The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
The Lowly Stay Takes Center Stage
SCOTUS' "stay jurisprudence" is a bit of a mess
SCOTUS' "stay jurisprudence" – the standards the Court uses to decide, inter alia, whether or not, during the weeks and months and sometimes years that are required before a final judgment is issued in any case, Executive Branch officials must obey the injunctions that have been entered against them – is of growing importance these days.
It's becoming increasingly clear – I'd call it obvious, even - that the lowly stay, heretofore a rather pedestrian and deeply unspectacular procedural device, is a central pillar of the President's strategy for dismantling statutory and constitutional constraints on executive power.
Here's how it works. The pattern is mind-numbingly familiar by now, having been repeated in dozens and dozens – probably hundreds, by now – of cases[1]:
1 The President issues an Executive Order, authorizing and directing his subordinates in the Executive Branch to do something (or to stop doing something that they had previously been doing).
Nobody connected to the Oval Office appears to spend much time worrying about whether the action complies with constitutional and statutory requirements,[2] or whether the President is authorized to do what he wants to do; he orders it done, and his subordinates start the process of getting it done. Rescind Harvard's federal grants. Close down US AID. Revoke the security clearances of everyone working at WilmerHale. Deport Venezuelan migrants. Close down the National Science Foundation. Impose a travel ban on 19 countries. Raise tariffs on Vietnamese textile imports. Lower tariffs on Vietnamese textile imports. Etc.
2 Lawsuits challenging the action assert that it violates some constitutional or statutory provision, or is otherwise beyond the power of the President to accomplish unilaterally through an Executive Order.
3 The challengers seek immediate, temporary relief, in the form [first] of a Temporary Restraining Order and [immediately thereafter] a Preliminary Injunction, ordering the Administration to cease whatever it was that it was doing (or had plans to do) while the case is being heard.
To obtain a P.I., the challengers must, of course, persuade the trial court, inter alia, that (a) they are "reasonably likely" to prevail on the merits (once the case reaches a final judgment), and that (b) they will suffer "irreparable harm" if the interim relief is not granted.
4 The Administration raises many objections – most of them procedural - to the motion for a P.I.
As it has, of course, every right to do: Standing, mootness, ripeness, justiciability, sovereign immunity, jurisdiction, the "political question" doctrine, etc.
5 But in the end, the court grants plaintiffs' requested relief.
Not always! But the Administration has lost and continues to lose an enormous and an alarming number of these cases at this preliminary stage[3] - "enormous" in comparison to any other presidency in US history,[4] and "alarming" because these are, after all, cases in which a federal judge (and often more than one federal judge) has said: "It is likely – at least on my first look at the case - that the Administration is breaking the law."[5]
- The Administration immediately seeks a stay of the judgment and the order for an injunction. And, increasingly often:
7 The Supreme Court grants the Administration's emergency application for a stay of the lower court's order.
Again, not always. But it has happened a lot, including in some of the highest-profile challenges to Administration action.[6]
*****
Obtaining a stay means, as both a practical and a legal matter, that the Administration may disregard the injunction entirely – i.e., it can do pretty much whatever it wants – until a final judgment has been reached and all avenues of appeal have been exhausted. These cases take a great deal of time to litigate to final judgment – months at a minimum, often years – even in the best of circumstances[7]; when the entry of a stay gives one of the parties such an obvious incentive to use every device imaginable to delay the proceeding as much as humanly possible - because it gets to do pretty much whatever it wants, lawful or not, in the interim – it can stretch out for years.
Hence the importance of the Court's rules regarding stays. If we are ever unfortunate enough to have a Chief Executive who would like total, one-man control over everything the federal government does or doesn't do, one who acts on dozens of different fronts simultaneously as though there were no legal or constitutional constraints on his power, a Court that was too generous with its stay-granting power would play right into that authoritarian-wannabe's hands.
Of course, the President will, ultimately, prevail in some, perhaps even many, of those cases. He will also lose in many; I am quite confident that the highest Court will find that he cannot unilaterally revoke birthright citizenship, that he cannot unilaterally set tariff rates, that he cannot unilaterally order federal troops into our cities to quell civic unrest or "fight crime," that he cannot order ICE agents to round up anyone who looks Hispanic and poor and take them downtown for questioning, he cannot unilaterally turn over the information in the Social Security Administration databases to his friends and political supporters, that he cannot unilaterally revoke all of a university's federal contracts because he doesn't approve of their DEI policies, etc.
But because the Court has granted the Administration's request for a stay of the lower court injunctions regarding each of these matters, he can do all of those things. And more. At least, for a good long while.
