Donald Trump thankfully survived an assassination attempt last weekend and may well win back the presidency in November.  What constitutional theories might a second Trump administration advance?

Trump himself has suggested one possibility:  He has promised to assert a “Constitutional power to stop unnecessary spending through what is known as Impoundment.”  In other words, a second-term President Trump would likely claim the constitutional authority to “impound,” or refuse to spend, funds appropriated by Congress.  A recent report from the Center for Renewing America, written by former OMB General Counsel Mark Paoletta along with Daniel Shapiro and Brandon Stras, has now spelled out the theoretical basis for this view, arguing that historical practice before and after the Constitution’s adoption supports it.

The next President (whether Biden or Trump) should not take this bait.  Although the CRA report is well-crafted, its historical analysis is misleading and its bottom line is mistaken.  The President cannot disregard statutory mandates to spend funds, and under current law agencies normally must spend the amounts that Congress appropriates for them.

The Constitutional Text and Structure

The Constitution gives Congress the “power of the purse,” meaning the power to control government funds.  As the CRA report notes, most historical controversies over government expenditure involved executive officials—kings, governors, or presidents—exceeding the sums provided by legislatures.  Reflecting this historical concern with executive over-spending, the U.S. Constitution’s Appropriations Clause is framed negatively:  “No Money shall be drawn from the Treasury, but in Consequence of Appropriations made by Law.”

The report infers from this text and history that the framers meant “to establish a ceiling on Executive spending, not a floor, and certainly not an authority for Congress to compel the President to expend the full amount of an appropriation.”  But that inference is incorrect.  The Appropriations Clause is not the only constitutional provision that bears on executive spending.  

On the contrary, the Constitution also obligates the President to “take Care that the laws be faithfully executed.” Appropriations statutes are laws and, as such, fall within this presidential duty of faithful execution.  As a general matter, then, Presidents are constitutionally obligated to expend appropriated sums if applicable statutes make that expenditure mandatory.

As I explained in a 2018 article, there are some exceptions to this rule.  As a matter of text, structure, and historical practice, Congress cannot employ its power of the purse (either through ceilings or floors) to control presidential constitutional authorities that are “resource-independent,” meaning powers that the President could exercise even without funds from Congress (beyond his or her own constitutionally guaranteed salary).  Congress could not, for example, compel the President to issue a particular pardon by mandating expenditure for its issuance.

This principle, however, does not extend to governmental functions such as law enforcement, benefits administration, and the use of military force that Presidents cannot discharge without resources from Congress.  If Congress imposes constraints, conditions, or mandates with respect to use of funds in those areas, the Take Care Clause obligates Presidents to abide by them.

Historical Impoundment Practice

The CRA report accepts this basic picture of congressional supremacy with respect to spending caps— “ceiling[s], rather than “floor[s],” in the report’s terms.  The report argues, however, that spending mandates are different.  According to the authors, an “unbroken Executive impoundment practice and congressional acquiescence” supports a constitutional authority to decline to spend appropriated funds.

Though the report offers a survey of historical impoundment practice to support this conclusion, its analysis makes a fundamental mistake:  Essentially all the examples it relies on were consonant with governing statutes, not contrary to them.  To establish a practice-based prerogative to defy spending mandates, the authors would need to show a pattern of presidential actions in actual defiance of governing statutes.  In terms of Justice Robert Jackson’s celebrated three-part framework from Youngstown Sheet & Tube Co. v. Sawyer, they would need to show presidential action at the “lowest ebb”—when a statute has banned some action but the President undertook it anyway.  Yet the report identifies no such pattern of practice.

