This week a federal court in San Francisco enjoined the Trump Administration from terminating the Deferred Action for Childhood Arrivals (DACA) program, which has granted a form of immigration relief to some 700,000 undocumented immigrants who entered the United States without authorization as young children. This decision, however attractive as a matter of policy, strikes me as mistaken under the law. It warrants swift reversal by higher courts.
I happen to agree with the administration that DACA was illegal to begin with—a conclusion the district court rejected. But that is not the problem (or at least the only problem) with the court’s decision. The argument I and others made against DACA was that, by granting a large category of undocumented immigrants an unusually determinate prospective guarantee of non-deportation, it effectively changed statutory law without an adequate delegation of authority from Congress.
The legal justification (such as it was) for DACA was that it was instead just an exercise of enforcement discretion, no different from a beat cop turning a blind eye to a speeding car. It follows from this asserted legal justification that the deferrals must also be revocable. That was what made them (at least arguably) an exercise of enforcement discretion rather than a change in law. If the promises had not been revocable, then the deferrals clearly would have altered substantive immigration law without any delegated authority from Congress to do so.
I have discussed this principle (and related reliance questions) at length in a recent article. But don’t take it from me. OLC’s opinion approving the related DAPA program (though not opining directly on DACA itself) emphasized that “deferred action confers no lawful immigration status, provides no path to lawful permanent residence or citizenship, and is revocable at any time in the agency’s discretion.” Indeed, curiously, the district court itself recognized this aspect of the program. In defending DACA’s legality, it observed: “Even for enrollees approved under the program, DHS expressly retained the authority to terminate their deferred action at any time, in the agency’s discretion. DACA provided no guarantee against removal.”
Why then did the court say the administration cannot terminate the program? The court reasoned that, in light of immigrants’ and third parties’ reliance on grants of deferred action, the government failed to adequately justify its decision to terminate the program. The government’s apparent rationale was the program’s illegality; the district court concluded DACA was lawful; ergo, the program’s termination was arbitrary and capricious.
This logic strikes me as flawed in this context. Insofar as DACA was simply an exercise of enforcement discretion, any explanatory burden with respect to its reversal must be minimal, and the government’s stated legal misgivings (and associated litigation risks) seem more than adequate. After all, federal courts had enjoined a closely related program (DAPA).
To the extent legal misgivings and litigation risk are inadequate justification, moreover, the government could simply announce that it has chosen to allocate its limited enforcement resources differently so as to better advance overall immigration compliance. Requiring any more justification than that would effectively eliminate the central formal limitation on nonenforcement policies—the principle, acknowledged by OLC and even the district court itself, that any nonenforcement assurance adopted as a matter of grace must also be revocable at will.
It is true that in some cases, such as the Encino Motor Cars decision cited in the district court’s opinion, the Supreme Court has required express consideration of reliance interests when an agency departs from a prior policy. It is also true that courts sometimes review legal interpretations embodied in enforcement policies. But these lines of cases are distinguishable. They involve situations where agencies followed or adopted some interpretation of the substantive conduct-rules enforced by the agency. Here, in contrast, there is no credible argument of which I am aware that the immigrants in question are not subject to potential deportation under applicable statutes. That is why the government could grant relief only through an exercise of enforcement discretion—but that is also why, once again, the government’s choice to do so must remain revocable at will.
Furthermore, to the extent reliance factors in the analysis, the district court’s reliance theory here was startlingly broad. Under the court’s reasoning, DACA beneficiaries could reasonably rely not only on their own individual deferrals (which were themselves both time-limited and revocable), but also on the indefinite continued operation of the program itself. The new administration, after all, is not reneging on existing grants of deferred action; it even temporarily continued renewing them. In effect, then, under the district court’s reasoning, an expressly time-limited and revocable reprieve from enforcement may become, by judicial fiat, a permanent entitlement.
Trump’s opponents should care about these principles. In administrative law, what goes around comes around, and this administration has adopted any number of permissive, deregulatory policies and enforcement practices. You can bet that if this decision is allowed to stand, regulated parties will invoke it in the future to prevent a new administration from revoking those policies. At the same time, the core principle that the district court neglected in this case—that agencies cannot change substantive law through mere nonenforcement assurances—has already proven to be an important constraint on this administration: courts have invoked it repeatedly to invalidate unlawful repeals of Obama-era substantive regulations.
The dreamers’ situation in this country is tragic. Like most Americans, I hope very much that Congress provides them with relief. But in the meantime courts have authority to overturn the administration’s policies only insofar as they are unlawful. At best, flawed decisions like this one establish precedents that will thwart the next administration as much as the current one. At worst, by making judges appear political, they squander the legal credibility on which courts’ capacity to check the executive ultimately depends. For those reasons, this is one case the administration deserves to win on appeal.