For the Symposium on Administrative Reform of Immigration Law
There is much to cheer in OLC’s careful opinion, but nonetheless cause for concern about the precedent the administration has set.
Let me say at the outset that as a policy matter I support the President’s action and find Congress’s inability to provide immigration relief incomprehensible. I view this issue, however, as one instance of a recurrent separation-of-powers problem that arises across many areas of regulation: What authority do Presidents have not to enforce laws for policy reasons?
From that perspective, the most important thing the OLC opinion does is to cabin this precedent to the immigration context—and the most significant risk it poses is that this effort will not prove successful.
For those who haven’t yet read it, OLC’s opinion approves a program making some four million undocumented immigrants who are parents of U.S. citizens or legal permanent residents eligible for “deferred action.” Deferred action is an exercise of prosecutorial discretion in which immigration authorities choose not to seek removal of a removable immigrant. Though the government retains the right to reconsider, deferred action typically extends for a specified period—in this case renewable periods of three years. Under immigration regulations, it entails eligibility for work authorization.
Although this practice arose as a means of handling particularly sympathetic individual cases, previous programs have extended deferred action to large groups. The controversial Deferred Action for Childhood Arrivals (DACA) program adopted in 2012 enabled certain undocumented immigrants who arrived in the United States as young children to apply for deferred action. A footnote in OLC’s opinion describes previous “preliminary” oral advice approving DACA, but the opinion does not otherwise address DACA’s validity. In fact, it specifically rejects a proposed program to extend deferred action to parents of DACA beneficiaries.
OLC grounds its conclusions in a thoughtful analysis of prosecutorial discretion that gets many important things right.
In my view, which I have elaborated elsewhere, prosecutorial discretion is, at its constitutional core, an authority to make case-specific exceptions from enforcement of general laws. In many areas of modern regulation, broader exercises of discretion become necessary because executive agencies must enforce broad mandates with limited budgets. Even in such areas of regulation, however, the constitutional architecture supports an important background norm that executive officials still must seek to effectuate statutory policies. Enforcement discretion is an authority to pursue one case rather than another because you lack the time and money to do both. It is not an authority to change underlying legal obligations.
One important way in which the executive branch gives effect to this basic norm is by framing its enforcement choices as priorities rather than guarantees and thus avoiding, to the greatest degree possible, either definite prospective assurances of non-enforcement or rule-like categories to distinguish between violations.
I read OLC to be in basic agreement with all of this. Indeed, I am aware of no other legal opinion addressing the fundamentals of enforcement discretion more clearly and thoughtfully. The opinion correctly recognizes that although line-drawing with respect to enforcement practices is hard—it is often “more a matter of mindset than any sort of bright-line rule,” as I have argued—lines nevertheless must be drawn.
The opinion also recognizes, with admirable candor, that the immigration practice of deferred action, as it has come to be exercised, is hard to square with these norms.
Of course, it is entirely conventional for agencies to adopt internal guidelines that prioritize certain cases over others. Case-by-case deferred action (what the opinion calls “ad hoc” deferred action) also appears fairly conventional. General deferred action programs, however, go beyond either conventional agency priority-setting or ad hoc deferred action by deeming broad categories of immigrants presumptively eligible for a prospective promise of non-enforcement.
In other settings—tax or environmental enforcement, say, or regulation of workplace safety—it would be unusual to invite a significant subset of regulated parties engaged in a continuing violation to apply for a promise (even a formally revocable one) that the law will not be applied to them. OLC is therefore correct to recognize that “categorical” uses of deferred action “may raise particular concerns about whether immigration officials have undertaken to substantively change the statutory removal system rather than simply adapting its application to individual circumstances.”
Contrary to what some argued before the OLC opinion, categorical deferred action cannot be justified as just another form of conventional agency priority-setting. A conventional agency priority is just internal agency guidance and avoids assuring non-enforcement to any specific party. Deferred action, although formally revocable, aims in practice to provide security from removal to its beneficiaries (and to enable certain other immigration benefits).
Nor is it sufficient justification that the Department of Homeland Security is spending its full budget on enforcement, as some also suggested. The OLC opinion implicitly rejects this theory by drawing limits on deferred action well short of what the government has adequate resources to enforce. In many politically volatile areas of law, only a tiny fraction of violations are realistically subject to enforcement. The signals the agency sends to regulated parties are often at least as important as its actual enforcement practice in assessing how faithful the agency is to its enforcement mission.
The OLC opinion therefore correctly seeks justification for this unusual program in considerations unique to immigration and immigration law. This framing of the issue is crucial, and the debate going forward should center on the immigration-specific arguments that OLC has put forth.
Here, the devil is in the details. Given the importance of the general background norm of agency fidelity to statutory policies, one might debate some inferences the opinion draws in its effort to link the executive action here to statutory objectives. Are past deferred action programs for immigrants with immediate visa-eligibility comparable to the deferred action here for parents with eventual, prospective visa-eligibility? What should we make of the REAL ID Act’s endorsement of deferred action as a basis for a qualifying driver’s license? Is it fair to read statutory provisions providing specific forms of immigration relief for family members as supporting broader use of deferred action to maintain family unity?
(It is also worth noting that the opinion’s careful effort to root deferred action for parents of citizens and LPRs in specific statutory policies concededly does not apply to DACA itself. DACA instead depends on an analogy to past humanitarian exercises of group-based deferred action, even though, as the opinion puts it, those past actions typically involved more “particularized and acute” concerns.)
More broadly, the opinion raises the question of what weight we should give to past executive practice. Past practice is an inevitable—and I believe appropriate—source of legal authority within the executive branch, particularly when there is reason to presume congressional acquiescence. But using executive practice to justify further executive practice carries the familiar danger of a one-way ratchet in executive-branch legal interpretation. For that reason and others, the weight of past executive practice in legal interpretation outside the executive branch remains an unsettled question.
What sort of precedent will OLC’s opinion itself set? One question is how courts will react. I view the judicial role in policing executive enforcement practices as a hard issue, and I hope soon to post a draft paper addressing the issue as a general matter. Suffice it to say that any litigation may well turn as much on justiciability as on the merits of the executive branch’s legal position.
The bigger question is what effect this example will have on the practice of future Presidents. Ironically, although current political alignments mask this reality, in the long run Democrats may have more to fear than Republicans from erosion of norms against policy-based non-enforcement. Amid our deeply polarized politics, Democrats are the party more committed in principle to the value of regulation in improving social welfare and restraining undue private power. Non-enforcement is a fundamentally deregulatory power: it is a power to strip force from statutory requirements though executive inaction. Republican Presidents have used it as such in the past. Democrats should worry about how they might use it in the future.
OLC has done a commendable job framing an expansive exercise of non-enforcement power as a narrow precedent. As every common lawyer knows, however, narrow precedents have a way of not remaining narrow. President Obama and his successors should ensure that this one does.
Zachary Price is a Visiting Assistant Professor at the University of California Hastings College of the Law. You can reach him by email at pricez@uchastings.edu.