Welcome back to Ninth Circuit Review-Reviewed, your monthly recap of administrative law before arguably “the second most important court in the land.” Let’s get straight to business.

Preliminary Panels: Politics by Other Means?

In normal times, injunctive relief is an “extraordinary remedy.” Over the last four years, by contrast, such relief became routine in the Ninth Circuit. While I don’t have the statistics, I’d be willing to bet the house that district courts in CA9 issued an unprecedented number of preliminary injunctions against the federal government during the Trump era (relative to prior presidential terms). 

On being enjoined by a district court, the government usually sought a stay pending appeal (of the PI) before a CA9 *motions* panel. Each month, three judges are assigned to this panel, which reviews applications “for procedural, summary, or discretionary relief,” including stay requests, petitions for interlocutory appeal, applications to file successive habeas petitions, among other things. On its own discretion, a motions panel may publish its opinion in the Federal Reporter. (For the nitty gritty details on motions panels, see CA9’s General Orders Chapter 6). 

In weighing stay requests, motions panels apply the Nken v. Holder framework, which requires the movant to demonstrate:

• a strong likelihood of success on the merits; 
• a showing of irreparable harm;  
• the balance of the equities favors a stay; and, 
• the public interest favors a stay. 

After resolution of the government’s stay request, the (typical) next step in these controversies is for the government to appeal the district court’s PI before a three-judge *merits* panel of the Ninth Circuit. For these PI appeals, CA9 reviews district courts for abuse of discretion. The district court’s decision making, in turn, is judged against the Winter v. NRDCframework for granting injunctive relief. Under Winter, a plaintiff seeking a preliminary injunction must establish:

• likelihood of success on the merits; 
• a showing of irreparable harm;
• the balance of equities favors a PI; and,
• the public interest favors a PI. 

For both the Nken and Winter frameworks, note that the leading factor is the likelihood of a party’s success on the merits. That is, both preliminary panels—involving requests for stays & PI appeals—address the same prospective legal question. Moreover, both preliminary panels can publish their legal opinions in the Federal Reporter. 

This brings me to the question that animates this post: What happens when the second preliminary panel disagrees with the first panel’s published opinion? By its nature, ascertaining the “likelihood” of success on the merits is an ambiguous endeavor, one that is amenable to differing opinions. 

Before Trump, I surmise that this sort of disagreement rarely occurred in a controversial context, such that the court rarely thought about it. But when it did lend thought to the matter, CA9 indicated that a published opinion becomes law of the circuit, period. Six years ago, in Lair v. Bullock, the court observed that “a motions panel’s published opinion binds future panels the same as does a merits panel’s published opinion.” 

In any case, the Trump administration’s immigration agenda forced the court to squarely address the question. Immigration policy, of course, is a values ladened subject, and the Ninth Circuit’s ideological composition became more polarized as it took in 10 new judges. In these circumstances, the odds became near certain that preliminary panels would come to opposite conclusions on the law. 

Curiously, the matter came to a head in an immigration controversy featuring harmony between the preliminary panels. Last February in East Bay Sanctuary Covenant v. Trump, a merits panel addressed an appeal of a PI brought by the government. Two years prior, a motions panel had denied the government’s stay request in a lengthy published order. Despite its agreement with the prior panel, the merits panel nonetheless chose to clarify the “the effect of the motions panel’s order on the present panel’s decision.” According to merits panel, the court’s statement in Lair v. Bullock had been “dicta.” Instead, the panel endorsed a “pragmatic” approach that “treat[s] the motions panel’s decision as persuasive, but not binding.” Judge Ferdinand Fernandez wrote a concurrence, agreeing with the result because he believed that the panel was doctrinally bound—by law of the case & law of the circuit—to abide the prior panel’s legal conclusions. That is, Judge Fernandez rejected the court’s “pragmatic” approach as being inconsistent with Lair v. Bullock and other precedents. 

On the same day the Ninth Circuit published East Bay Sanctuary Covenant v. Trump, that opinion was applied as precedent in a different immigration controversy—Innovation Law Lab v. Wolf. Here, however, the preliminary panels disagreed. Previously, a motions panel had determined that the government was likely to win on the merits in a published order granting an administrative stay. Yet the merits panel came to the opposite conclusion on hearing the government’s appeal of the PI. Regarding its latitude to depart from the motions panel’s published opinion, the court said that “we follow East Bay Sanctuary Covenant v. Trump,” which, again, was issued on the same day. Judge Fernandez dissented, raising the same arguments as in his concurrence in East Bay Sanctuary Covenant.

Fast forward to last month, which witnessed two more applications of the new East Bay Sanctuary Covenant principle. On New Year’s Eve, in Doe #1 v. Trump, a split panel granted the government’s PI appeal, even though a split motions panel had denied the government’s request for a stay in a published order. According to the second preliminary panel, “we do not view the motions panel opinion as precluding us from re-examining the merits of the issues afresh in light of the now-completed merits briefing and argument.”

Earlier, on December 2^nd, a merits panel performed the opposite flip-flop in San Francisco v. UCIS, yet another immigration controversy. Here, a split panel denied the government’s PI appeal, even though a split motions panel had granted the government’s stay request in a prior published order. 

Interestingly, neither Doe #1 v. Trump nor San Francisco v. UCIS makes any reference to East Bay Sanctuary Covenant. Instead, these merits panels independently distinguished the prior published orders with which they disagreed. 

So, in 2020, there were (at least) three instances where preliminary panels published contradictory opinions in the same case. That’s weird! 

Last month, I wrote about gamesmanship with respect to the Ninth Circuit’s en banc procedures. I wonder if something similar isn’t occurring with respect to preliminary panels. 

As a matter of first principles, what compels a motions panel to publish a lengthy legal analysis in deciding on the government’s request for an administrative stay? Before East Bay Sanctuary Covenant, did motions panels try to “lock in” subsequent merits panels by publishing its opinions? Or has the impetus always been to establish a record defending the panel’s decision in a subsequent proceeding before the Supreme Court’s “shadow docket”? I don’t know. But I think the court should be asking itself these questions. 

At the very least, it seems like a silly waste of resources for multiple panels to write conflicting opinions regarding the same prospective legal question in the same case. After all, CA9’s massive backlog accounts for nearly one-third of all pending federal appeals—it’s not as though the court has time & energy to spare. 

William Yeatman is a research fellow at the Cato Institute.