Concept
Standard for Granting Relief
Preliminary injunction
Standard for Granting Relief
Rule 65(a) says when a court may issue a preliminary injunction, but not what the movant, the party asking for the order, must prove to get one. Rule 65(a)(1) provides that “The court may issue a preliminary injunction only on notice to the adverse party,”1 and Rule 65(a)(2) permits the court to consolidate the hearing with the trial on the merits. Neither provision states a substantive standard.2 Federal courts have filled the gap by reading Rule 65 to incorporate the common law requirements that traditionally govern equitable orders. Under those requirements the movant must show a likelihood of success on the merits, irreparable injury, a balance of hardships in its favor, and an injunction that serves the public interest.3
Those four factors are also the content of the court’s explanation. Rule 52(a)(2) provides that “In granting or refusing an interlocutory injunction, the court must similarly state the findings and conclusions that support its action,”4 so a judge who rules on a preliminary injunction motion states findings about likelihood of success, irreparable harm, the balance of hardships, and the public interest. A later section returns to the findings requirement itself. This section teaches what the findings are about.
Each factor guards against a different way an order entered before trial can go wrong. The order binds the defendant before any court has decided who is right, so the court first asks whether the plaintiff will probably win, since restraining a defendant who is likely in the right serves no purpose. It asks next whether the plaintiff faces an injury that a damages award could not repair, because waiting for final judgment is acceptable when the judgment can make the plaintiff whole. The balance of hardships then sets the plaintiff’s injury against the injury the order would inflict on the defendant, and the public interest factor looks past the parties to the people the order may affect. The circuits differ on how the four combine. Many require a threshold showing on each factor, so that failure on any one ends the inquiry, while the alternative approach lets a stronger showing on one factor offset a weaker showing on another.5 The last subsection compares the two approaches after the factors have been taught one at a time.
Likelihood of Success
The Supreme Court states the movant’s burden in Winter v. Natural Resources Defense Council: a plaintiff seeking a preliminary injunction “must establish that he is likely to succeed on the merits, that he is likely to suffer irreparable harm in the absence of preliminary relief, that the balance of equities tips in his favor, and that an injunction is in the public interest.”6 The Court calls the order “an extraordinary remedy.”7 Likelihood of success heads that list, so it is the first question a court asks. The next three subsections take up the other factors in turn.
The factor exists because of when the order issues. A preliminary injunction binds the defendant before any court has decided whose account of the dispute is correct, so a defendant who is in the right carries the restraint, and every cost that comes with it, for nothing. The order is justified only as a way to hold matters in place for a plaintiff who is likely to win at the end of the case. A plaintiff with a weak, doubtful, or unproved claim gives the court no reason to impose that cost, however sympathetic the threatened injury, and the court demands a real showing on the merits before it restrains anyone.
The showing has a floor and a practical ceiling. The floor is a prima facie showing of a cause of action against the defendant, meaning evidence that, if believed, would establish each element of the claim; the court cannot exercise its discretion over interim relief without it.8 The ceiling is higher. A temporary injunction can do substantial injury to the defendant even though it issues before the merits are determined, so courts award preliminary relief only when the plaintiff demonstrates a clear case of entitlement.9 A merely arguable claim can have a prima facie case without a clear one, and a court has little reason to restrain a defendant on an arguable claim before anyone has tested it at trial.
Disputed facts show the ceiling at work. When the facts on which the claim depends are genuinely contested, the court normally refuses the injunction.10 The hearing is shorter than a trial and often rests on affidavits, so the judge usually cannot tell which sworn version is true, and an order issued anyway would restrain the defendant on the strength of the plaintiff’s account alone. The appellate court in Atari Games Corp. v. Nintendo of America, Inc., an antitrust dispute, voiced the same caution: a district court should be wary of issuing an injunction based solely on allegations and conclusory affidavits submitted by the plaintiff.11 It pointed to Ninth Circuit decisions that reversed a grant resting on conclusory affidavits and that denied relief supported by only two affidavits.12 A conclusory affidavit hands the judge the plaintiff’s conclusion that the claim is good without the facts from which the judge could test it, so it cannot supply a clear case.
Courts word the factor differently, and the wording sets how much more than a prima facie case the movant needs. Winter requires that the movant be likely to succeed.13 The Ninth Circuit, as the Atari opinion recites, has long required a strong likelihood of success, yet it lets a movant carry its burden by showing either a combination of probable success and the possibility of irreparable injury, or that serious questions of law are raised and the balance of hardships tips sharply in its favor.14 The district court in Atari had granted relief on the ground that the alleged practices “raised serious questions going to the merits” of Atari’s antitrust claims.15 Likely, probable, strong likelihood, and serious questions are labels for different points on a range, and the final subsection explains how the alternative formulation lowers the merits showing when the other factors weigh heavily. In every version, the claim must look strong enough on an incomplete record to justify restraining the defendant before trial.
