Last night the Supreme Court refused to stay a three-judge district court order which required North Carolina to come up with a new redistricting plan within two weeks which did not constitute a “racial gerrymander” violating the equal protection clause. The court issued the order when absentee balloting had already begun. It would require a new election under the new rules. I had predicted a stay of the district court’s order, based upon what I term in an upcoming paperthe Purcell Principle. Looking at the pattern of the 2014 emergency orders issued before the 2014 election in cases from Ohio, North Carolina, Texas, and Wisconsin, it appeared the Supreme Court stopped courts from changing the rules of the game just before the election. So the Court said courts could not stop Texas from using its voter identification law before the election, but could stop Wisconsin, who was freed by the 7th Circuit to immediately use its law which had been put on hold. The difference appeared to be that Texas had already been using its voter id law but the id law had not been rolled out in Wisconsin before the 2014 election.
Talking about the Purcell principle (whose name comes from an earlier case where the Court stopped the 9th Circuit from putting a last minute stop to implementation of AZ’s voter id law, which was already underway) is guess work because these orders form part of what Will Baude calls the shadow docket: the Court decides these cases generally without issuing opinion or reasoning, and so we have to use concurrences, dissents, and other clues to figure out what’s really going on.
So was the Supreme Court’s decision to allow this last minute change to NC’s congressional elections a repudiation of the Purcell Principle, as some have suggested? Perhaps.
To begin with, as I noted earlier this week, Justice Scalia’s death may have changed the calculation. I had assumed that Justice Kennedy (who I think would go along with the liberals on the merits of NC’s voting claim) could well be persuaded that changing the election date and rules this close to the election violated Purcell. If all the conservatives agreed, that would have given 5 votes to stay the change for this election. With Scalia gone, perhaps the Court went 4-4 on this issue. We don’t know, because the Court’s order denying the stay is just as consistent with a 4-4 vote as a 0-8 vote for a stay. So maybe the Purcell principle has lost its majority. Then again I counted Justice Breyer, one of the liberals, and perhaps Justice Kagan, as agreeing with the principle in part too, based on the votes in the 2014 cases. So maybe that’s not what’s going on.
Another possibility is that the Justices were too upset/preoccupied by Justice Scalia’s death to devote the energy to duking this one out. Given that the Court does not have to give reasons, a decision here has really no effect on other cases. The fact that the order came a few hours after NC notified the Court it HAD come up with a redistricting plan may have put some Justices at ease that the lower court order was not too onerous for the state. (Though the new lines could well be an overstep by the NC legislature, violating both section 2 of the Voting Rights Act and perhaps convincing Justice Kennedy there is a partisan gerrymander—having a 50/50 state divide congressional seats 10/3 looks pretty egregious.)
Further, it might be that this kind of case presents a different kind of concern than the other Purcell principle cases. In the 2014 elections, voting was already underway or about to start, and it was not going to be delayed. Here, there is considerable confusion, and voting has begun, but there will be ample time to educate the public in time for a special election in June.
The last possibility is that the Court has reconsidered the Purcell principle. In my paper, I argue that the principle is inconsistent with how the Court usually decides emergency litigation. Aside from the question of changing the rules in the process of the election, which goes to the public interest factor, the Court should consider irreparable harm to both parties and likelihood of success on the merits. So it might be that the Court used all of these factors, and Justice Kennedy and the liberals were persuaded not to grant a stay because NC is so likely to lose when the case comes up on appeal.
The worst part of all of this is that we have to guess. One recommendation I make in my paper for emergency election litigation is that the Court should explain itself, even if the explanation has to come later. Lower courts and litigants need guidance, which is sorely lacking in shadow docket cases.
As you can tell, this issue is quite difficult to disentangle, and raises difficult questions of institutional rules and competence to adjudicate these issues with a potentially deadlocked Court. So despite Marc Elias’s facile trolling, it was reasonable to expect the Court to apply the principle again in this case, even though, after the death of one of the Justices, the Court did not.