Gabriel J. Chin is Martin Luther King Jr. Professor of Law at UC Davis School of Law.
As is customary, the Court’s per curiam affirmance of the Fifth Circuit by an equally divided court was a one-line opinion. Although the ruling means that the injunction against DAPA remains in place, the case is some distance from being over.
What was at issue was the propriety of a preliminary injunction. As the Court held in University of Texas v. Camenisch, “where a federal district court has granted a preliminary injunction, the parties generally will have had the benefit neither of a full opportunity to present their cases nor of a final judicial decision based on the actual merits of the controversy.”
Thus, the states and the United States now head to trial. To be sure, in cases like this where most of the issues are purely legal rather than factual, it is likely that “the preliminary injunction was ‘the whole ball game,’ and [the lower courts’] analysis of the propriety of preliminary relief is applicable to any permanent injunction as well.”
But the Fifth Circuit’s injunction purported to apply nationwide. If, technically, the Court had affirmed the issuance of a permanent injunction, DAPA might well be dead, and this administration or another might well have to start from square one: Design and prepare for a new judicial test of a new plan, with all of the delay that would entail.
There are other long-shot alternatives. Perhaps the administration could go to district courts in the D.C., Second, or Ninth Circuit, seeking a declaration that the writ of a district court judge from Texas does not extend from sea to shining sea. But even if that would eventually work, it would take time. The administration probably should revise the policy guidance to remove the words which turned out to be a flashpoint, “lawfully present”; immigration specialists understand that lawful presence means, roughly, “likely unlawfully present, but tolerated by the Attorney General.” That subtlety turned out to be very hard to explain.
It is possible, though, that the Court missed an opportunity here to give some guidance. To be sure, most four-four affirmances are unexplained. They date back at least to The Antelope, a nightmarish 1825 case dealing with claims to possession of enslaved persons seized from a foreign vessel. Chief Justice Marshall explained:
Whether, on this proof, Africans brought into the United States, under the various circumstances belonging to this case, ought to be restored or not, is a question on which much difficulty has been felt. It is unnecessary to state the reasons in support of the affirmative or negative answer to it, because the Court is divided on it, and, consequently, no principle is settled. So much of the decree of the Circuit Court as directs restitution to the Spanish claimant of the Africans found on board the Antelope . . . is affirmed.
But The Antelope was not a one-liner.
In United States v. Texas, was there nothing upon which five or more Justices could agree? This case affects, potentially, the ability of millions of people to move out of the shadows, as well as the right of the president to exercise his authority to enforce the law. If so, there is precedent for affirming some piece of the case by an equally divided court, while issuing merits rulings on other questions. Thus, in American Electric Power Co. v. Connecticut, the Court was stymied on standing, but addressed other issues implicated in a lawsuit by states against carbon dioxide emitters. In United States v. Zolin, the Court was equally divided on the power of district courts to conditionally enforce IRS subpoenas, by imposing dissemination restrictions, but evaluated other aspects of their enforcement.
There is also an interesting, although dated, history of rehearings leading to a reversal of fortune. One would think that affirmance by an equally divided Court would come only after the Justices explored and rejected all other options, including ordering further briefing, reargument, holding a case over to the next Term – or doing more to persuade colleagues on the other side who had not yet seen the light. But four-four decisions do not necessarily represent true deadlocks. In Halliburton Oil Well Cementing Co. v. Walker, a patent case, an equally divided court affirmed in January 1946. The Court granted rehearing, and in November 1946, reversed with only Justice Harold Burton dissenting.
Similarly, in January, 1938, an equally divided court affirmed in Railroad Commission of California v. Pacific Gas & Elec. Co., a utility rate case, with Justice George Sutherland not participating. But after Justice Hugo Black replaced Willis Van Devanter, the Court reversed, six to two. By my count, at least one of Brandeis, Cardozo, Hughes, Stone, or, of course, Owen Roberts, had to have switched sides.
A potentially promising case for the United States is Indian Towing Co. v. United States,, a reversal initially affirmed by an equally divided court with Justice John Marshall Harlan II not participating. Harlan had been a member of the Court at the time of the initial decision, but sat out, perhaps because he had not heard oral argument. He turned out to be the deciding vote in the ultimate five-four decision. Thus, any movement toward the seating of a replacement for Justice Antonin Scalia might warrant rehearing.
On what I assume to be the merits, I remain puzzled. No one claims that the administration – the leading deporter in U.S. history – has failed to expend all available resources in carrying out the immigration laws. The issue is limited to the exercise of prosecutorial discretion.
There also seems to be no dispute that any given agency can permissibly issue: (1) legally binding guidance, which must go through APA notice and comment, (2) general, non-binding policy guidance, creating no enforceable rights, which need not go through notice and comment, or (3) no guidance, leaving it to individual officers and agents in the field to indulge their personal policy preferences. In “a government of laws, not of people,” option 3 would seem to be, by some distance, the least desirable.
The argument was that DAPA, drawing on the history of DACA, purported to be permissible non-binding category 2 guidance, but really was binding category 1 law, because it was likely to be followed in almost all cases. This argument seems silly. Sometimes, policy guidance is so sensible (e.g., “FBI Agents: Prioritize anti-terrorism investigations over investigations of people smoking marijuana in public”) that it will be nearly universally followed. Even so, the guidance could provide (as DAPA does), and courts should decide, that it creates no enforceable rights, and does not constrain the exercise of case-by-case discretion, if, in the view of the agency, there are reasons to depart from it. It is startling to see a case decided by courts based on the reasoning that government officials creating, understanding, and applying policy coherently and consistently is so unusual that something nefarious must be afoot.