Trump v. CASA and the future of the universal injunction

This is part of SCOTUSblog’s term in review series, in which scholars analyze some of the most significant cases of the 2024-25 Supreme Court term.

The best that can be said for Trump v. CASA is that it could have been far worse. Its methodology and conclusions are myopic and wrong, and it is especially unwelcome during a period in which each passing day seems to bring new incursions by the executive branch upon individual rights, the separation of powers, federalism, and the rule of law. The court held that federal courts may not give universal injunctions, which are orders that block the application of a law or an executive branch action to anyone who might be harmed by it, not just its application to the plaintiffs. But despite the court’s seemingly categorical rejection of the universal injunction, the ironic possibility exists that the decision will turn out to be, as Justice Samuel Alito put it, essentially “academic.”  

The court has announced that one narrowly defined category of broad relief is no longer on the table. But the core of the fight will now shift to a new set of questions: How broadly will the malleable standard of “complete relief” sweep? Can plaintiffs reliably use the nationwide injunctive class action – including the pre-certification injunction protecting a putative class – to achieve the same result as the universal injunction? And perhaps most critically, will the court in a future case preserve the universal remedies available under the Administrative Procedure Act that this decision leaves untouched? The answers to these questions, which are explored in greater depth throughout this piece, will determine the extent to which universal relief will persist even absent the universal injunction. 

The decision’s bottom line is easily summarized. Relying on a cherry-picked recounting of equity’s historical constraints, the court held that the Judiciary Act of 1789 does not authorize federal courts to give universal injunctions. Thus, the court granted the Trump administration’s request for partial stays of the lower courts’ universal injunctions, “but only to the extent that the injunctions are broader than necessary to provide complete relief to each plaintiff with standing to sue.”  Complete relief to the plaintiffs, the court cautioned, is not “synonymous with” universal relief. Yet the court simultaneously recognized that giving complete relief to the plaintiffs may sometimes require a court to give relief that benefits non-plaintiffs – for example, when a court orders that a nuisance cease or that a racially gerrymandered voting map be re-drawn. Notably, the court did not itself actually narrow the injunctions; it merely announced its “complete relief” principle and left the work of reconsidering the injunctions’ scope to the lower courts.  

On remand, the states challenging the executive order, and quite possibly CASA and ASAP, the immigrants’ rights groups also challenging the order, may well be able to demonstrate that they are entitled to universal relief.  The states have argued that because children and families freely move across state lines, a blanket bar on the birthright citizenship order is the only workable solution that will give them complete relief. And the associations have contended that an injunction limited to CASA and ASAP members would not provide them with complete relief: A limited injunction would require parents to disclose their membership in CASA and ASAP to establish their newborn’s citizenship, which raises First Amendment problems, while imposing unworkable administrative burdens on hospitals, government agencies, and families. (Although Justice Clarence Thomas, in a concurring opinion, asserts that the majority’s opinion “readily dispatches” the groups’ contention that they need a universal injunction, the majority does not actually address this question.) Litigants will also seek certification of nationwide or state-wide injunctive classes to shield those vulnerable to the Trump administration’s illegal birthright citizenship order. Consequently, the lower courts on remand may determine that they can and should retain the existing universal injunctions against enforcement of the order (because such relief is necessary to provide complete relief to the states or to these associations) or issue new ones (because newly crafted class actions demand it). And if and when the Trump administration issues guidance on how it plans to implement the order (as it is now free to do), litigation under the APA – a law that allows those aggrieved by unlawful agency action to attack it – will presumably also swiftly ensue to challenge those agency directives. 

As even this brief description reveals, several carve-outs from the court’s decision potentially cabin its systemic damage. Most importantly, the ruling is a statutory holding, not a constitutional one: “Our decision rests solely on the statutory authority that federal courts possess under the Judiciary Act of 1789. We express no view on the Government’s argument that Article III forecloses universal relief.” Nor did the court hold that Article III standing doctrine has any relevance to the question of universal relief, as the government had urged.  

It would have been a serious error, as I have argued in my scholarship and in an amicus brief, for the court to have read into Article III an invented constitutional rule that injunctions may only protect the plaintiffs. Happily, the court did not go down that path. Had the court grounded its ruling in Article III rather than in the Judiciary Act, the decision would have cast constitutional doubt on the scope of remedies that are today available under the APA (and related laws, such as the Hobbs Act). The APA is a landmark statute that has long been used by litigants to seek universal remedies against unlawful executive branch regulations – for example, orders that decree that a federal regulation must be paused or vacated across the board, for everyone. Indeed, many of the most important cases of the past decade that have sought universal remedies against the federal executive branch have been brought under the auspices of the APA or related statutes. (Claims that do not rest on the APA or its sister statutes are generally either those against the president directly or the relatively rare “ultra vires” cases involving agency action that contravenes a specific statutory prohibition and that can’t otherwise be meaningfully reviewed –  a kind of case that Justice Brett Kavanaugh has called a sort of “Hail Mary pass” that “in court as in football … rarely succeeds”.) Crucially, however, the court expressly reserved the question whether the APA authorizes universal vacatur of agency action: “Nothing we say today resolves the distinct question whether the Administrative Procedure Act authorizes federal courts to vacate federal agency action.” When the court eventually confronts this issue, the correct outcome would be for it to sustain the long-recognized authority of Congress to empower federal courts to vacate and stay agency action universally. Until then, the court has at least limited the immediate disruptive systemic consequences of its truncation of federal equitable authority elsewhere by preserving APA remedies, including preliminary APA remedies, intact. Justice Kavanaugh expressly recognized the latter point in his concurring opinion, noting that “in cases under the” APA, “plaintiffs may ask a court to preliminarily ‘set aside’ a new agency rule.”