The standard also shapes what the judge must write. A court that grants or refuses an interlocutory injunction must state the findings and conclusions that support its action, so its conclusion on likelihood of success belongs in its stated reasons.16 The Atari court added that a district court need not make binding findings of fact but must at the very least find probabilities that the necessary facts can be proved.17 The finding, then, is a judgment about what the movant will probably be able to prove, made before the movant has had to prove it.
Students make two opposite errors on this factor. Some treat a colorable claim as enough. They have stopped at the floor, and a prima facie case against a sharp factual denial is not the clear case courts require. Others treat the factor as a prediction of final victory and demand too much, because the court judges likelihood on an incomplete record and need not make binding findings of fact.18 Even a strong merits showing does not carry the motion alone, since the movant must also show harm that money cannot repair, the subject of the next factor.
Irreparable Harm
The Supreme Court’s four-part statement in Winter places irreparable harm second: the plaintiff must establish that it is “likely to suffer irreparable harm in the absence of preliminary relief.”20 Irreparable harm is an irreversible, harmful change in circumstances that lies ahead and that money damages cannot compensate.21 Courts treat it as the most important of the four factors, and the applicant must show that the harm will probably occur before the court can decide the merits.22 An applicant who has an adequate remedy at law, a legal remedy such as a damages award that would fully compensate the loss, is denied the injunction.23
The first half of the factor, that money cannot repair the harm, comes from the nature of the remedy. An injunction is an equitable remedy, and equity has long refused to act when the plaintiff has adequate redress at law.24 The mechanism is easy to trace. An interim order restrains the defendant before any court has decided who is right, so a defendant who wins at trial will have carried the restraint for nothing. If a damages award at the end of the case would put the plaintiff where it would have stood anyway, the order adds nothing the judgment cannot supply, and its cost to the defendant has no justification. The same reasoning explains the tradition that interim injunctive relief is unavailable in actions where the plaintiff merely seeks money damages.25 The second half of the factor concerns timing. The order holds matters in place until trial, so it can protect only against harm that will arrive before the merits are decided and that the final judgment could not then undo.
A loss that sounds serious can still fail this factor when it is a sum of money in disguise. In one early case, speculators bought tickets to a baseball game that looked likely to sell out, planning to resell them at a substantial profit, and they sought a preliminary injunction to keep the club from refusing admission to the people who bought those tickets. The court denied the injunction because the speculators could sue for the price of any ticket the club wrongfully refused to honor, which gave them an adequate remedy at law.26 A house about to be torn down sits at the other pole. Once the walls are gone, no award of money returns the house, so a plaintiff seeking to preserve it shows exactly the irreversible change in circumstances this factor demands.27 The two situations differ in one respect: whether money can stand in for what is lost. The size of the loss does not decide that question, since the speculators’ profits may have been large and the house may be modest.
Timing can make the same point. If a city is about to enforce an ordinance against the plaintiff, the plaintiff may be ruined or may irreparably lose important rights unless an injunction issues immediately, and a full trial cannot be scheduled for months at best.28 A preliminary injunction can forbid enforcement until a trial is held and a final decision made.29 A damages award at the end of the case cannot give the plaintiff back what the interval destroyed.
Even a harm of the classic kind can fail when most of it has already occurred. In Narragansett Indian Tribe v. Guilbert, a tribe sought a preliminary injunction to stop completion of a dwelling. The First Circuit reasoned that although alteration of real property is a classic example of irreparable harm, the possibility of irreparable damage without an injunction was “very faint,” because site preparation was already extensive: trees and underbrush had been cleared, the foundation excavation was complete, and the invasive work for a septic system had been done.30 Given that transformation, the tribe failed to show that completing the house would do further irreparable harm.31 The court treated the strength of the other three factors as irrelevant, which shows how much weight this factor carries under the traditional test, where failure on any one of the four ends the inquiry.32 The last subsection shows how the alternative test changes that result.
Two mistakes recur. The first treats a large loss as an irreparable one. The question is whether money can repair the loss, and a large but calculable sum is the speculators’ situation, where the injunction failed. The second assumes that any injury to land satisfies the factor. Narragansett shows that a classic harm to land still fails when the land has already been transformed and the order could prevent no further irreparable injury.
