As readers are no doubt aware, the Trump administration has attempted to “pause” various spending programs for an initial period at the start of Trump’s new term.  Two first-day executive orders mandated such pauses for certain grants and foreign aid pending a review for compliance with administration policies.  On January 27, the Office of Management and Budget then ordered a government-wide pause in certain spending.  A district judge swiftly stayed the OMB order, and the administration rescinded it the next day.  Now two district courts have issued temporary restraining orders barring its implementation.

This litigation raises a host of issues, including threshold questions about standing, mootness, finality, and whether litigants other than the Comptroller General can challenge violations of the Impoundment Control Act (ICA).  On the merits, the pauses’ legality likely depends on the details of the various affected grants and programs and also on whether the funds in question have already been obligated or only allocated internally to certain purposes.  In some instances, the administration may have authority to make a “programmatic delay” because operational considerations justify a brief pause while the administration sorts out its approach.  In other instances, the delays may amount to a “deferral” based on policy objections to the spending itself—something the ICA prohibits.  (For more on these concepts, see this explainer I wrote for Lawfare.)

In this post, I want to highlight an important question lurking in the background that implicates some broader issues:  Now that the Supreme Court has rejected Chevron deference in Loper Bright Enterprises v. Raimondo, what weight should established government practices carry in statutory construction?

The question is relevant here because the ICA does not expressly contemplate “programmatic delays” and indeed arguably restricts them.

According to the Government Accountability Office, the congressional agency that polices appropriations compliance, a programmatic delay is a delay “in which operational factors unavoidably impede the obligation of budget authority, notwithstanding the agency’s reasonable and good faith efforts to implement the program.”  As Eloise Pasachoff has pointed out, a delay for these reasons would seem to satisfy the statutory definition of a “deferral.”  And while the ICA allows deferrals, they must be made for non-policy reasons (or based on specific statutory authority), they must not prevent prudent obligation of the funds before the end of the fiscal year, and the administration must report them to Congress.  GAO, however, views programmatic delays as falling outside the ICA altogether, meaning that they don’t need to be reported, and in fact no President appears to have reported a deferral since fiscal year 2000.

To the extent courts reach the question of such delays’ validity—and again, for any number of reasons, they may not—they will need to decide what to make of this practice.  Before Loper Bright, one might have thought courts should defer to GAO’s views, given its expertise regarding appropriations law, or perhaps in some instances to the agencies receiving the appropriations in question.  (For thoughtful discussion of these possibilities, see these articles by Gillian Metzger and Matthew Lawrence.)  But Loper Bright seems to foreclose these forms of deference.

For its part, the administration may argue that its views should get deference because of the President’s responsibility for faithful execution of federal laws.  On the last day of the first Trump administration in 2021, the OMB Director and General Counsel (both of whom have been renominated for those roles) sent a letter to Congress staking out several interpretive positions, including authority to make programmatic delays.  And as part of its overall pitch, the letter argued:  “The President must faithfully execute all the laws, and where those laws provide discretion or are ambiguous as to implementation, the Constitution grants the President exclusive authority to determine the best and most efficient manner in which to implement such laws.”  Those are some of the same arguments that justified Chevron—and that the Supreme Court rejected last June in Loper Bright.

So if deference is out the window, should courts also disregard government practices in applying statutes?  At least when it comes to important framework statutes like the ICA, I believe the answer is no.

As the ICA example suggests, applying conventional textualism to some statutes could be quite disruptive.  The ICA is one of several statutes that give effect to Congress’s constitutional authority over government funding.  For that very reason, interpreting these statutes implicates the balance of power between Congress and the executive branch, often in areas of acute policy conflict.  Much as courts should be wary of disrupting constitutional understandings reflected in entrenched inter-branch practice, so too should they hesitate to disrupt settled understandings of statutes like these that have structured the two branches’ relations over time.

Loper Bright does not foreclose accounting for such practices.  On the contrary, in its discussion of pre-Chevron decisions, the Court observed (quoting earlier cases) that “the longstanding practice of the government—like any other interpretive aid—can inform a court’s determination of what the law is.”  Similarly, one of the Court’s recent “major question” cases pointed to “the longstanding practice of [the agency] in implementing the relevant statutory authorities” as support for upholding the action at issue.

Government practice, then, should properly inform statutory interpretation in some contexts.  At the same time, however, accepting settled practices should not mean accepting anything the government does.  It should mean asking whether some new action genuinely fits within a settled pattern of practice that matches some coherent view of the law.

In the immediate example of programmatic delays, accepting settled practice would mean investigating what implicit understanding has in fact governed executive behavior and congressional responses to it.  Courts should not necessarily embrace the precise legal formulations that either OMB or GAO have used.  (Those formulations may carry some problems, as Mark Thomas observed on this blog.)  Nor should courts necessarily accept every past action—let alone every past legal assertion—as valid.  Courts, rather, should aim to discern the overall pattern of practice and the implicit theory underlying it.  To the extent this theory is credible, they should then hold the executive branch to its limits, even as they allow actions that accord with accepted past conduct.

Beyond the pause litigation, this problem may come up in other cases too.  Many important statutes governing executive behavior have been rarely litigated in the past.  If Trump administration policies now spark legal challenges, courts might fruitfully hold both sides to the practical understandings that the political branches worked out in calmer days gone by.

Zachary S. Price is a professor at UC Law San Francisco (formerly UC Hastings).  He is the author of Constitutional Symmetry:  Judging in a Divided Republic and Funding Restrictions and Separation of Powers.