Can a single federal district court judge issue an injunction binding in every state? And if so, when should they do so? That question has been on the minds of many watching the whiplash-inducing orders from judges these last few years. In 2015, a district judge in Texas issued an injunction barring the federal government from enforcing an executive order granting temporary reprieve and work authority to immigrant parents of persons lawfully in the United States (either citizens or permanent residents). The plaintiff was the State of Texas – the only state found to have standing – but the injunction encompassed the entirety of the United States. In 2016, a district judge in Texas issued an injunction barring the federal government from enforcing a Department of Education policy requiring that public schools provide facilities (such as restrooms) that match their students’ gender identity. The plaintiffs were sixteen states that did not wish to comply with the policy, but the injunction encompassed the entirety of the United States. In case you think that everything is bigger in Texas, in 2017, a federal district judge in Illinois issued an injunction barring the federal government from withholding federal funding to “sanctuary jurisdictions,” states or local governments that refused to cooperate with immigration enforcement. And in 2017 several district court judges enjoined the federal government from enforcing various versions of an executive order barring nationals from several predominantly Muslim countries from entering the United States.
One of the coolest things about procedure is how procedural questions destabilize our political preferences. As a matter of principle, if you agree that judges have the power to issue one kind of national injunction, then you must agree that they have the power to issue the other. As a political matter, you might like the injunctions against the Obama administration and not those against the Trump administration, or vice versa. The controversy over this exercise of judicial power was, until recently, mostly academic (in the sense that mostly academics cared about it), but a bill recently was introduced in Congress barring federal judges from issuing “an order that purports to restrain the enforcement against a non-party of any statute, regulation, order, or similar authority” except where the case has been certified as a national class action. The impetus for the bill is probably political (its sponsor is Republican Representative Goodlatte), but that does not answer the question: Is this a good bill? Should we limit the reach of a federal district court to the parties before it or in some other way that would bar these types of injunctions?
High-profile and well-reasoned articles published last year by Samuel Bray, Michael Morley, and a more recent piece by Howard Wasserman argue that such injunctions overstep the bounds of what a federal court can and should do. Amanda Frost has now written an excellent defense of nationwide injunctions in response. Importantly, Frost situates the issue in larger debates over the role of the federal courts. Her claim is that your priors about what courts are for will determine to a significant extent your reaction to a national injunction. If the purpose of the courts is primarily dispute resolution, an injunction should be limited to the parties. If a substantial purpose of the courts is law declaration, a national injunction could be an important tool in the court’s toolbox, especially in the face of a recalcitrant governmental defendant. The possibility of a recalcitrant government defendant is especially important because, as Frost points out, national injunctions have surfaced in the years when the executive has exercised more unilateral power than any time in recent memory. Indeed, it is impossible to understand the phenomenon of national injunctions without understanding the recent relationship among the courts, Congress, and the executive branch. This is the powerful and important contribution of Frost’s article: it forces us to confront assumptions about both the structure of government and the court’s place within that structure.
Frost’s diagnosis of why we are here is the most valuable contribution of her piece, but her constitutional arguments are important, as well. First, Frost responds to the argument that the judicial equity power historically was limited to the parties before the court and that this limit should remain today. To the extent you think the judicial power today should be defined by its parameters during the Founding Era, Frost points to the “bill of peace,” which applied to non-parties. Other historical work supports her. Although non-party preclusion was rare, it was not unknown in the early common law. (If you are a proceduralist reading this and have not read Robert Bone’s article, Rethinking the “Day in Court” Ideal and Nonparty Preclusion, stop and read that instead.) Indeed, the very nature of property rights is that they are in rem – “they serve not only to bind successors in interest but the whole world.” Contrary evidence is always the problem with common law arguments steeped in history, especially procedural history; such arguments are rarely (if ever) definitive. The famous Talmudic saying applies to instrumental legal history: turn it and turn it again, for all is in it.