Balance of Hardships
The balance of hardships asks which side stands to lose more if the court guesses wrong. The harm the plaintiff will suffer if the injunction is denied must outweigh the harm the defendant will suffer if it is granted.35 The court ordinarily balances the relative inconveniences to the parties, and if granting the injunction would do more damage to the defendant than denying it would do to the plaintiff, the request should be refused.36
The comparison exists because a court that rules before trial does not know who will win and can err in either direction. A mistaken grant restrains a defendant who turns out to be in the right, and the restraint’s costs were borne for nothing. A mistaken denial leaves a plaintiff who would have prevailed exposed to harm while the case proceeds. The court cannot eliminate the risk of error, so it asks which mistake would leave the smaller injury and favors the order that does less harm if it proves wrong. The Seventh Circuit frames the comparison in those terms in Turnell v. Centimark Corp., 796 F.3d 656 (7th Cir. 2015), weighing the irreparable harm the movant will endure if the injunction is wrongfully denied against the irreparable harm the nonmovant will endure if it is wrongfully granted.37
This factor differs from irreparable harm. Irreparable harm looks at the plaintiff alone and asks whether the loss is one that money cannot repair. The balance looks at both sides at once and asks which stands to lose more, so a plaintiff can show harm of the irreparable kind and still be refused, because preventing it would cost the defendant more than it saves the plaintiff. The court may also count injury to the public welfare in this calculation, and the next subsection treats that consideration as a factor of its own.38
The same comparison comes out the other way when the larger loss sits on the plaintiff’s side.
What moved the answer in the two scenes was the relative size of the losses. A student who stops after finding that the plaintiff will be hurt has answered half the question, because the factor also requires a finding about what the defendant stands to lose. That finding belongs in the court’s explanation of its decision, since a court granting or refusing an interlocutory injunction must state the findings and conclusions that support its action.39 The balance does not always operate as a free-standing hurdle, either. In some circuits, how heavily the comparison must favor the plaintiff depends on how strong the plaintiff’s showing on the merits is, a link the last subsection takes up.40
Public Interest
The last element of the showing in Winter is that the movant establish “that an injunction is in the public interest,” and the Court adds that courts must “pay particular regard for the public consequences in employing the extraordinary remedy of injunction.”42 Irreparable harm and the balance of hardships measure injury to the people standing in the courtroom. This factor looks past them. An injunction runs against the parties, but a restraint on a hospital or a lumber mill changes things for many people who never filed anything and cannot object. If the court counted only the parties’ injuries, nobody would speak for those people, and an order that is fair as between plaintiff and defendant could still do harm that neither side has any reason to mention. The factor therefore principally takes into account the interests of non-parties, and courts often frame it negatively by asking whether granting the order would be adverse to the public interest.43 The hardships balance can also take in injury to the public welfare, so a public consequence may surface under either label, and an answer should credit it wherever the court puts it.44
Because the interest belongs to people who are absent, the law offers no single test for finding it. The notion of public interest has no uniform or constant formulation, and a court may find it in a statute, in the nature of the subject matter in dispute, or in a judicial determination that topics such as protecting the environment or civil rights are traditionally seen in a favorable light.45 In Woodhouse, a court reasoned along the last of those lines and found the public interest favored a preliminary injunction requiring the state to permit campaign contributions consistent with Equal Protection requirements.46 A party arguing this factor therefore has a task of proof: it must point the court to the statute, the subject matter, or the settled view of the topic that shows where the public stands.
A tempting shortcut treats a favored subject as a winning card, so that an environmental plaintiff is assumed to prevail on this factor. Two logging cases show why the shortcut fails, because the defendants in each could point to public benefits of their own. In League of Wilderness Defenders/Blue Mountains Biodiversity Project v. Connaughton, environmental organizations sought a preliminary injunction to halt timber-cutting in mature forests, alleging violations of federal environmental statutes. The defendants argued that an injunction would be adverse to the public interest because the logging would reduce the risk of forest fires and insect infestation and would bring jobs, school funding, and improved health services to the local community. The court granted the injunction, determining that a countervailing public interest in preserving an elk habitat in the forest outweighed the other considerations.47
In Alliance for Wild Rockies v. United States Forest Service, environmental organizations challenged logging operations and a road construction proposal, and the court denied injunctive relief. It found a strong public interest in increased economic and recreational benefits for the local citizenry, in watershed restoration, and in reduced wildfire risk.48 The two decisions involve the same kind of dispute and opposite results, and the difference lies in the weighing. In the first, the habitat interest outweighed the fire, insect, and economic points the defendants raised. In the second, the interests favoring the project were strong enough to defeat the injunction, and one of them, watershed restoration, is itself an environmental interest. No subject carries a fixed weight, so the outcome on this factor depends on the competing interests each side can establish on the record before the court.