So perhaps I was wrong when I wrote that the President and his legal advisers give no thought whatsoever to the legality or illegality of the President's actions. Maybe they are cleverer than I give them credit for; maybe they understand more than I do about how the process works, and the Oval Office conversation has actually been going like this:
President to Legal Advisor: "I want to deny all federal disaster assistance to states that voted for Kamala Harris in 2024. Write me up an Executive Order to that effect."
Legal Advisor: "Um, Mr. President, I don't think you can do that."
P: "No? What'll happen if I do it?"
LA: "The courts will surely tell you to stop."
P: "How long will that take?"
LA: "Probably just a few days to slap you with an injunction – you really don't have a very good case, I'm afraid. However, …."
P: "'However' what?"
LA: "Well, If we can get a stay of the injunction – and the Supreme Court's been pretty generous in handing those out - we can probably stretch it so that you can do it for a year or two."
P: "Now you're talkin'! Do it."
So we really do need to understand how the Court decides whether or not to grant a stay in cases in which the Administration is acting – or at least appears, at first glance, to a federal district judge and possibly to an Appeals Court panel as well, to be acting – unlawfully, if we want to understand whether and how our constitutional system can withstand this kind of assault by some dictator-wannabe.
*****
It's a little hard to say how the Court is deciding these stay cases, given the custom that has arisen whereby the Court rarely reveals its reasoning when deciding cases on its "emergency docket."[8]
Here, though, is the standard formulation of the "familiar rules that govern stay applications," taken from an opinion by Justice Gorsuch, concurring in a case granting the State of Idaho a stay last term (Labrador v. Poe, 2024):
"This Court, like every other federal court, is guided by the same 'sound principles.' Nken v. Holder, 556 U. S. 418, 434 (2009). We ask (1) whether the stay applicant has made a strong showing that it is likely to succeed on the merits, (2) whether it will suffer irreparable injury without a stay, (3) whether the stay will substantially injure other parties interested in the proceedings, and (4) where the public interest lies. Nken, 556 U. S., at 434.
Notice that the Court does not ask whether the stay applicant is "likely to succeed on the merits"; it asks whether the stay applicant has made a "strong showing" that it is likely to succeed on the merits.
That strikes me as a pretty significant difference. And one that actually matters, too. Here's why:
-
In any contest or experiment with two possible outcomes – e.g., plaintiff wins/defendant wins, petitioner wins/respondent wins, Brewers win/ Dodgers win, the stock market goes up/the stock market goes down, it will rain tomorrow/ it will not rain tomorrow, etc. – an outcome is "likely to occur" only if it has a probability of occurrence that is greater than 50%.[9]
-
That means that the other outcome has a probability of occurrence that is less than 50%.[10]
-
And that means that only one of the two possible outcomes can be "likely" to occur. If plaintiff is likely to prevail, the defendant is not likely to prevail (and vice versa). They're mutually exclusive categories; the plaintiff and the defendant cannot both be likely to prevail, just as it cannot be likely to rain tomorrow and likely not to rain tomorrow, or that the Brewers and the Dodgers are both likely to win tonight's game.
"Strong showings," though, aren't like that; they're not mutually exclusive outcomes in this way. Anyone who has ever attended a high school debate tournament (or a Supreme Court oral argument) knows that both sides in a 2-person contest can make a "strong showing that they're likely to prevail." Happens all the time.
So in this sense, the "strong showing" requirement is a lower bar, easier to satisfy, than the "likely to prevail" requirement. There's nothing illogical about the position both that the Administration has made a strong showing that it should prevail, and that the challengers are likely to prevail (because they have made a stronger showing that they should prevail).
Persuading a court that you have made a "strong showing" that you are likely to prevail on the merits does not require you to persuade the court that you have a greater than 50% chance of winning. As a logical matter, persuading a court that you're "likely to win" is equivalent to persuading a court that the other side is "likely to lose." But persuading a court that you've made a "strong showing" that you're likely to prevail does not require you to show that the other side is likely to lose; it is perfectly logical to hold the position that the Administration, though it has made a "strong showing" that it is likely to prevail, is actually likely to lose – just as the lower courts in all of these cases have found.
Remember: the cases we're talking about are cases in which a lower court has entered an injunction against the Administration. The Administration is applying for a stay precisely because it lost in the court(s) below, and it lost because the challengers persuaded the court below that, among other things, they were "likely to prevail" on the merits, and that the Administration was therefore not "likely to prevail" on the merits.
