A little bird (ok, a normal-size adult human being) has asked me a question about the U.S. Supreme Court’s recent decision in Trump v. CASA. It might not be the question on the tip of everyone’s tongue, although it’s a natural one for those interested in competition policy, administrative law, or all things (or at least some things) related to the Federal Trade Commission (FTC). That is, what does—or might—Trump v. CASA mean for the ultimate fate of the FTC’s now-stayed noncompete rule?
Spoiler alert: not much. At least, I shouldn’t think so. Then again, there may be significant implications down the road for other competition matters—especially those involving private plaintiffs.
I’ll get to those, although I’ll start with the decision itself and its possible implications for the rule. And depending on the reader, this is either a trigger warning or a trigger assurance: in what follows, there’s a whole lot of law and precious little economics.
For those who haven’t read it, Justice Amy Coney Barrett’s majority opinion in CASA decided “three emergency applications challenging the scope of a federal court’s authority to enjoin Government officials from enforcing an executive order.” Three different federal district courts had each issued a “universal” preliminary injunction barring implementation and enforcement of Executive Order No. 14160, inaptly titled “Protecting the Meaning and Value of American Citizenship.”
The controversy over the executive order is well-familiar by now, and entirely understandable. It seems clear to me that the executive order is inconsistent with the 14th Amendment, but the Court has not said otherwise and I am, to paraphrase the fictional Dr. McCoy, just a poor country law and econ guy with a focus on antitrust, and not a constitutional law scholar. To hew closer to my lane, I want to address the scope of the injunctions that were at issue in CASA, what the Supreme Court has said about them, and what that might imply for the status of the FTC’s noncompete rule, and perhaps other antitrust matters.
In each of the three challenges “the District Court concluded that the order is likely unlawful and entered a universal preliminary injunction” against the executive order.
What’s a “universal” injunction, you ask? Several issues seem to be at play. First, can a single federal district court issue an injunction that applies nationwide? Second, can a federal district court issue an injunction that provides relief to nonparties to the case before it? And third, has Congress “granted federal courts the authority to universally enjoin the enforcement of an executive or legislative policy.”
To that third question—and, it seems, the second—the Court says “no.” It’s a bit complicated, but I won’t delve into the details of Barrett’s historically grounded majority opinion, the three concurring opinions (by Justices Clarence Thomas, Samuel Alito, and Brett Kavanaugh), or the dissents, which include a 44-page dissent written by Justice Sonia Sotomayor, and joined by Justices Elena Kagan and Ketanji Brown Jackson and an additional dissent by Jackson, which agrees “with every word of Justice Sotomayor’s dissent” and then some.
The majority opinion is defensible, interesting, and, like most of the Court’s forays into historiography, eminently debatable. Or, as Barrett writes (more charitably, with charity beginning at home), “[a]s with most disputed issues, there are arguments on both sides.”
Skipping ahead to the holding, it is this:
Nothing like a universal injunction was available at the founding, or for that matter, for more than a century thereafter. Thus, under the Judiciary Act, federal courts lack authority to issue them.
So, federal courts may not issue “universal” injunctions—whatever “universal” means.
That might seem a sweeping limitation on the power of judicial review, and perhaps it is. Or perhaps it portends one. But the holding doesn’t mean as much in context as it seems to when read in isolation. That is, the limitation has its limitations. The Court itself notices some of these potential limitations, as do the concurring opinions from Alito and Kavanaugh. And as do an increasing number of commentators (see, e.g., Steve Vladek, Jack Goldsmith, Noah Feldman, and Curtis Bradley). “Universal” may not be so universal, after all.
One limitation on the holding’s reach is that the Court’s opinion isn’t a decision on the merits. It’s not even a decision on the merits of the question of the scope of relief—let alone a decision on the merits of the underlying Citizenship Clause question. Rather, it is predicated on a determination that “[t]he Government is likely to succeed on the merits of its argument regarding the scope of relief” (emphasis added).
