choice that effectively confers discretion on executive officials to decide how funds will be used. In addition, appropriations statutes or other laws often provide agencies with some authority to “transfer” funds between appropriations or “reprogram” them for different purposes within a single appropriation. Although these authorities typically carry their own limits and requirements, they may give executive officials significant policy discretion. President Trump, for example, employed such provisions to spend greater amounts on border wall construction than Congress had specifically provided for that purpose, although courts deemed some of this spending unlawful.

It is also worth noting that Congress attempted to establish an additional form of impoundment authority in the Line Item Veto Act of 1996. That statute allowed presidents to cancel certain spending items within five days of an appropriation’s enactment, subject to a congressional override through expedited new legislation. The Supreme Court, however, held in Clinton v. New York that this cancellation power amounted to an unconstitutional line-item veto (meaning a power to veto particular clauses in a law rather than the bill as a whole). Although the Court acknowledged the president’s “traditional authority to decline to spend appropriated funds,” it rejected the government’s argument that this practice supported the Line Item Veto Act’s cancellation power. Unlike all prior statutes invoked by the government, this one, the majority reasoned, gave “the President the unilateral power to change the text of duly enacted statutes.”

As a result of this holding, any impoundment power afforded by the Line Item Veto Act is defunct; presidents hold only the limited deferral and rescission authorities afforded by the ICA. 

Interpretive Questions About the Statute

Several interpretive questions about the ICA arose during the first Trump administration and may well arise again.

Pocket Rescissions

One key question is whether the ICA allows “pocket rescissions.” Recall that the ICA allows presidents to put a 45-day hold on spending that they have proposed for rescission. But what if they submit the proposal less than 45 days before the spending expires? Can they run out the clock and allow the money to go unspent, or must they release it in time for it to be employed before the fiscal year runs out? (The name “pocket rescission” derives from the “pocket veto,” which occurs when a president prevents Congress from overriding a veto by taking no action on legislation presented late in a congressional session.)

During his first administration, President Trump twice proposed rescissions from foreign aid accounts less than 45 days before the funding in question would expire. In a subsequent defense of this practice, OMB, maintained that any resulting rescission of the spending—any lapse in authority for it due to an appropriation’s expiration—would be lawful because nothing in the ICA’s text restricts the timing of rescission proposals. By contrast, the Government Accountability Office (GAO)—a congressional agency with statutory authority to monitor spending and enforce the ICA and other spending-related statutes—reached the opposite conclusion. It issued an opinion holding that such pocket rescissions are unlawful because they defeat the statute’s purpose of preventing rescissions without new legislation.

This interpretive question may be a close call; it implicates the sort of conflict between statutory text and purpose that features in many difficult cases of statutory interpretation. In the end, however, political pressure from Congress led the Trump administration to back down from its attempted pocket rescissions and release the funds in question. That practical outcome might carry some weight in future interpretive disputes. At any rate, such historical practice has often factored in the interpretation of the ICA and other spending-related statutes, in part because standing limitations and the compressed time frame for disputes have limited courts’ role in resolving them.

Programmatic Delays

 Another interpretive issue relates to the scope of the statutory limits on deferrals. To avoid treating every good-faith delay as a deferral requiring a report to Congress, GAO has long understood the ICA to allow what it calls “programmatic delays”—essentially, delays caused by external factors that prevent expeditious employment of funds to achieve statutory goals.

Though it may make practical sense, this interpretation, as Eloise Pasachoff has observed, fits uneasily with the ICA’s text, which defines deferrals very broadly. In practice, it could even swallow the entire category of non-policy deferrals, which are the only sort of deferrals the statute allows. Perhaps for that reason, no president appears to have reported a deferral to Congress since fiscal year 2000, though they have surely delayed spending on programmatic grounds during that period.

In any event, the boundary between lawful programmatic delays and unlawful deferrals is contestable and generated acute controversies during the past two administrations.

 One such dispute involved delays in military assistance to Ukraine during the first Trump administration. As evidenced by the famous phone call that led to his first impeachment, President Trump apparently sought to leverage Ukraine aid to induce a Ukrainian criminal investigation into Joe Biden’s family dealings in that country. As a practical matter, OMB aided this apparent goal by placing appropriated funds on hold and releasing them only toward the end of the relevant fiscal year, after Trump’s phone call came to light.

OMB argued that this hold was programmatic because it stemmed from a policy review oriented toward “ensur[ing] that funds were not obligated prematurely in a manner that could conflict with the President’s foreign policy.” GAO, however, rejected this characterization. It concluded that the administration made an unlawful deferral because its reasons for the delay were rooted in policy considerations rather than external obstacles to efficient release of the funds.

A second dispute then arose in the Biden administration, this time relating to expenditures for border wall construction. Congress had appropriated over a billion dollars a year for border wall construction during several years before Biden’s inauguration, and it made the funds available each time for a five-year period. On his first day in office, however, Biden issued a proclamation directing federal agencies to “pause” border wall construction. His administration then effectively delayed further expenditures by reimposing environmental reviews and other legal requirements that the Trump administration had waived and by engaging in statutorily required consultations with various constituencies before settling on plans for construction.