It follows, then, that when the Court grants a stay to the Administration, it is not overturning the lower court's finding that the challengers are "likely to prevail" on the merits, nor is it saying that the Administration is likely to prevail on the merits.
It is surprising how many otherwise perspicacious observers of the legal scene and developments in legal doctrine appear to misunderstand this point. Justice Gorsuch, for one. In the case I quoted from above, immediately after stating that the Court, when evaluating a stay application, first asks "(1) whether the stay applicant has made a strong showing that it is likely to succeed on the merits," he writes:
"Applying that traditional stay test here yields a ready answer. Start with the first question: whether Idaho [the applicant for the stay] has shown it is likely to succeed on the merits . . ."
Weird, no? He just told us the first question is "whether the applicant has made a strong showing that it is likely to succeed on the merits," and now he says the first question is "whether the applicant has shown it is likely to succeed on the merits." Just sloppy drafting? Or is he signaling that the Court's position is that the lower court was incorrect when it granted the challengers a preliminary injunction?[11]
And things get weirder. Though the Court as a whole is generally silent when dealing with these "emergency" stay requests, Justice Kavanaugh, to his credit, has written separately to explain his reasoning for joining the majority. In the recent case of Noem v. Vasquez Perdomo, where the Court stayed a lower court injunction prohibiting ICE agents from detaining individuals based on nothing more than their race, ethnicity, clothing, and job-hunting activities,[12] he wrote:
"To obtain a stay from this Court, the moving party must demonstrate a fair prospect that, if the District Court's decision were affirmed on appeal, this Court would grant certiorari and reverse. The moving party also must show a likelihood that it would suffer irreparable harm if a stay were not granted. Those two factors are the "most critical." Nken v. Holder, 556 U. S. 418, 434 (2009)."
A "fair prospect" that the Court would reverse the 9th Circuit's decision here – what, pray tell, does that mean?
It sounds like something out of Jane Austen or Thomas Hardy, no?
From Blenheim peak, 'tis a fair prospect/Over the hills of Wessex to the sea …
That's presumably not what Justice Kavanaugh has in mind. I invite commenters to take a crack at what they think he does have in mind. To my mind, a "fair prospect" that the Administration will ultimately prevail is an even lower standard than either (a) a "strong showing" that it will prevail, or (b) a finding that it is "likely" to prevail. But maybe that's just me.
Not surprisingly, here too commentators have, I think, been confused, and have missed the main point. Amy Howe on ScotusBlog, for instance, writes this:
"In a concurrence, Kavanaugh explained that the Trump administration had met both of the "most critical" factors that the court considers in deciding whether to grant temporary relief. First, he explained, "given the significance of the issue to the Government's immigration enforcement efforts," the Supreme Court would be likely to grant review if the 9th Circuit were to uphold [the district court's] injunction.
Second, Kavanaugh continued, the Trump administration has shown not only that the Supreme Court would be likely to grant review, but also that it would be likely to reverse the lower court's ruling in the challengers' favor. [13]
I don't think Justice Kavanaugh said that. By his own terms, he has determined that the Administration has shown it has a "fair prospect" the the Court would reverse, not that it is likely to reverse. It makes a big difference, I think.
All of which leaves the question: Why should the applicant for a stay have a significantly lighter burden than the applicant for a preliminary injunction?
[1] See the excellent compendium of the 498 (as of 9/18/2025) cases filed against Trump Administration policies at https://clearinghouse.net/collections/38759, or the one at JustSecurity.com.
[2] A week after the inauguration in 2025, I wrote:
"He [Trump] has figured it out: He can do whatever he wants, and nobody can stop him. Unilaterally impose 25% tariffs on Colombian imports? Do it! Withhold federal disaster relief to cities that don't assist ICE agents carrying out their raids? Absolutely! Halt all payments due for dispersal under NIH research grants until grantees dismantle their DEI programs? Sure!
Does anyone actually believe that, during the discussion in the Oval Office concerning those moves, Trump asked: 'Are we sure we have statutory and constitutional authority to do this?'?"
"One Man Rule", Jan 27, 2025.
[3] In his guest post here on the VC several months ago, Prof. Abner Greene helpfully compiled a list of 28 cases (pending as of May 31, 2025 – 4 months or so after Inauguration Day) in which Administration officials had been enjoined from carried out a presidential policy directive. The pace has picked up considerably since then. See the compendia listed in note 1. [It does make me wonder: Are U.S. Attorneys doing anything besides defending the Administration in these lawsuits?] Visit either of those websites and click on any of the categories and notice how many "Preliminary Injunction Granted" labels you see. It is a remarkable record of lawlessness on the part of our government. That is has become so commonplace – another day, another outrage, and another P.I. - should not stop us from remarking, at every opportunity, on just how awful it is in its lawlessness.
