Government by Injunction

We’ve all imagined what we would do if we were king for a day. Even federal judges. Remember the joke, “Question: What’s the difference between God and a federal judge? Answer: God knows that he’s not a federal judge.” Well, for some federal judges, that’s no joke; they genuinely don’t know the difference.

President Donald Trump has decided (among other things) to enforce federal immigration laws and remake the administrative state. Not everyone is happy. Numerous lawsuits—almost more than “the stars of the heaven” or “the sand which is upon the sea shore”—have been filed to slow or halt some of his policies and practices. So far, he has had some successes and some failures, but we are still in the early innings, with more to play.

Consider a current, trendy practice that some federal district court judges have applied with alacrity: issuing so-called “nationwide injunctions” outside the context of a properly certified nationwide class action. Those orders bar the President from implementing a law or policy in a certain manner not only against the plaintiff who won his or her lawsuit, but against everyone else, whether in this nation or elsewhere, even though nonparties are, by definition, strangers to the litigation that gave rise to the injunction. Some judges have even issued those injunctions at the earliest possible stage of a case: when a plaintiff asks a judge to enter a temporary restraining order (TRO) against the federal government (usually at 4:55 pm on a Friday) because his client not only is likely to prevail when the judge finally resolves the case, but also needs immediate injunctive relief to avoid the return of Vigo the Carpathian before the government can even be advised that there’s a lawsuit afoot. It’s the legal version of asking a parent for help with a term paper due that day just before leaving for the school bus.

Injunctions were a traditional form of relief in England when this nation was founded. Since then, plaintiffs have sought and obtained injunctive relief if they can persuade a judge that such immediate relief is legitimate and critical to prevent the government from irreparably harming them. What makes nationwide injunctions novel is that, nowadays, a plaintiff (read: the public interest lawyers representing the carefully chosen plaintiff) seeks far more than protection from the government for himself. No, the plaintiff wants to halt an entire government program applicable to him and thousands of others. To do so, the plaintiff importunes a federal district court judge to stop that program in its tracks nationwide. You can fund or free Willy and all of his friends if you can persuade one federal district court judge (to mix movies, not metaphors) to freeze a federal program in carbonite until a higher court sets aside that injunction or Congress revises the statute at the bottom of the dispute.

Some might say that this is just whining over nothing. As its name provides, a  TRO is temporary. It may last for no more than 14 days; a district court judge can lift it at any time before that period expires; and, if the court leaves it in effect, once that two-week period expires, it becomes a preliminary injunction that an appellate court of appeals can review and set aside if need be. At its worst, therefore, a TRO can cause only a little delay in implementing a federal program. What’s the big harm?

Start with the most obvious harm: financial. The Trump Administration has delayed the issuance of some federal grants so that the new administration’s officials at the Department of Government Efficiency can decide whether a variety of grants made by the Biden Administration were fraudulent or unwise and should be clawed back, grants such as the following: “(1) $56 million to boost tourism in Tunisia and Egypt;” a “$20 million grant to create an Iraqi version of ‘Sesame Street;’” “$2.5 million to advance Diversity, Equity, and Integration (DEI) in Serbia’s workplaces;” “$2.5 million to build electric vehicle charging stations in Vietnam;” “$2 million “for promoting sex changes in Guatemala;” “$500,000 for the expansion of (inter alia) atheism in Nepal;” “$70,000 to produce a DEI musical event in Ireland;” “$47,000 for a transgender opera in Colombia;” “$68,000 for dance classes in Wuhan, China”; and “$32,000 for a transgender comic book in Peru.” Likely 90 percent of Americans (or 99 percent of those not civilly committed or registered Democrats, pick ‘em) are irate that the Biden Administration took their money by force then to shovel it out the door for such inglorious activities. Trump Administration officials are certainly entitled to ask whether $56 million is better spent on food or ammunition for our military than on tourism anywhere, even here. That inquiry is doubtless a legitimate one.

The problem, however, is this: If a district court enters a TRO ordering the federal government to hand over funds to the plaintiff organization, the government might never get that money back even if the plaintiff was not entitled to receive it. Suppose the plaintiff is a non-governmental organization whose mission is to distribute federal funds to people in foreign lands, perhaps ones like Yemen, which lacks a central government and is in the midst of a civil war, chaos, or both. Once the plaintiff obtains a TRO, it might try to “win” the case by immediately disbursing those funds before the TRO expires. If the plaintiff has not been required to post a bond so that the government can ultimately recover that money if the plaintiff loses—which Rule 65(c) of the Federal Rule of Civil Procedure requires, but district court judges do not always demand—those funds are like one of Aaron Judge’s homers: gone, forget it, goodbye. Good luck trying to claw back money paid out to anyone in Yemen. I’m sure that the Yemeni courts (assuming that there are any) would be very receptive to a claim for relief filed by lawyers representing the nation that is bombing it as I write.

