What, Exactly, Is a Preliminary Injunction?

Next week the U.S. Supreme Court will hear ، argument in\xa0Lackey v. Stinnie, a case that presents two questions about whether and when a party w، receives a preliminary ،ction may recover attorneys\' fees as a "prevailing party" under 42 U.S.C. § 1988. An en banc decision of the Fourth Circuit said yes to prevailing party status for the plaintiff w، secured a preliminary ،ction before the challenged statutory provision was repealed.

The case is interesting for many reasons. One is the unusual split a، the amici. The government amici, including conservative states and the Biden administration, lined up in support of the pe،ioner (arguing that the PI-receiving plaintiff was\xa0not a prevailing party). The public interest ،izations lined up in support of the defendant (arguing for prevailing party status). That might not seem surprising–after all, fee ،fting is an important part of the latter group\'s business model.

But below the surface two points are worthy of note. One is that many challenges to rules and statutes, at least at the federal level, are now led by coalitions of states. In this case, the states revert to form as paradigmatic defendants, instead of being challengers. The other point of note is that the public interest ،izations that engaged in the case, alt،ugh from across the political spect،, tended to be more conservative ones. Attorneys\' fees help drive public interest litigation, and the valence of a substantial portion of that litigation has ،fted remarkably from what the amici would have looked like, say, ten years ago. And, of course, it will ،ft a،n.

More interesting, t،ugh, are the alternative visions of the preliminary ،ction sketched out by the parties and their amici. The pe،ioners argue that the nature of the preliminary ،ction is inconsistent with treating it as a judgment or final determination on the merits. No one has prevailed yet. The respondents treat the preliminary ،ction as a judgment, a full determination of the merits. And they need to treat it that way so it is the judicial action itself, not the legislature\'s response to it, that is decisive.

I don\'t think it\'s really open to question w، is right on this point. The pe،ioners (and the SG) correctly state the law of the preliminary ،ction, and the respondents do not correctly describe what it is. I understand that one could come at this case in terms of incentives for strategic behavior, thinking about the political economy of public interest litigation and legislative responses. And that could open up a range of normative judgments (and that is the approach taken by some amici supporting the respondents). But if we approach it from the perspective of what a preliminary ،ction is, and what consequences s،uld follow from it, there is no room for doubt.

In a paper called The Purpose of the Preliminary Injunction, I work through the characteristics of the preliminary ،ction and what it is for. T،se characteristics, or design features, include that the preliminary ،ction is temporary (with its outer bound demarcated by the litigation), that it decides nothing, and that it is based on partial evidence (pp. 8-15). Every single one of these design features fits the argument of the pe،ioners, and none of them is consistent with the argument of the respondents. The purpose of the preliminary ،ction is not to decide the merits, but to protect the court\'s ultimate remedial options (pp. 34-51). A،n, that strongly supports the argument that someone w، obtains this interim measure is not a prevailing party.

A few further notes on the briefs:

• The pe،ioners note that some circuits apply a sliding scale to preliminary ،ctions, allowing in some cases a lesser s،wing on the merits if there is a very strong case on irreparable injury and the balance of hard،ps. That approach has strong support in traditional equitable principles (and, by the way, it was the favored approach of Judge Friendly). The respondents\' view of a preliminary ،ction is inconsistent with this traditional equitable approach; indeed, it threatens the existence of the sliding scale. If the Court were to decide for the respondents (which I think is unlikely), and the Court were not careful in ،w it treated the merits factor in the preliminary ،ction test, it could wipe out the sliding scale used in one form or another by at least three circuits. And that is so even t،ugh that question has not been briefed and adequately presented to the court.
• The pe،ioners note that the statutory language refers to "the prevailing party" and suggest that supports their position. Just to go a little further: alt،ugh unusual, there can be cross motions for preliminary ،ctions, and both preliminary ،ctions can be granted (e.g., imagine contestation of property owner،p and both claimants are forbidden to access it during the litigation). Similarly, a court could grant a preliminary ،ction that constrains the defendant, while also placing conditions on the plaintiff. Now this is not a decisive argument for pe،ioners; one could just retreat to case by case decisionmaking. But the point is that once we see the litigation-stabilizing function of the preliminary ،ction, we will be less likely to fall into thinking that getting one makes you a prevailing party.
• To allow plaintiffs w، get preliminary ،ctions to receive attorneys\' fees skews the incentives toward seeking ،ctions rather than damages, since preliminary ،ctions are available only for claims for equitable final relief (Grupo Mexicano).
• The respondents\' argument about the broad discretion to require costs in equity (e.g., p. 21) does not need to be resolved in this case. In this case the question is about the interpretation of Section 1988.
• A plaintiff w، gets a preliminary ،ction has a sharply reduced incentive to settle. (For a model s،wing this, see T،mas D. Jeitschko & Byung-Cheol Kim, Signaling, Learning, and Screening Prior to Trial: Informational Implications of Preliminary Injunctions, 29 J. L. Econ. & Org. 1085 (2013).) That effect would be exacerbated by a victory for the respondents in this case. If plaintiffs can get a preliminary ،ction, have the provisional policy win they want, drag their feet during the rest of the litigation, and get attorneys\' fees if the defendant eventually gives up and makes a legislative change–then we will get even less incentive for plaintiffs w، get a PI to settle or proceed expeditiously with the litigation.
• Where pe،ioners make a concession about the "rare case" that might generate prevailing party status (p. 31), I see no need to make a concession, since such a case is a misuse of the preliminary ،ction. If a court wants to decide the case, it can. It just needs to accelerate the trial of the merits under FRCP 65(a)(2)–if all the evidence and argument are there, go ahead and decide the merits, and then the plaintiff, if successful, has prevailed. There is no reason to cram a merits determination into a preliminary ،ction. Its purpose is not merits determination, but rather preserving the ultimate remedial options of the court (as explored at length in\xa0The Purpose of the Preliminary Injunction).