[4] I don't know of any actual data on the number of times that, say, Presidents GW Bush, or Clinton, or Eisenhower, were enjoined, even preliminarily, from doing what they wanted to do. I am nonetheless quite certain that the number was lower – probably by an order of magnitude or two -- than is the case with the Trump Administration (which is now in triple digits on this metric. See note 6 below.), despite having had a mere 8 months to achieve this grim milestone.
[5] They're not, of course, final judgments that the Administration is acting unlawfully. Final judgments are, in most cases, months away – in part because there are some pretty complicated procedural and substantive legal issues in many of these cases, in part because the Administration has many tools, and the incentive, to delay the proceedings for as long as possible. These P.I.s only represent a judge's initial take on the matter. But still – 100+ and counting?
[6] Here is my hastily-compiled and undoubtedly incomplete list of cases where SCOTUS has stayed preliminary injunctions against Trump Administration officials: Boyle v. Trump (D. MD) (firing CPSC Commissioners); Harper v. Bessent (DDC) (firing of MSPB members); Am. Fed. Gov't Employees v. Trump (ND CA) (mass firings of probationary employees); W.M.M. v Trump (ND TX) (Alien Enemies Act deportation power); AFSCME v. Social Security Administration (D MD) (challenge to SSA data disclosures to DOGE); Wilcox v Trump (DDC) (firing of NLRB member); AFGE v OPM (ND CA) ; New Jersey v. Trump (D MA) (birthright citizenship executive order); Massachusetts v. Kennedy (D MA) (research funding freeze); California v. US Dept of Education (D MA) (teacher training funding freeze); Citizens for Responsibility and Ethics in Gov't v DOGE (DDC); D.V.D. v. DHS (D MA) (deportation removals); Doe v. Noem (D MA) (immigration status changes); National TPS Alliance v. Noem (ND CA) (status change for Venezuelan immigrants); Massachusetts v. Kennedy (D MA) (research funding freeze); Am. Public Health Assn. v. Kennedy (D MA) (termination of NIH grants); Oregon v. Trump and V.O.S. Selections v. Trump (Ct of Int'l Trade) (tariffs); Shilling v. Trump (D WA) (military trans ban).
[7] The wheels of justice, as the saying goes, turn slowly. If you're unfamiliar with the pace and process of complex constitutional take a look at some of the complete docket listings in some of the cases compiled at at https://clearinghouse.net/collections/38759, or JustSecurity.com. Scores, even hundreds, of docket entries for pretty much every case – motions challenging plaintiffs' standing or the case's ripeness for resolution, motions to dismiss, motions for summary judgments, motions requesting permission to file amicus briefs, motions to admit lawyers pro haec vice, motions requesting extensions of filing deadlines, motions requesting class certification, motions contesting any of the other motions, status conferences, transcripts of status conferences, motions to seal certain documents, motions to otherwise limit public disclosure of the court's deliberations, declarations averring or contesting the factual allegations in plaintiff's complaint, etc. Not to mention that the legal issues raised may be themselves thorny and complex, requiring time to research and review. It's more surprising that they ever reach final judgment than that it takes them a long while to do so.
[8] See Justice Sotomayor's dissenting opinion in a recent case granting the Administration's request for a stay, Noem v. Vasquez Perdomo:
"The Court's order is troubling for another reason: It is entirely unexplained. In the last eight months, this Court's appetite to circumvent the ordinary appellate process and weigh in on important issues has grown exponentially. See, e.g., Trump v. J. G. G., 604 U. S. ___ (2025); Trump v. Wilcox, 605 U. S. ___ (2025); Noem v. Doe, 605 U. S. ___ (2025); United States v. Shilling, 605 U. S. ___ (2025).
Its interest in explaining itself, unfortunately, has not. See Trump v. Boyle, 606 U. S. ___, (2025) (KAGAN, J., dissenting from grant of application for stay) (slip op., at 2–3)
There may be good justification for issuing an unreasoned order in some circumstances. See Labrador v. Poe, 601 U. S. ___ (2024) (Kavanaugh, J., concurring in grant of application for stay) (explaining how opinions on emergency applications "can sometimes come at a cost"). But some situations simply cry out for an explanation, such as when the Government's conduct flagrantly violates the law, or when lower courts and litigants need guidance about the issues on which they should focus.