But that could be the least of the government’s problems. Stick with the above example. The ostensibly poor, needy, harmless people in Yemen seeking federal funds could turn out to be fronts for the Iran-backed Houthis, who are trying to sink ships that traverse the Red Sea, as well as kill Israelis with their rockets. (Or terrorists could wind up stealing the money from the needy. It’s happened before.) In such a case, there are surpassing foreign policy interests at stake on the government’s side of the balance, as well as the lives of people aboard the targeted Red Sea vessels and in Tel Aviv. Any funds that make their way into the hands of the Houthis, or anyone else up to no good, enable a plaintiff organization to fund illegal, terroristic, or murderous action. So, more than just a 14-day delay is at stake even if the government ultimately gets its—no, my mistake, our—money back.

The practice of issuing nationwide injunctions has little English or American common law pedigree. At most, there was a case here and there since 1939. The pedigree—and frequency—of such orders truly began in the George W. Bush Administration, but each of his successors has also suffered under them too. Yet, that practice has matured like a salp on anabolic steroids. There are several explanations offered in its defense, such as the desire by judges to achieve uniformity in the interpretation of the law, consistency in its application to multiple parties, and efficiency in the adjudication of new cases. None of those rationales, however, justifies disregarding the constitutional law that reveals the practice’s illegality.

Start with the basics: the Convention of 1787. The Framers spent the bulk of their time in Philadelphia discussing and designing the Article I branch—Congress. They defined the architecture of that institution, they specified who may hold office under it, they limited the tenure of whomever the citizens elected (or the states chose) to represent them, and they defined how it would exercise its delegated authority. The Framers debated those factors with care because they saw Congress as being the most powerful element of the new central government, and therefore the one most likely to cause trouble. The Framers gave far less attention to the Article II and III branches because, while important, the Constitution grants each one far less authority than Congress, which means that what they are empowered to do is far less threatening. 

Article I vests “[a]ll legislative Powers” in Congress, which Congress may exercise (ordinarily with President’s assent) by producing a “Law”—viz., a binding rule of conduct generated by a legitimate source of political authority and backed by governmental power. That term—“Law”—is highly significant because of how it appears in that article as well as the other two. Article II vests in the President the “executive Powers,” which includes the responsibility to “take Care that the Laws be faithfully executed,” by himself or by using one of the “Officers of the United States” that he may appoint. Article III vests the federal judiciary with the “judicial Powers,” which includes the power to preside over “Trials” that raise issues “in Law and Equity, arising under” the Constitution, one of Congress’s laws, or a treaty, but only as part of a “Case” or “Controversy.” The term “Law” or “Laws” appears in each of the first three articles, but has a different meaning in each one. Congress may create them, the President may execute them, and the judiciary may interpret them.

That distinction is important. Article III offers no hint that the “judicial Power” vested in the federal courts over the parties in “Cases” and “Controversies” is identical to the “legislative Power” vested in Congress over the nation. Nor does Article III imply that, in a “Trial,” a federal court may enter a judgment that is comparable to the “Law” that only the Congress (and President) may create. The Framers certainly knew the difference between the legislative and adjudicatory processes because there were English governmental institutions—such as the King’s court, the Privy Council, and the House of Lords—and at least one in the Colonies—the New York Council of Revision—that combined one or more legislative, executive, and judicial powers into a single entity. Yet, persuaded by Montesquieu’s belief that a government of divided powers was less likely to become autocratic, the Framers rejected the English and early American approaches and assigned different powers to each of the three newly created branches.

The Supreme Court of the United States has found that choice illuminating, compelling, and binding. In a long series of cases, the Court has concluded that the text of Articles I, II, and III assign federal governmental powers and responsibilities to specific branches and that the political branches cannot vary from the now 230-plus-year-old architecture that the Philadelphia Convention proposed and that the states adopted. Congress cannot grant itself a legislative veto regardless of how efficient that twentieth-century supervisory device might be. Congress may not delegate executive authority to the Comptroller General, an official appointed and removable only by Congress. Congress cannot (generally) grant the “judicial Powers” to a decisionmaker lacking the tenure and salary protections granted only to Article II judges. And so forth.

