Should the Government Have Won by Losing in Vasquez Perdomo?

Overlapping Stays on the Irregular Docket

Last week, the Supreme Court issued a controversial stay in Noem v. Vasquez Perdomo. I have offered thoughts on the case; so have many others. (See eg Kerr, Ifill, Vladeck, and Dorf.)

Here, I want to focus on an aspect of the case that could reasonably be dismissed as an academic wrinkle or footnote—namely, the relationship among the case’s three orders: the district court’s injunction, the Ninth Circuit’s partial stay, and the Court’s stay. Focusing on the meaning and interrelationship of these orders suggests a conclusion that is as surprising as it is attractive: the Administration should have effectively won the case by losing.

We should start with some simplified background. Plaintiffs filed suit in connection with an alleged practice of unconstitutional immigration-enforcement stops in the Los Angeles area. About ten days later, after expedited briefing and a similarly expedited adversarial hearing, the district court issued a temporary restraining order. Here is that injunction’s key text:

b. In connection with paragraph (1), Defendants may not rely solely on the factors below, alone or in combination, to form reasonable suspicion for a detentive stop, except as permitted by law:

i. Apparent race or ethnicity;

ii. Speaking Spanish or speaking English with an accent;

iii. Presence at a particular location (e.g. bus stop, car wash, tow yard, day laborer pick up site, agricultural site, etc.); or

iv. The type of work one does.

As drafted, the district court’s injunction was critically ambiguous, as several observers have noted. Here are two salient ways of reading it, both drawn from my own earlier post.

First, the injunction could prohibit stops based entirely and exclusively on the four factors or a subset thereof. That reading finds support in the words “rely solely.” However, it also renders the injunction a virtual nullity ab initio. For officers can almost always adduce a fifth factor, thereby taking themselves out of the injunction’s scope. The starkest example: officers could rely in part on the high numbers of undocumented persons in the area, a factor that seems present literally everywhere that the injunction applies.

Second, the injunction could mean that any consideration of the four factors is prohibited, subject to after-the-fact review for compliance with the Fourth Amendment and other legal requirements. That reading trades on a different meaning of “rely” and finds support in the “except as provided by law” phrase. On this reading, the injunction at least initially sweeps in a vast amount of law-enforcement activity, much of which is constitutional.

When the Ninth Circuit confronted the injunction, it saw two problems. First, the injunction contained a typo: there is no “paragraph (1)”. Rather, there are only two paragraphs: (a) and (b). The court of appeals quite sensibly overlooked this difficulty, in effect applying in a mild “scrivener’s error” principle to the order. Second, and more significantly, the court of appeals recognized that the “except as permitted by law” phrase rendered the injunction impermissibly vague. In other words, the court of appeals concluded that the injunction as issued was unlawful due to a form of ambiguity.

In the hope of curing the injunction’s defect, the court of appeals purported to stay the “except as permitted by law” phrase—and only that phrase—thereby leaving the rest of the injunction in effect. The court posited that, as stayed, the injunction would carry the first, narrow meaning described above. In its clearest statement to that effect, the court of appeals noted: “The TRO does not expose Defendants to the threat of contempt when they make a stop based on other factors—even if a court later concludes that Defendants lacked reasonable suspicion for the stop.”

But the appellate court did not seem to think that that purported re-writing of the injunction effectively rendered it a nullity. On the contrary, the court of appeals thought that the injunction-as-partially-stayed would still matter. For example, the Ninth Circuit appears to have concluded that officers would transgress the injunction if they relied not just on the fact that “a person is present at a business” such as a car wash, but also on the "officers' knowledge that illegal immigrants have frequented or sought work at that location.” (As we saw in my earlier post, Justice Sotomayor’s later dissent cast a very similar scenario as outside the injunction.)

Against that backdrop, the Administration applied to the Supreme Court for a stay of the district court injunction. Now, this procedural posture is quite unusual—or at least used to be. Normally, the Court hears cases that have proceeded through an appellate process, resulting in an appellate judgment. The appellate judgment is then appealed or (to similar effect) challenged via certiorari. The Court is then able to affirm, reverse, vacate, and so forth.

That, however, was not the posture in Vasquez Perdomo (or a number of other recent cases). Instead, the Court was being asked to act directly on the district court order issuing the original injunction. By rule, the government could not file that request until it had sought a stay in the court of appeals; but that requirement is effectively one of exhaustion. Unlike in a normal appeal or typical case on cert, Court was not being asked to act on the judgment entered by the court of appeals.