That’s not to make light of the “likely to succeed” determination. Antitrust folk will no doubt recognize that preliminary decisions of “likely” or “not likely” to succeed on the merits are neither final nor small beer in, e.g., merger challenges. Still, I expect that a decision on the merits of the remedy will be longer and clearer than the preliminary one just issued, even if it’s highly unlikely to adopt the position of Sotomayor’s dissent. And “highly unlikely” might be an understatement. By a lot.
For what it’s worth, I also doubt that the Court will, ultimately, adopt the administration’s reworking (the White House might say “clarification”; others might say “fictionalization”) of the Citizenship Clause.
Second, and not for nothing, but even as a preliminary matter, the Court granted the government only partial relief: the government did not seek, and the Court did not grant, full stays of the district court injunctions at issue. Those injunctions stand as they apply to individual named plaintiffs in the underlying suits (albeit with some residual question about the individual states that were named as plaintiffs). That may be little comfort to those left out in the cold by the partial relief, as narrowed by the Court. But it’s a limitation now, and it may prove the basis of significant limitations going forward.
Third, the limitation, such as it is, seems to be a “thee but not we” limitation. That is, when the Court says that “federal courts” lack the authority to issue universal injunctions, it seems to mean “lower federal courts”—certainly federal district courts, and probably federal courts of appeal—but not the U.S. Supreme Court itself, notwithstanding that careful readers might note that the Supreme Court is, in fact, a federal court. Or that nothing in the Judiciary Act of 1789 makes quite this distinction between the Supreme Court and lower federal courts or expressly limits the authority granted the federal courts over, e.g., “all suits of a civil nature at common law or in equity . . .”
Jack Goldsmith, the Learned Hand Professor of Law at Harvard and a former U.S. Justice Department (DOJ) official, provides a useful summary of the proceedings and the repeated emphasis at oral argument on the Supreme Court’s ultimate authority in both questioning by the justices and assurances by the solicitor general. As Goldsmith (and others) have noted, the Court’s decision in CASA:
…leaves open many avenues to lower-court universal relief: class actions, APA vacatur, certain suits by states, other forms of suit that demand complete relief, and “indivisible” remedies that incidentally benefit third parties.
He also points to perhaps the clearest claim of supremacy on behalf of the Supremes; that is, a long footnote (number 18) in which “the Court made clear its expectation that the executive branch will comply with its ‘judgments and opinions.’” In sum, “Trump v. CASA did not diminish the Supreme Court’s authority vis-a-vis the presidency.” Judicial review remains undisturbed in that regard, at least as an in-house affair.
You can take that limitation to the bank. A bank will give you nothing for it, but you can take it anywhere you like. That Goldsmith is right is my best guess, and I think it’s a good one, but . . . yeah.
Fourth, there are various statutory limitations suggested by the Court’s opinion, including two discussed expressly in Kavanaugh’s concurrence. The basis of the Court’s decision in this matter is statutory—or, at least, the Court’s reading of the Judiciary Act of 1789—which, the court notes:
…endowed federal courts with jurisdiction over “all suits . . . in equity,” §11, 1 Stat. 78, and still today, this statute “is what authorizes the federal courts to issue equitable remedies,” (citations omitted)
But, but, but . . . the Court in CASA also recognizes that the Judiciary Act is not the only basis on which the federal courts can issue other types of remedies in civil matters, including remedies that might incorporate injunctions that seem, say, universal-ish. These include, for example, class-action suits, in which named plaintiffs may be numerous, and challenges to federal regulations.
The potential for suits seeking injunctive relief for a plaintiff class is provided under the Federal Rules of Civil Procedure (as noted by the Court, and in concurring opinions by Alito and Kavanaugh).
And then there’s the statutory “loophole” that might matter most for the FTC’s noncompete rule (or, indeed, any competition or consumer protection rule adopted by the FTC). As I noted way back in August 2024:
courts have uniformly held that invalid federal regulations can be subject to vacatur, and hence “set aside” by reviewing courts under the Administrative Procedure Act (APA). See, for example, the D.C. Circuit here and the 5th Circuit here.
That’s because there is a statutory basis for such relief, and it’s in the 20th century APA, not the 18th century Judiciary Act:
Section 705 of the Administrative Procedure Act authorizes federal courts to stay enforcement of a rule pending judicial review; and Section 706 gives reviewing courts the power to “set aside agency action, findings, and conclusions” that are found to be unlawful.
