Zachary Price is a professor of law at UC Hastings Law in San Francisco.
In its decision in Department of Homeland Security v. Regents of the University of California barring repeal of the Deferred Action for Childhood Arrivals program, known as DACA, the Supreme Court reached an attractive result through flawed legal reasoning. The decision may carry implications that progressives will regret, but it is hard to tell because Chief Justice John Roberts’ opinion seems deliberately designed for one day and case only.
Within the context of other recent decisions, furthermore, the decision appears to reflect an impulse I have called “symmetric constitutionalism”—deliberately spreading case outcomes and doctrinal benefits across major partisan divides so as to preserve judicial legitimacy and shared legal commitments. Yet by engaging in essentially ad hoc reasoning to reach such a politically significant result, the court in Regents violated a central requirement that such symmetry should entail—namely, the imperative to anchor decisions that disappoint one faction today in generalizable principles that may disappoint their rivals tomorrow.
Without rehearsing all the details here, DACA is a program that invited large numbers of unauthorized immigrants who entered the United States as young children to apply for renewable two-year promises of nondeportation and obtain work authorization and other benefits. (I addressed DACA’s legality in this article on “Enforcement Discretion and Executive Duty”; the reasons why such policies should be revocable in this article on “Reliance on Nonenforcement”; and the reasons why the government should have won this case in this earlier blog post.)
The essential problem with DACA, as I see it, is that it goes beyond simply directing agency officials to assign low priority to some cases relative to others and attempts to provide a form of durable, prospective protection to a large category of immigrants who were concededly deportable by statute. In my view, such relief requires specific statutory authorization or delegated interpretive authority; it is not a proper implication of mere organic enforcement discretion, as DHS claimed when it first adopted the program.
The key reason for DACA’s revocability is that such policies’ unreliability places an essential constraint on executive authority. Enforcement discretion is generally a power to prioritize some cases over others and turn a blind eye to particular violations; it is not a power to alter the law itself or prospectively authorize violations. Yet executive officials could convert enforcement discretion into a power to change law if inviting reliance on promised forbearance were sufficient to freeze a permissive policy in place.
To avoid these problems, courts have previously shown extraordinary reluctance to recognize any form of estoppel based on government assurances. They have likewise repeatedly rejected claims of reliance based on promised forbearance. The main exception has been in cases quite different from Regents—cases in which officials were not merely exercising enforcement discretion, but instead employed delegated interpretive authority in a manner that reasonably led regulated parties to believe their conduct was lawful.
The majority in Regents appeared oblivious to this pattern of decisions and the danger of executive self-aggrandizement to which it responds.
In the majority’s view, although “DHS may rescind DACA,” its effort to do so here was “arbitrary and capricious” simply because it was insufficiently explained. Though bound by the attorney general’s determination that DACA was unlawful, the secretary of DHS, in the court’s view, should have considered in greater depth whether to preserve enforcement forbearance while cancelling other benefits associated with that forbearance.
In addition, as a separate and apparently independent defect, the majority faulted the secretary for inadequately considering the reliance interests of DACA beneficiaries. Instead of the terse statement it issued upon initially rescinding DACA, the secretary should have “asses[ed] whether there were reliance interests, determine[d] whether they were significant, and weigh[ed] any such interests against competing policy concerns.” (The secretary did in fact make clear later that she considered it “critically important for [the government] to project a message that leaves no doubt regarding the clear, consistent, transparent enforcement of the immigration laws against all classes and categories of aliens,” but the court disregarded this statement as a “post hoc rationalization.”)
Regents’ basic logic, if generalized across other cases, could have dramatic implications. Consider some examples. Right now, a multibillion-dollar marijuana industry operates openly in multiple states, in flagrant disregard of federal criminal narcotics laws. This industry arose in response to an announced Obama administration policy assigning low priority to enforcing federal criminal laws against state-compliant marijuana consumers and businesses.