In fact, until Congress enacted the Impoundment Control Act in 1974, governing statutes generally allowed presidential impoundment—indeed, sometimes they did so explicitly.  For example, in 1803, in what the CRA report calls “[t]he most famous early impoundment precedent,” President Thomas Jefferson “refused to spend a congressional appropriation of $50,000 for 15 gunboats for use on the Mississippi.”  Jefferson invoked changed circumstances as a justification:  The sums “remain[ed] unexpended,” he told Congress, because “[t]he favorable and peaceable turn of affairs on the Mississippi rendered an immediate execution of that law unnecessary.”  In any event, Jefferson’s impoundment was entirely consistent with the underlying statute.  As was typical of appropriations laws at the time, the statute authorized expenditure without requiring it:  By its terms, it merely “authorized and empowered” the President to order construction of “a number not exceeding fifteen gun boats,” using “a sum not exceeding fifty thousand dollars.”

Some later impoundments did involve statutes that appropriated sums without including such explicit “not exceeding” language.  Both Congress and the executive branch, however, seem to have followed a default understanding that appropriations statutes conferred implicit authority to forego spending when it proved unnecessary or when statutory goals could be accomplished with a smaller expenditure.  Late nineteenth century Attorney General opinions thus indicated that, while executive officials should ultimately “look[] to the purpose of Congress,” appropriations statutes generally should not be considered mandatory “to the extent that [executive officials] are bound to expend the full amount if the work can be done for less.”  Employing this approach, an opinion in 1896 understood “emphatic language” in one appropriation to be permissive, while another three years later construed permissive language as mandatory.

Reflecting this same outlook, the Supreme Court held in 1838 in Kendall v. United States ex rel. Stokes that executive officials could not withhold funds that a statute obligated them to pay.  “To contend,” the Court observed, “that the obligation imposed on the President to see the laws faithfully executed, implies a power to forbid their execution, is a novel construction of the constitution, and entirely inadmissible.”  The CRA report distinguishes Kendall on the grounds that it involved payment of a defined debt to particular party—specifically, “a contract claim against the government that was adjudicated by Congress” through enactment of a private bill.  But this reading of Kendall is unconvincing.  The expenditure in Kendall may have been mandatory because it paid a debt, but it was its mandatory character, not the nature of the spending, that implicated the President’s duty of faithful execution.

In general, presidential impoundment comports with faithful execution so long as Congress intended the expenditures in question to be permissive rather than mandatory.  In fairness, as time went on, Presidents stretched this understanding to the breaking point, particularly in the military context.  As the authors note, Presidents from Herbert Hoover and Franklin Roosevelt forward made quite large impoundments.  Yet Congress, again, seems to have generally viewed their actions as consistent with statutory authority rather than at odds with it.  For example, in a 1950 report, the House Appropriations Committee rejected any power of impoundment that could be used to “thwart[] a major policy of Congress,” yet it observed at the same time that “[i]t is perfectly justifiable and proper for all possible economies to be effected and savings to be made.”  Indeed, the report specifically indicated that appropriated sums should generally be understood as “only a ceiling upon the amount which should be expended for that activity.” 

The rare instances of inter-branch conflict during this period do not support inferring any clear presidential prerogative to defy statutory commands.  As the CRA report notes, President Truman claimed in a signing statement that requiring him to issue a large loan to Spain “would be unconstitutional,” but the basis for this objection was unclear and Truman ultimately disbursed the loan anyway (a fact the report omits).  Similarly, President Kennedy objected to legislative language directing his administration to expend specified sums on a bomber program, but his objection pointedly stopped short of asserting any preclusive presidential authority to defy Congress.  Instead, Kennedy argued only that permissive language would be “more clearly in line with the spirit of the Constitution” (not its letter) and that “a spirit of [inter-branch] comity” “implicit in the Constitution” made “unwise if not impossible any legislative effort to ‘direct’ the Executive on matters within the latter’s jurisdiction.”  Congress ultimately omitted mandatory language from the bill, giving Kennedy no occasion to defy any statutory directive.

In short, although Presidents over time asserted broad authority to impound appropriated funds, this practice did not involve any assertion of constitutional authority to act contrary to congressional directives.  It instead reflected an operative interpretation of the underlying statutes—an understanding that Congress accepted and that Presidents generally exercised in a manner consistent with Congress’s overall policy goals.