As it happens, in ruling against the FTC’s authority to issue the noncompete rule in the Ryan case, vacatur is precisely what Judge Ada E. Brown of the U.S. District Court for the Northern District of Texas provided:
The Court sets aside the Non-compete Rule. Consequently, the Rule shall not be enforced or otherwise take effect on its effective date of September 4, 2024, or thereafter.
While a prior preliminary ruling had “preliminarily enjoined implementation and enforcement of the Rule [only] as to the named Plaintiffs,” there was no such limitation in Judge Brown’s ultimate ruling on the merits: her issuance of vacatur applies nationwide, as authorized by the APA.
Put it another way: Does anyone really think that the CASA majority meant to repudiate the Court’s own power to review federal regulations and, if necessary, to remedy agency overreach “universally”? The Court that repudiated the Chevron doctrine in Loper Bright in 2024? The West Virginia v. EPA Court?
These are not, of course, the only reasons to doubt reanimation of the FTC’s sweeping noncompete rule. Both Chairman Andrew Ferguson and Commissioner Melissa Holyoak issued stinging dissents when the commission adopted the rule on purely party-line votes. They both argued that the FTC had overstepped its authority in adopting the rule. In March 2025, the FTC filed motions to stay its appeals of both the Texas decision and another by a federal court in Florida.
And as it happens, none of the three commissioners who voted to adopt the rule (then-Chair Lina Khan and then-Commissioners Rebecca Slaughter and Alvaro Bedoya) remains to support reinstatement of the appeals, although I should note ongoing legal challenges to the dismissals of Bedoya (who has resigned his position, in any case) and Slaughter (who, so far as I know, hasn’t).
I suppose it’s possible that the commission could change course yet again. Ferguson is reported to have acknowledged that possibility in public remarks—somewhat odd, in my view, given the strength of his dissent from the rule’s adoption—but there’s nothing to suggest that such a reversal is imminent. And the likelihood that this Supreme Court would, ultimately, sustain such a challenge seems slimmer still, for all the much-discussed reasons.
I’ve written quite a bit about noncompetes and the FTC’s rulemaking already, and I’m not going to recapitulate all of that. In addition to my August 2024 post “Vacatur’s All I Ever Wanted,” there were posts here, here, here, here, and here at Truth on the Market; here in a peer-reviewed journal article; and here in extensive comments submitted with my International Center for Law & Economics (ICLE) colleagues and scholars of law & economics.
But this post is not about the merits of the noncompete rule. It’s not even about the question whether the FTC Act grants the FTC the authority to adopt competition rules generally. Rather, it’s about the power of the judiciary to review such rules, and to remedy, nationwide or “universally,” those it finds unlawful.
And to return to our initial question, the rule’s slumber should be undisturbed by the Court’s decision in CASA.
Private Plaintiffs and a Potential Loophole to Some of the Loopholes
None of the above should suggest that, for antitrust purposes, there’s nothing to see here. There may be plenty. And that brings us back to my earlier suggestion about antitrust suits brought by private plaintiffs. As Daniel Francis and Christopher Sprigman note in their casebook “Antitrust”:
Despite the prominence and importance of the federal agencies, most federal antitrust cases in the United States are not brought by the Department of Justice, the FTC, or the State Attorneys-General. Rather, most antitrust litigation and adjudication is the result of private enforcement: that is, lawsuits brought by consumers, competitors, and trading partners.
And we might add, despite the prominence and importance of the federal antitrust agencies and state enforcers, suits brought by private parties have led to some of the Supreme Court’s major antitrust decisions: Leegin, Trinko, NCAA v. Alston, etc.
Private plaintiffs can bring antitrust complaints, provided the plaintiffs satisfy the requirements of standing—both Article 3 standing and antitrust standing, which requires not just an antitrust violation, but injury-in-fact to the plaintiff itself, with that injury suffered by reason of the antitrust violation.
But suppose a plaintiff satisfies the requirements of standing and, moreover, prevails in establishing antitrust liability. What might CASA suggest about a (lower) federal court’s ability to issue universal injunctive relief as a remedy?