Two Supreme Court decisions are directly applicable to this issue. One is Williams v. Zbaraz. Zbaraz held that there is no “Case” or “Controversy” between the parties to a case—there, parties challenging abortion funding restrictions and the federal government—over an issue that is not in the case—the Hyde Amendment, which prohibits federal funding of abortions unnecessary to save the mother’s life. There is certainly no “Case” or “Controversy” between a party and a stranger to litigation. A nonparty must file a lawsuit and prevail for an Article III judge to exercise federal judicial power and award that party relief. The other case was United States v. Mendoza. Mendoza ruled that the federal government is not subject to the “Win or go home” consequence of litigating a case. That is, while the government is bound by the judgment entered in favor of a prevailing party in a lawsuit, strangers to that litigation cannot prevent the government from relitigating its liability in future cases. (The technical explanation, for the lawyers out there, is that nonparties cannot make affirmative use of the doctrine of collateral estoppel or issue preclusion to bind a party to the resolution of an issue it once lost.) The reason is that applying any such rule of preclusion harms the public by impeding reasonable development of the law.

Handcuffing the government to a loss prevents it from relitigating the issue in different cases decided by different judges, perhaps coming at the issue from different perspectives. That slow-but-sure, common-law-like process of case-by-case adjudication increases the likelihood that every aspect of a legal question, along with every benefit and cost to the law and life from resolving it one way or the other, will be fully aired before the Supreme Court might need to resolve it for the nation.

That last point also is significant for another reason, one that relates to the architecture of the federal judicial system. At common law, there was but one judge who could award injunctive relief, a Chancellor, who sat in a court of equity for the entire nation. By contrast, the Framers created a judicial system with more than 670 district court judges, each of which has authority over both “Law” and “Equity.” Above those judges are more than a dozen federal circuit courts of appeals, with the Supreme Court sitting atop the entire system. Given its place at the apex of the federal judicial system, a Supreme Court ruling sets the law for each federal court below it. That is one of the roles that the Supreme Court serves: to provide a definitive ruling for the nation on a question of federal law. A district court judge who issues a nationwide injunction effectively elevates himself to the same rank in the federal system that the Supreme Court occupies. That certainly is a most peculiar result. (It also reminds me of a remark that a law professor of mine made about the New York state judicial system’s practice of denominating trial-level courts as “Supreme Courts,” which he said tells us something about New Yorkers. But I digress.) No district court judge in Maine can bind the action of a district court judge in Alaska. Nor can the First Circuit (which includes Maine) bind the Ninth Circuit (which includes Alaska). That result was intentional because it fosters consideration of issues by multiple judges and courts, which increases the likelihood of the right or best outcome. The nationwide injunction practice applied by some courts—which is not authorized by any act of Congress—is a bizarre aberration to the systematically organized federal judicial system architecture that Congress has created. That is yet another powerful argument against the legitimacy of this practice.

If it were necessary to offer a sound policy argument against nationwide injunctions, there are four of them at hand. First, nationwide injunctions encourage judge shopping. Some divisions within federal judicial districts have but one judge, so filing a case there guarantees who will decide it, thereby defeating the benefit of the random assignment of cases to judges to prevent forum shopping. Second, judge shopping leads to further politicization of the courts. Much of the public already sees little difference between the actions of the Article I and II political branches and those of the Article III courts, but that cynicism could turn fatal from blessing the nationwide injunctions practice. Third, the practice raises the possibility that different courts could issue different, even conflicting, nationwide injunctions. That would create an administrative mess until the inconsistency is resolved, which benefits no one. Fourth, the practice would weaken the doctrine of stare decisis. If the Supreme Court cannot wait until several circuit courts of appeals have resolved an issue before taking up a case to decide it themselves, the Court will issue some rulings that it later finds mistaken or regrets handing down for some other reason. Take away the front-end opportunity to wait for the views of other courts, and you force the Supreme Court to wash, rinse, and repeat its consideration of a subject at the back end, thereby eroding the certainty that a final Supreme Court decision affords everyone. In sum, the practice of issuing nationwide injunctions is mistaken as a matter of law and misguided as a matter of policy.

Does everyone hate them? No. Any organization that hates the policies of a newly installed presidential administration—yes, folks, some people on the extreme Left and Right hate each other with a passion not seen since the last Yankees-Red Sox game—and wants “one ring to rule them all” might find nationwide injunctions desirable to the delay occasioned by repeated litigation. For the rest of us, however, not so much. The traditional slow-but-steady way to resolve controversies makes up in the long run what it might cost in the short run. Hopefully, the Supreme Court will make that point soon.

‍Paul J. Larkin is the John, Barbara & Victoria Rumpel Senior Legal Research Fellow at The Heritage Foundation.