[9] Think about a single toss of a fair coin. Two outcomes: Heads or Tails. "Heads" is not a "likely" outcome of that toss. Neither is "tails." Though each has exactly a 50% probability of occurrence, 50% is not sufficient to make something "likely." The "likely" outcome of a two-outcome contest must have a greater than 50% chance of occurring.
[10] Because the probabilities of the two outcomes summed together cannot be more than 100% (by definition).
[11] Justice Sotomayor, too, appears to drop the "strong showing" language from time to time. See, e.g., her dissent from the Court's decision to grant a stay in Noem v. Vasquez Perdomo, a case involving ICE's detention policies.
"A stay is an 'intrusion into the ordinary processes of administration and judicial review.'" Nken v. Holder, 556 U. S. 418, 427 (2009). This Court will not grant one absent "extraordinary circum_stances." Ruckelshaus v. Monsanto Co., 463 U. S. 1315, 1316 (1983). The applicant bears "an especially heavy burden" where, as here, the matter remains pending before the Court of Appeals, and two lower courts have already denied such relief. Packwood v. Senate Select Comm. on Ethics, 510 U. S. 1319, 1320 (1994).
Ordinarily, the Court considers the applicant's likelihood of success on the merits, the likelihood of irreparable harm absent emergency intervention, and the balance of the equities. See Nken, 556 U. S., at 434; see also Hollingsworth v. Perry, 558 U. S. 183, 190 (2010). . . . An applicant must not only show a likelihood of success on the merits, but must also demonstrate that "the implementation of the judgment pending appeal will lead to irreparable harm." Graves v. Barnes, 405 U. S. 1201, 1203 (1972).
As in Justice Gorsuch's case, one cannot help but wonder if this is a strategic omission. Justice Sotomayor has dissented from most of the Court's recent stay decisions, and would clearly like the Court to be applying a higher standard to the applicant's request than merely that it had made a "strong showing."
[12] Noem v. Vasquez Perdomo is one of the more inexplicable of the Court's recent "emergency docket" rulings staying a district court's preliminary injunction against an Administration policy. [See ScotusBlog entry here with links to the relevant primary documents, including the district court opinion here].
Long-standing precedent allows immigration enforcement agents to "briefly detain an individual for questioning if [the agents] have a reasonable suspicion, based on specific articulable facts, that the [detainee] is an alien illegally in the United States." See U.S. v. Brignoni-Ponce, 422 U. S. 873, 884 (1975). ;
This past summer, roving partrols of ICE agents were deployed to car washes, Home Depot parking lots, tow yards, bus stops, and other locations in and around L.A., detaining various individuals and demanding proof of US citizenship; individuals who could not provide such proof on the spot were transported to an ICE questioning center in downtown L.A. for further questioning.
The district court (CD CA) found that the agents' decisions about whether or not to detain someone were based "solely on four factors: (1) the individual's apparent race or ethnicity; (2) whether they spoke Spanish or English with an accent; (3) the type of location at which they were found (such as a car wash or bus stop); and (4) the type of job they appeared to work. Concluding that that that the presence of these four factors cannot satisfy the requirement of "reasonable suspicion based on specific articulable facts," the court concluded that the challengers were "reasonably likely to prevail" in their challenge and that they would suffer "irreparable harm" if an injunction were not granted pending a full trial on the merits, and accordingly preliminarily enjoined ICE from using these factors, "alone or in combination," in deciding whom to detain.
The Court's decision to stay the injunction is appalling, insofar as it allows continuation of the practice whereby armed agents of the United States government confront and detain individuals solely because they speak English with an accent, have brown skin, and are hanging around a Home Depot parking lot looking for construction work. I'm no expert on the 4th Amendment, but I would naively assume – and hope - that the 4th Amendment's prohibition on "unreasonable searches and seizures" prohibits this practice. The district court certainly took this position, in holding that the challengers are "likely to succeed" on their claim, as did the 9th Circuit in upholding the injunction and denying the Administration's request for a stay. Indeed, I would think a challenge to such a practice would be something of a constitutional slam dunk.
But in the meantime, the practice can continue.
[13] See also Steve Vladek, "Brett Kavanaugh's Defense of the Shadow Docket is Alarming":
Stays are supposed to be an 'extraordinary remedy' meant for extraordinary cases where the three different things are true: First, the party seeking the stay must be likely to win their appeal. Second, it must be the case that not freezing the lower court ruling while it appealed would cause the appealing party 'irreparable harm.' And third the public interest must be served by such relief.