 Republicans in Congress objected that these delays amounted to a deferral just as Trump’s Ukraine delays had, but this time GAO concluded that the delay was programmatic because the administration was simply fulfilling legal prerequisites to the spending.

 GAO’s distinction is plausible: Biden’s slow-walking of border funds seems to have been rooted in legal requirements in a way that Trump’s Ukraine delays were not. (Although a statute allowed the Trump administration’s waivers to expedite border wall construction, GAO reasoned that the law gave Biden the discretion to reconsider those choices.)

Nevertheless, both of these examples involved pretextual justifications for delaying spending that Congress had mandated. It was perfectly apparent, after all, that Biden opposed expenditures on a border wall, and his administration went out of its way to reimpose legal obstacles to construction that the Trump administration had removed.

A further controversy along these lines is now brewing. In one of his first-day executive orders, President Trump declared it the policy of his administration to “unleash” American energy production and end electric-vehicle mandates. This order directed federal agencies to “pause the disbursement of funds” under two of Biden’s signature legislative achievements, the Inflation Reduction Act of 2022 and the Infrastructure Investment and Jobs Act. This pause is meant to ensure that any such disbursements are “consisten[t] with the law and the policy” outlined in the order. A second first-day order likewise imposed a 90-day pause on disbursement or obligation of foreign aid funds “pending reviews of such programs for programmatic efficiency and consistency with United States foreign policy.” 

More recently, OMB directed federal agencies to pause “all activities related to obligation or disbursement of all Federal financial assistance” apart from Medicare and Social Security. OMB ordered this pause, which must take effect by 5:00 p.m. on Jan. 28, to enable a review of grant programs and other forms of assistance for compliance with the president’s executive orders, including those ending diversity, equity, and inclusion (DEI) in the federal government, reversing green energy requirements, and defining gender in biological terms.

All these pauses might constitute unlawful deferrals under the ICA. If challenged, the administration will likely argue that it is doing just what the Biden administration did with the border wall: pausing expenditures to ensure compliance with legal prerequisites and enable it to make its own discretionary choices about exercising various authorities. In reality, of course, policy concerns may well be motivating some of these delays, but the same was true of Biden’s border wall pause. Sorting out whether the administration’s position is plausible, or whether the delays are instead unlawful policy deferrals, will depend on canvassing the legal requirements and practical exigencies applicable to the huge range of programs affected by these orders. 

Looming Constitutional Questions

Beyond advancing permissive interpretations of the ICA, the second Trump administration might claim that the law is altogether unconstitutional. Indeed, to the extent the current pauses in federal spending cannot be justified as programmatic delays, the administration might well assert that the constitution grants President Trump an indefeasible authority to order them.

Mark Paoletta, the OMB general counsel from Trump’s first term whom Trump has renominated for the role, argued in a pair of memos last year that presidents hold a constitutional power to impound funds notwithstanding the ICA. These memos are not entirely clear about the scope of the purported impoundment power that the new administration might advance. Overall, the memos appear to defend an across-the-board presidential power to refuse to spend funds. Yet they also advance rationales for impoundment that would apply only in more limited circumstances, such as when two statutes conflict or when an appropriation’s goals can be fully accomplished for less than the amount provided.

Although statutory conflicts could indeed require narrow spending adjustments in some situations, the broader argument against the ICA’s constitutionality is weak. As support for their position, Paoletta and his co-authors point to the historical practice of impoundment, arguing that it establishes a valid presidential authority as a matter of “history and tradition” (something the Supreme Court has considered controlling in key recent cases). But examples from pre-ICA practice do not support a constitutional authority to defy the ICA’s specific limits on impoundments. At best, they demonstrate a practical background understanding, shared by Congress as well as many presidents, that governing statutes before the ICA allowed presidents to impound appropriated funds in some circumstances. By stepping in to regulate this practice, Congress abolished that prior understanding and limited impoundments by statute. Presidential practice from before the ICA thus should not provide precedent for actions defying it.

In fact, even before the ICA, the Supreme Court rejected one of President Nixon’s impoundments in Train v. City of New York, reasoning that the appropriations statute in question required the expenditure. Although the Court did not address constitutional issues in Train, neither did it hint at any constitutional problem with a spending mandate, despite contemporaneous claims by Nixon that he held a constitutional impoundment power. In a partial dissent in the Clinton case, Justice Antonin Scalia observed that the Court in Train “proved [Nixon] wrong.” By the same logic, the ICA’s limits now bind the president. They are part of the law that the Constitution obligates the president to “faithfully execute[].”

***

 Statutory limits on impoundments were a major issue in the first Trump administration, and they seem poised to present issues again in this one. Although Republicans control both houses of Congress at present, unobligated funds from some past appropriations are already in the administration’s crosshairs, and Congress might well impose new spending obligations that Trump opposes, particularly if Democrats regain control of the House or Senate in the next midterm elections. In this context, it is not surprising that Trump hopes to “restore” the impoundment authority exercised by presidents before 1974, but that does not mean that doing so will be lawful.


Zachary Price is a professor at the University of California College of the Law, San Francisco. His article “Funding Restrictions and Separation of Powers” appeared in the Vanderbilt Law Review in 2018, and his book “Constitutional Symmetry: Judging in a Divided Republic” was published by Cambridge University Press last year.

Subscribe to Lawfare