Second, Frost responds to the view that standing doctrine requires that the scope of the federal court’s power to order injunctions be limited to the parties, because the parties frame the “case or controversy” under Article III. Frost’s most powerful argument is that just as courts can strike down statutes as to all, they can strike down executive orders. Frost also points out that the equitable relief necessary to provide complete relief to the plaintiff must include other parties in many cases, such as a desegregation order requiring the defendant to allow all nonwhite students to attend a school. It may be that the problem raised by the plaintiff is so intertwined with others similarly situated that relief in the plaintiff’s case requires relief to others. The argument against allowing such intertwined interests to be heard reminds me of Lon Fuller’s objection to courts deciding “polycentric” disputes and brings us back to the basic point that one’s assumptions about what courts are for is at the heart of this question. That debate has been going on since the founding of the republic. To put it in terms of twentieth century scholarship, are you more inclined to agree with Fuller or Chayes?
Frost also responds to the argument that national injunctions should be available only by certifying, when appropriate, a nationwide class under Rule 23(b)(2). I imagine that the underlying appeal of this argument is that if the injunction is subject to Rule 23’s requirements (especially that the named plaintiff adequately represent the entire class), then we can rely on the rule to cabin the exercise of judicial power and this feels less overreaching. I use the word “feels” for a reason. As a practical matter the court would end up in the same place. Frost points out that if an injunction is permitted pursuant to 23(b)(2) then it must be constitutional as well; if a court can do something through a rule (which the courts themselves promulgate), why would it be unconstitutional to do this same thing in the absence of the rule? It cannot be that the rule dictates what Article III allows. The best argument for insisting on certification of a Rule 23(b)(2) class action is that the rule makes absent class members into parties, but I am not convinced for three reasons. First, the “representative litigation” concept that provides the constitutional foundation for class actions is a legal fiction in which the adequacy of the representative is deemed to be sufficient, as a constitutional matter, to bind all others in the class. Legal fictions should be avoided whenever possible, and courts should be direct about what they are doing. The legal fiction of binding by representation ought not be sufficient if the exercise of judicial power is illegitimate. (The ink spilled on the principal-agent problem in class actions is evidence of the discomfort with this fiction.) Second, we have a history of non-party preclusion that supports a broader conception of who can be bound by a lawsuit. Third, absent class members aren’t parties for many purposes, so one needs a policy reason to say they are parties for this purpose.
This brings us to the real issue: policy. Should one judge, potentially an outlier judge with radical ideas, make policy for the nation? If so, why not make them go through the process of certifying a national class action and ascertaining that there are no conflicts that require rethinking the idea? Frost doesn’t seem to be against requiring judges to certify injunctive classes in the run of cases, but argues that injunctive class actions are sometimes very difficult to certify, often requiring hearings that are tantamount to trials. Indeed, an assault on injunctive classes may be coming, and those who want to be able to bind the elected branches should be worried about allowing the class action device to tie judicial hands in the face of extreme government wrongdoing. That said, it is not always hard to certify injunctive classes. Classes can be certified provisionally (in fact, all classes are to some extent certified provisionally until a final judgment is entered), and more courts might use the power to certify a provisional class in order to issue a national injunction in a timely manner if that hurdle were required. For example, in Mrs. L v. Immigration and Customs Enforcement, the court certified a 23(b)(2) class in three-and-a-half months — plaintiffs filed their class certification motion on March 9, 2018, and the judge certified the class on June 26 — very fast for a class certification motion. But we cannot expect every class to proceed with such haste. This was a special case, involving very young children being separated from their parents and suffering likely permanent, debilitating psychological and potentially physical injuries. . . . Upon reflection, three-and-a-half months was much too slow.
The issue, then, is whena national injunction might be appropriate. Frost is closer to her critics than I anticipated because she does not seem to think national injunctions should be often used. She explains the risks of forum shopping and conflicting injunctions, concerns about infringing on the rights of non-parties, and the need to let the law percolate. And she urges courts to be sensitive to these risks as well as the costs and benefits in the specific fact situation in which the need for an injunction arises. In the end, she seems more inclined to believe in the integrity and capacity of judges than her critics. Reasonable minds can differ on the question of judicial competence to make these difficult decisions in a high-stakes environment and on the question of whether the costs outweigh the benefits. Before making up your mind, I highly recommend reading Frost’s article both for its summary of the debate and for its compelling arguments.