President Nixon’s Constitutional Theory—and Congress’s Rejection of It

This implicit understanding finally broke down when President Nixon overturned the apple cart by impounding gigantic sums for domestic programs that Congress supported and he did not.  Unlike his predecessors, President Nixon squarely claimed constitutional authority for his actions.  As he put it in a 1973 press conference, “[t]he constitutional right for the President of the United States to impound funds, and that is not to spend money, when the spending of money would mean either increasing prices [through inflation] or increasing taxes for all the people, that right is absolutely clear.”  

The response was swift, and far from acquiescing to Nixon’s view, Congress forcefully repudiated it.  In 1974, Congress enacted the Impoundment Control Act, which largely forbids impoundments.  As modified by later amendments, this statute precludes the “rescission” (meaning cancellation) of budget authority without new legislation.  It also bans even the “deferral” (meaning postponement) of spending unless the reasons for the delay are programmatic rather than policy-based.  Specifically, it generally allows deferrals only within a single fiscal year and even then 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 of these changes is to make most spending mandatory:  Executive officials cannot cancel or even delay spending for which Congress has appropriated funds unless they are doing so on narrow programmatic grounds and follow specified reporting procedures.  Once this law was in place, a continued practice of presidential impoundment could have reflected a claimed constitutional authority to defy congressional mandates.  Yet no such practice developed.  On the contrary, as Eloise Pasachoff recently observed, “[t]he Act has been generally successful in restricting illegal impoundments.”  In other words, as Josh Chafetz has put it, “presidents have largely adhered to the act’s requirement to report impoundments and . . . have released funds when required to.”

Accordingly, presidential practice in the only context that matters—at the “lowest ebb” of presidential power—supports rather than contradicts the validity of spending mandates.  Even the CRA report does not suggest otherwise:  Despite asserting that the statute “represented an unprecedented break with the Nation’s constitutional history and traditions,” it halts its historical survey with the statute’s enactment and thus identifies no post-enactment practice supporting its theory.

The General Challenge of Faithful Execution

In sum, appropriations statutes are laws that the President must faithfully execute.  When a statute mandates expenditure, as is now often the case, the President must therefore comply, unless the law infringes on some specific resource-independent power of the presidency.  It is true that a significant historical practice of impoundment developed before 1974, but this practice was a practical gloss on the statutes in place at the time, not on the Constitution itself, so Congress was free to abrogate it by altering those statutes, as it did by enacting the Impoundment Control Act.

Renewed claims of impoundment authority, then, would not draw support from longstanding practice.  Instead, they would fit within a more recent, and more troubling, pattern of executive behavior, namely, the recurrent penchant of recent Presidents for self-aggrandizement.  President Obama’s actions might provide the closest analogy:  On the heels of President George W. Bush’s Article II maximalism, President Obama corroded the President’s duty of faithful execution by unlawfully claiming power to suspend enforcement of statutes and programmatically authorize violations.  Since then, executive authority to reshape the law through overt and deliberate nonenforcement has become an article of faith among state and local progressive prosecutors, though not yet a generalized practice in federal administration.  A claimed constitutional authority of impoundment would compound the error in these examples by stretching the same theory beyond law enforcement to government expenditure.

The theory, however, is flawed in both contexts, and extending it to government expenditure would be quite damaging.  Existing statutory and constitutional authorities give contemporary presidents broad powers of initiative in setting national policy.  Because Presidents are dependent on annual appropriations for most of their policy goals, the appropriations process is a place where Congress can push back, provided it can overcome its own internal divisions and reach sufficient agreement to embody its contrary preferences in law.  Further degrading this important check on the modern presidency would be misguided even if it weren’t also unconstitutional.

Zachary S. Price is a professor at UC Law San Francisco (formerly UC Hastings).  His book “Constitutional Symmetry:  Judging in a Divided Republic” is forthcoming with Cambridge University Press this fall.