Another important boundary to Friday’s decision is that the court does not revisit the federal courts’ power to issue preliminary injunctions that protect putative classes – that is, classes that are not yet certified. (Indeed, the Supreme Court itself issued such an order roughly six weeks ago.) The government and other critics of nationwide injunctions have long urged that a court may not give injunctive relief to absent parties unless a case is certified as a class action, because only certified classes can result in judgments that bind the whole class. As I have elsewhere explained, however, that argument ignores the long history of pre-1966 representative suits in equity in which “plaintiffs did not have to do anything more than allege in their bills of complaint that they were suing on behalf of many other persons, on a question of common interest to all of them, in order to obtain preliminary injunctions shielding all those absent parties from the law’s enforcement.” As that tradition reveals, “there’s nothing even new, let alone illegitimate, about a court giving injunctive relief that shields members of a putative class.” The court – thankfully – nowhere endorsed the contention that class certification must precede preliminary class-wide injunctive relief. Thus, and though the overlap may not be total, in many cases that would have produced universal injunctions, as Justice Brett Kavanaugh wrote, district courts may still be able to give “the functional equivalent of a universal injunction” by “granting . . . a preliminary injunction to a putative nationwide class.”  

The court’s paucity of guidance on the key yardstick of what makes relief “complete” (or “indivisible”) will also be significant. In many of the most important universal injunction cases of recent years, lower courts have rested their decrees either exclusively or in the alternative on their conclusion that a universal injunction was necessary to provide complete relief to the plaintiff. This presents the possibility that decrees that would formerly have cited alternative justifications for universal relief will now rest solely on the ground that they are required to ensure complete relief for the plaintiffs. Though Justices Clarence Thomas and Neil Gorsuch perceive this possibility and do not like it at all, they offer no guidance for fending it off other than the vague admonition that sometimes “traditional equitable limits will require courts and plaintiffs to make do with less than complete relief.” They do not tell us what those “traditional equitable limits” are or how lower courts should apply them; they also omit that the very article they cite points out that “equitable remedies” may either “go beyond, or stop short of, the strict right of the plaintiff” (emphasis added). Last but not least, they conspicuously fail to assert that the complete relief principle does not justify a universal injunction for the plaintiff states in this case. 

As Justice Gorsuch wrote in dissent in Gundy v. United States, “When one legal doctrine becomes unavailable to do its intended work, the hydraulic pressures of our constitutional system sometimes shift the responsibility to different doctrines.” So it will be here. Even as litigation will continue to proceed apace under the APA and its sister statutes, the announced abolition of the universal injunction will redirect pressure toward principles of state standing; to doctrines of associational standing; to the law governing injunctive class actions; and to the opaque inquiries of what counts as “complete relief” (or as an “indivisible” remedy). To put it another way, the metes and bounds of all of these doctrines will matter more now that the universal injunction simpliciter is off the table. It is true that a pair of concurring opinions by Justices Thomas, Alito, and Gorsuch together demand that each of these alternative doctrines be rigidly and narrowly confined, lest the universal injunction “return from the grave.” But I do not see why a lower court should heed these three justices’ marching orders – given that it’s not clear that any of the other six justices agree with them – rather than faithfully applying existing blackletter law to these categories.  

In sum, while other doctrinal levies may yet break, the court’s ruling itself may prove less catastrophic than it first appears. Still, to be clear, I disagree intensely with the decision, which rests on an amalgam of bad history, fiat, and ipse dixit. Among its many flaws: The Founding era is lauded as of critical importance, but no American Founding-era cases are cited. Rule 23 matters greatly because it allows class actions, but Rule 65 – the rule that actually addresses injunctions, and which does not limit preliminary or final injunctive relief only to the plaintiffs – apparently does not matter at all. Cases involving state laws (such as Scott v. Donald) count when they can be portrayed as casting doubt on universal injunctions, but not when they uphold such relief and explicitly deem it “appropriate” (see Pierce v. Society of Sisters). A landmark case decided in 1908, Ex parte Young, is an acceptable evolution of equity even though it worked an enormous change in our law, but then the clock mysteriously stops: Nothing else in the 20th century merits consideration. 