Consider Epic v. Apple as a case in point. In that case, Epic sued Apple, alleging that Apple’s operation of its app store and in-app payment (IAP) system violated both federal antitrust law—specifically, the Sherman Antitrust Act—and California’s Unfair Competition Law (UCL). The district court found in favor of Apple on Epic’s Sherman Act claims, but it found that Apple’s anti-steering provisions violated California’s UCL. The 9th U.S. Circuit Court of Appeals took issue with various aspects of the lower court’s opinion but, in the end, sustained both the federal antitrust decision (against Epic) and the state UCL decision (against Apple).
More to the point of the present discussion, the 9th Circuit left undisturbed the district court’s remedy: a nationwide injunction against Apple restricting not just its conduct with regard to named plaintiff Epic, but with regard to all app developers. And just this year, Judge Yvonne Gonzalez Rogers ruled that Apple had violated the terms of her 2021 injunction and, in response to a contempt motion, issued an even more sweeping universal injunction against Apple.
It’s an odd and complicated set of decisions—for critical perspectives, see ICLE’s amicus briefs here, here, and here. But let’s leave for another day the details of the case. Its connection with CASA does not turn on the strengths or weaknesses of Epic’s repudiated antitrust allegations or its UCL allegations. Rather, there’s the simple fact that a lower federal court issued a nationwide injunction ranging over nonparties to the underlying complaint, as the only plaintiff in Epic v. Apple was Epic.
Leave aside that neither the 2021 injunction nor the more sweeping 2025 order is predicated on violation of a federal law. That itself raises a different sort of question about the proper scope of a district court injunction: can a federal district court (for example, the U.S. District Court for the Northern District of California) order nationwide equitable relief to remedy a violation of state law (say, California’s UCL)?
That’s an interesting question. I’m not sure that CASA implies a specific answer to it, although at least several justices (Alito concurring in CASA, Gorsuch concurring in U.S. v. Texas, and Thomas concurring in Trump v. Hawaii) have raised various concerns about overbroad injunctions, as has the majority elsewhere, as in Lewis v. Casey.
Still, the district court in Epic v. Apple could have issued a similar injunction—indeed, the very same injunction—had it found a violation of the Sherman Act. But a federal cause of action would have been immaterial to the extent that the Court’s concern about universal injunctions has to do with the authority to provide a complete (or partial) remedy to the parties to the suit. As noted above, Epic is the only plaintiff in Epic v Apple, and as Geoff Manne and I noted elsewhere, the district court’s order in Epic v. Apple raises Article 3 standing concerns. But it also:
…risks harming the vast majority of app developers, who have not requested the injunction and are now operating on the iOS for free.
Neither those many other developers, nor the millions of consumers who use Apple’s App Store, were parties to the suit. And on those grounds, the Supreme Court’s decision in CASA suggests, if only as a preliminary determination about a different remedy, that the District Court for the Northern District of California has no such authority.
One might imagine a class-action suit attacking the same conduct, but that would require class certification, and it might still limit the scope of relief. Numerosity does not imply universality.
The CASA decision says nothing expressly about Epic or Apple, and its precedential value for antitrust matters, among others, may be limited in the near term. Subsequent decisions may sharpen the implications of CASA, both with respect to limits on the lower federal courts’ jurisdiction and with respect to “loopholes” or “workarounds,” via the APA, class-action suits, and antitrust claims brought by the states. Some of the sharpening may push one way, and some another.
Still, an ultimate holding on the merits of the remedy could matter across diverse antitrust suits—especially suits in which private parties allege antitrust violations, and in which courts might enjoin conduct to benefit (or otherwise affect) nonparties. That’s of special concern, given the statutory authority for states, firms, and human persons to bring certain antitrust suits (or the statutory authority for issuance of vacatur under the APA). But it’s also a matter of concern given the requirements of antitrust standing, which are laser focused on questions of harms to individual plaintiffs, and the source of those harms, in addition the demands of showing harm to competition.
This is all hypothetical, which is not to say baseless. We shall see where it goes.