The third step does the real work, since it is where a student who has found a public interest on only one side goes wrong. A favorable showing here also does not carry the motion alone, because the movant’s showing under Winter has four parts, and the final subsection takes up how the circuits combine them.50
Traditional and Sliding-Scale Tests
The four factors appear, in one wording or another, in every formulation of the standard. The Supreme Court’s statement in Winter names them,52 the traditional common law test lists them,53 and the Ninth Circuit’s version, as recited in Atari, tracks them.54 The circuits differ on how the factors relate to one another. They can operate as independent thresholds, or a court can weigh them against each other on a sliding scale.
The traditional test makes each factor a separate hurdle. It sets a threshold for each of the four, and the plaintiff’s failure on any one ends the inquiry.55 Many federal circuits and most state courts follow this approach.56 The reason lies in what the order is. A preliminary injunction is “an extraordinary remedy,”57 and each factor answers a separate objection to imposing it before trial. Without a likely win, the court restrains a defendant who is probably in the right. Without irreparable harm, the plaintiff’s loss can wait for a final judgment. Without a favorable balance of hardships or public interest, the order costs more than it protects. A court that let overwhelming strength on one factor erase a failure on another would issue the order with one of those objections unanswered.
Narragansett shows the traditional test at its most unforgiving. The tribe failed on irreparable harm because the land had already been cleared, the foundation excavated, and the septic system installed, and the court treated its showing on likelihood of success, hardships, and public interest as irrelevant.58
The alternative test, usually called the sliding scale, keeps the same four factors but lets the court trade among them, so that a weakness on one can be offset by a stronger showing on another.59 The Ninth Circuit’s version, recited in Atari, allows a movant to carry its burden by showing either a combination of probable success on the merits and the possibility of irreparable injury, or that serious questions of law are raised and the balance of hardships tips sharply in its favor.60 The second route matters most to a plaintiff with a doubtful claim, because it replaces a probability of winning with a serious legal question and asks the balance of hardships to do more work. The harder the balance tips toward the movant, the less certain the movant’s win needs to be.
The Seventh Circuit reaches a similar place by a different path. Under the two-step analysis in Turnell, the movant first makes a threshold showing that includes irreparable harm and a likelihood of success, and the court then weighs the irreparable harm each side would suffer from a wrong decision, together with the public interest.61 The scale slides in the sense that the more likely the movant is to win, the less the balance of harms must weigh in the movant’s favor, and the less likely the movant is to win, the more heavily it must.62
The majority in American Hospital Supply Corp. v. Hospital Products Ltd. went further and put the same trade-off as an inequality: the district courts should grant a preliminary injunction “if, but only if,” P × Hp > (1 − P) × Hd, where P is the probability that denial of the injunction would be an error, Hp is the harm to the plaintiff if the injunction is denied, and Hd is the harm to the defendant if it is granted.63 A dissenting judge objected to reducing the law of preliminary injunctions to a simple mathematical formula, and the objection is fair.64 The formula is still useful as a picture of why a weak claim demands a lopsided balance. If the plaintiff has a 40 percent chance of winning and would lose $100,000 from a denial, while the defendant would lose $200,000 from a grant, then 0.4 × $100,000 = $40,000 is less than 0.6 × $200,000 = $120,000, and the injunction should be denied.
The two approaches diverge mainly at the edges. A plaintiff with a weak case on the merits but facing grave harm may win under the sliding scale where the traditional test would refuse relief. A plaintiff with a strong claim may win on a lesser showing of irreparable harm and balance of hardships, although such a plaintiff must still prove irreparable harm as a threshold matter.65 Narragansett invites the comparison. Had the alternative test applied, the tribe’s case would have turned on whether it could make a sufficient threshold showing of irreparable harm. If it could, a high likelihood of success on the merits, that the house infringed on the tribal reservation, might have offset the low showing of harm.66
The live question after Winter is whether the alternative test survives it. The Court’s formulation reads like four separate requirements: the movant must establish that it is likely to succeed, that it is likely to suffer irreparable harm, that the equities tip in its favor, and that an injunction is in the public interest.67 Whether courts using the alternative test may still issue an injunction on a serious question on the merits, when the movant shows grave irreparable harm and a balance of equities that tips heavily in its favor, remains an open question. The Second, Seventh, and Ninth Circuits have concluded that the flexible test survives; see, e.g., Alliance for the Wild Rockies v. Cottrell. The Fourth Circuit has concluded that Winter precludes the alternative test and that the movant must establish a likelihood of success.68
Two mistakes recur. Students treat the traditional test as a weighing exercise and let a strong merits case rescue a missing showing of irreparable harm, when the test makes each factor an independent hurdle. Others assume the sliding scale lets a movant skip a factor entirely, when irreparable harm stays a threshold requirement even for a plaintiff with a strong claim. Whichever formulation governs, the court granting or refusing the injunction must state the findings and conclusions that support its action, so those findings will track the factors that formulation requires.69 On an exam, follow the circuit’s formulation if the facts supply one, and address all four factors either way.