Apart from these and other internal inconsistencies and faulty logical jumps, the decision’s utter disregard of real-world consequences stands in stark contrast to the court’s own recent precedent. Just a handful of years ago, in Trump v. International Refugee Assistance Project, the court retained lower-court nationwide injunctions against an executive order issued by the first Trump administration. In deciding the stay applications, the court deemed it “necessary to balance the equities — to explore the relative harms to applicant and respondent, as well as the interests of the public at large.” That calculus led the court to leave in place injunctions that extended well beyond protecting the plaintiffs (or giving “complete relief” to them) to also protect the First Amendment associational interests of untold numbers of American “individual[s]” and “entit[ies].” In Friday’s ruling, by contrast, the court did not even acknowledge that the constitutionally (and statutorily) protected rights of myriad U.S.-born newborns now hang in the balance, let alone incorporate any consideration of those stakes into its one-clause, one-sided stab at evaluating “the balance of equities.”  

And though it is good that the court decided the case on statutory grounds, the flipside of this is that the court nowhere spared a thought to consider the strange consequences of its decision for relief against state and local governments. Because the court’s decision rested on Section 11 of the Judiciary Act, a successor provision to which is the familiar 28 U.S.C. § 1331, it speaks to the powers of federal courts to issue universal injunctions across the board – not just to challenges to federal executive branch action, but also to suits in federal court assailing state or local laws. All the familiar policy concerns about universal injunctions against federal laws and policies (that such remedies encourage forum shopping, diminish percolation, risk conflicting injunctions, and operate asymmetrically) are manifestly far less problematic in the context of challenges to state local laws and policies. And yet here we are: Under the court’s decision, a plaintiff in even a single-district state cannot obtain universal relief against the enforcement of even a blatantly unconstitutional state or local law or ordinance unless that remedy is necessary for that plaintiff’s “complete relief.” (Perhaps it wasn’t such a bad day for the state of New Jersey after all.) Certainly, if a state officer blares loud music at your house, your path to effective redress will be sure and swift, as the court’s trivial hypothetical reassures us. But if Oregon were to criminalize attending Catholic schools, then enjoining such a law for everyone will apparently require bringing a state-wide injunctive class action – despite the fact that 100 years ago, a federal court could and did simply grant a universal injunction against exactly such a law, as explored in my work. Equity, apparently, can evolve – as long as it shrinks.

And speaking of policy consequences: the court treats “the policy pros and cons” as “beside the point.” On that score, at least, the court’s cropped frame is apt, for its decision will not solve many, or perhaps even any, of the policy problems that have ignited so much ire against universal injunctions. After Friday’s ruling, injunctions must be tailored so that they provide complete relief to plaintiffs with standing; the universal injunction is verboten, but class actions and APA suits remain. 

What will this change as a policy matter? Last week, the government complained that universal injunctions “subvert[ed] the Article III hierarchy” by allowing one lowly district court judge’s decree to control the executive branch’s conduct nationwide. Next week, it will voice the same complaint when a district court judge certifies a nationwide class, grants an APA vacatur or stay, or issues a broad “complete relief” injunction. Last week, litigants had the incentive to forum shop for a court that would issue a universal injunction. Next week, litigants will have an incentive to forum shop for a court that will find that a universal remedy is necessary to give complete relief to the plaintiffs and/or for a court that will likely certify a nationwide class or vacate a rule.  And on and on we could go, through all the familiar litany: the sweeping up of non-parties (who may now be unwilling members of an injunctive class and/or unwilling beneficiaries of someone else’s complete relief or universal vacatur); the government’s “running the table” problem (for even after a string of victories, the government may face a court that decides that complete relief requires a sweeping remedy or that relief should issue to a nationwide class); the potential for “conflicting” injunctions (for courts may disagree on when complete relief requires protecting non-parties, or on whether a class may be certified); and the pressure on the Supreme Court’s emergency docket (because numerous emergency applications will doubtless flow from the lower courts’ adjudication of APA cases, nationwide class actions, and complete relief injunctions). These policy problems will not go away even if lower courts conscientiously adhere to all of the court’s instructions. 

By eliminating one form of relief while preserving multiple pathways to achieve functionally identical results, the court may have in the end accomplished little beyond handing the executive branch a litigation victory – one it has eagerly embraced.  The executive branch has already – and wrongly – seized upon the court’s decision as a ringing vindication of its claim that lower courts have unlawfully stymied the Trump administration’s breakneck barrage of executive orders. The decision seems destined to only embolden the resolve of the executive branch to push the envelope in its attacks on prevailing constitutional understandings and statutory constraints. But it bears remembering that the federal courts’ authority to provide meaningful remedies has always been crucial, both functionally and symbolically, to their capacity to pronounce robust constitutional norms and to maintain adequate checks on government power. When that remedial authority erodes – or even when it is perceived to erode – so does the rule of law. CASA has proclaimed a diminution of that authority at precisely a moment when that authority is needed urgently.  

Yet there remains room to hope that the lower courts will discern pathways to fulfill their constitutional obligation to secure the rule of law within the parameters that the court has pronounced. Now that the court has rejected the universal injunction, it must do its part to ensure that the remedial devices it has preserved remain fast, flexible, and sturdy enough to ensure that government officials stay within the bounds of legality.

Mila Sohoni is a Professor of Law and the John A. Wilson Distinguished Faculty Scholar at Stanford Law School, where she teaches Federal Courts and Civil Procedure. Her scholarship is available here.