But was the original injunction still there to be stayed? Again, the court of appeals had stayed a key phrase in the center of the injunction. Perhaps that partial stay had the effect of revising the original injunction, somewhat in the way that a partial reversal can alter or revise a judgment issued below. Another analogy might be to the supposed “writ of erasure,” or the idea (now decidedly out of fashion) that injunctions can effectively revise statutes.

I disagree with that view of the partial stay. After the court of appeals ruled, there were two relevant legal decrees, not just one: there was a district court injunction and a separate partial stay issued by the court of appeals. Imagine that the partial stay had to be vacated, perhaps because the Ninth Circuit realized it had lacked jurisdiction to issue that relief. The appellate court would then dissolve its partial stay, and the automatic effect of the partial stay’s dissolution would seem to be the full implementation of the original district court injunction. The original injunction would have been there the whole time.

Even so, the Court had to consider the effects of the partial stay that had been issued by the court of appeals. To obtain a stay from the Court requires a showing of irreparable injury (among other things, such as a likelihood of success on the merits). And it is possible that a partial stay issued by the court of appeals might eliminate any such injury. So the Administration (or any other party seeking a stay from the justices) should have to show irreparable injury even after accounting for any stays issued by the court of appeals.

Was that requirement met in Vasquez Perdomo? The answer depends on what the partial stay in the court of appeals did. As we have seen, the Ninth Circuit purported to narrow the district court’s injunction, but it also thought that the injunction would continue to have significant implications. The plaintiffs seemed to take a similarly broad view, noting in their opposition filing at the Court: “the TRO does not meaningfully constrain the government at all, and contempt would never be an issue”, but only “[s]o long as the government is not violating the Fourth Amendment on a going-forward basis (as it claims it is not).”

The ambiguous but evident threat of after-the-fact review via contempt allowed the government to claim irreparable injury: because of the actual or threatened operation of the partially stayed district-court injunction, the government’s enforcement operations were impaired. And under recent (albeit disputable) precedent, that impairment qualifies as irreparable injury. So if the government could satisfy the other stay factors, such as likelihood of success, the result would be overlapping stays: a partial stay by the court of appeals and a complete stay by the Court. The justices may in fact have done precisely this.

Yet there is, or was, a more interesting possibility. Again, the court of appeals seemed to think that its partial stay brought the district court’s injunction into line with the first, narrow reading noted above. If that apparent understanding were credited, then the district court’s injunction would no longer have any significant effect. Virtually all executive-branch enforcement actions would, or at least could, involve an additional factor, apart from the four factors listed in the injunction. The Court could have said as much, effectively clarifying the injunction-as-partially-stayed. The result is ironic: the justices could have denied the Administration’s request for a stay on the ground that the district court’s injunction was already an effective nullity.

That disposition would have had the considerable benefit of being closely tailored to the specifics of this case. The Court would have made little if any new law, whether on standing, the Fourth Amendment, or anything else. Again, there would have been no need to reach the likelihood of success factor. At the same time, the Court would also have supplied an explanation, rather than leaving the lower courts and legal community to guess at what had happened. Narrow explanations are especially attractive these days, given the visible stresses placed on trial courts, the courts of appeals, and the Court itself. The justices might also have pointed out that this outcome was not simply due to some slip up by the district court but instead reflected the inherent difficulty of issuing programmatic Fourth Amendment relief, especially when expedited in a fast-evolving case involving factual nuance.

An irreparable-injury disposition also fits with the other claims that the government put forward in the case. The Administration’s stay application substantially involved arguments that the district court injunction was extra-jurisdictional, including because the plaintiffs lacked standing. And one of the government’s jurisdictional arguments resembles the one presented here: because the government rarely if ever relies exclusively on the listed factors, the plaintiffs assertedly lacked a sufficient risk of injury from enforcement based exclusively on those factors. Yet, in requesting a stay from the justices, the government’s threshold burden was to show that it was entitled to relief. Under the government’s own logic, the best answer was, No.

This no-irreparable-injury disposition may have been the best option available to the justices, or at least better than the unreasoned order that actually issued. But, as you can see, the argument in favor of that view is not without doubt. And the whole operation might reasonably be viewed as too clever by half.

If there is a broader lesson here, it is that the Court’s shadow / emergency / interim docket – perhaps we should just call it the “irregular” docket? – is creating not only new procedural complications, but also novel decisional options.

Thanks to several interlocutors for provoking or engaging some of the above ideas.