Though it has since revoked this policy, the Trump administration has not meaningfully resumed enforcement. The Regents court’s reasoning implies that it actually could not do so even if it wished to—even though the Obama-era policy made crystal clear that the government was not authorizing what Congress prohibited and that enforcement remained possible. If any future administration wishes to reassert the primacy of federal marijuana laws over contrary state policies, Regents implies that it will need to publish a self-flagellating litany of all the multiple ways in which regulated parties have relied on the prior invitation to law-breaking. Even then, the government would face litigation risk that a court would consider its deliberation inadequate.
That result may strike you as sensible with respect to marijuana or DACA. But the game can be played the other way, too. In multiple lower court decisions, courts have invalidated Trump administration policies that purported to undo environmental regulations, among other things, with the stroke of an administrative pen. Courts relied in these decisions on the principle established in Accardi v. Shaughnessy that, because regulations bind agencies as well as those they regulate, it takes a new notice-and-comment rule to undo a past notice-and-comment rule. Under Regents, it appears these courts had it backwards: Instead of overturning lawless deregulatory actions, courts should have frozen them in place until the agency took adequate account of any resulting reliance interests.
Still more broadly, to the extent Regents’ conclusions are generalizable, the Trump administration might use its remaining time in office to adopt permissive enforcement polices across any number of areas, from gun control to labor regulation, the environment and public corruption. To the extent such policies invite reliance, a future administration will be foreclosed from reversing them unless its self-explanation meets the Regents court’s meticulous standards.
In all likelihood, though, Regents will not have these effects. When one dives into the details, the majority’s reasoning becomes so narrow and self-contradictory as to preclude confident generalization beyond the example at hand. The court expressly declined to consider DACA’s legality—yet faulted the secretary for failing to consider options she believed were unlawful. It held that the secretary was bound by the attorney general’s legal conclusions—yet faulted her for failing to independently parse the reasoning in a court decision. It emphasized the importance of considering reliance interests—yet never addressed the troubling incentives that protecting reliance could create for the executive. Overall, the court repeatedly emphasized the particular features of the government’s shambolic rescission process and parsed the particular words of DHS’ explanation, seemingly advertising the decision’s narrow, fact-specific character.
Given all this muddled narrowness, it is hard to escape the conclusion that the court, whether self-consciously or not, took advantage of the Trump administration’s characteristically erratic decision-making process to benefit sympathetic immigrants and freeze in place a policy the court knows to be both quite popular and quite polarizing. In effect, it reached a result that Congress should have reached on its own—but in a manner that spares Congress any need to strike compromises and disappoint some constituents to please others.
Taking into account other cases across recent terms, Regents also seems to reflect an impulse among at least some justices to advertise their good faith, and perhaps lower the political temperature, by distributing wins and losses across partisan lines. Regents and the Title VII decision on Monday are progressive “wins,” for example; last term, Rucho v. Common Cause and American Legion v. American Humanist Association, on partisan gerrymandering and the establishment clause, were major progressive “losses.” (We will soon see what happens with abortion and religious freedom.)
But the flaws in Regents’ reasoning show that there are both good and bad ways to go about pursuing this sort of symmetry. Dispensing wins and losses based on narrow or one-sided reasoning feeds a perception that the court is simply channeling its members’ preferences. By contrast, anchoring those decisions in generalizable principles—ideally principles with potential benefits and costs straddling major ideological divides—may reinforce shared commitments. At the least, doing so can signal the court’s support for what Justice Stephen Breyer has called “the law’s demand for even-handedness.” In crafting such general principles, moreover, courts should appreciate their limited role in our constitutional system, which relies principally on the political process, and not judicial decisions, to enable self-government and liberty under law.
The court could have done all that here by allowing DACA’s rescission while making clear that any comparable policies from this administration will be equally revocable in the future. Instead, it chose to save the political process from itself by freezing a justifiably popular program in place with transparently ungeneralizable reasoning. The trouble is that here, as in most contexts, it is not the court’s job to spare us the hard work of political compromise. If carried too far, that impulse risks saving democracy by killing it.
CLICK HERE FOR FULL VERSION OF THIS STORY