* * *
A part of the Pandemic Elections series.
In a time when normal life has ground to a halt, it may be reassuring that one American tradition—suing over electoral rules—is still going strong. As of this piece’s writing in mid-May, by my count, there were more than twenty court decisions involving electoral policies allegedly rendered overly burdensome by the pandemic. These decisions were handed down by federal and state courts across the country—including the Supreme Court—and addressed, among other things, signature requirements for candidates and ballot initiatives, deadlines for voter registration, deadlines for absentee ballots, witness requirements for absentee ballots, cancellations of elections, and refusals to cancel elections.
Cases of this kind are governed by a framework known as sliding-scale scrutiny. A court first has to determine the severity of the burden imposed on the relevant right. (The right to vote is the most notable one covered by this regime, but others include the rights to associate and to run for office.) The court then calibrates its scrutiny based on the heaviness of the burden. Severe burdens thus trigger strict scrutiny, under which policies are upheld only if they’re narrowly tailored to serve compelling state interests. More moderate burdens, however, require only reasonable tailoring and important state aims to be sustained. And light burdens result in highly deferential rational basis review under which rules almost always survive.
In this brief essay, I consider how courts have deployed the framework of sliding-scale scrutiny in the time of the pandemic. In particular, three novel issues have arisen in recent cases: (1) how to conceptualize burdens that are attributable to both state action and the pandemic; (2) whether to fault plaintiffs for not having taken precautionary steps before the pandemic hit; and (3) what weight to give to the so-called Purcell principle, which frowns on late-breaking judicial changes to electoral rules. Overall, I think several courts have reached the right answers on these issues. The Supreme Court, however, is the glaring exception to this encouraging trend. This leads me to two conclusions. One is that sliding-scale scrutiny is an impressively flexible doctrine, able to resolve adequately new kinds of claims in the midst of an unprecedented calamity. The other is that the current Court remains what I have called the anti-Carolene Court, implacably hostile to efforts to vindicate democratic values.
Let’s start with identifying the burdens imposed by electoral regulations during the pandemic. In ordinary (non-pandemic) cases, courts never ask whether voting would become easier or harder if external circumstances shifted—if, say, people were more or less educated, had more or less access to cars, lived closer to or farther from polling places. Courts do often investigate whether voting burdens vary by social group, in particular, for the poor versus the non-poor and for racial minorities versus non-minorities. But courts never contemplate the possibility of temporal rather than inter-group variation in voting burdens. Does past judicial practice, then, mean the pandemic is irrelevant to the assessment of plaintiffs’ burdens? Is it just another shifting external circumstance that courts can safely ignore?
It should be obvious the answer is no. In run-of-the-mill cases, courts have no reason to speculate about the effects of different external circumstances because the conditions they actually face are relatively fixed. People aren’t going to acquire college degrees, new cars, or homes next to polling places overnight. But even in run-of-the-mill cases, the burdens imposed by electoral regulations are never functions of the rules alone. Rather, the burdens are always products of the rules’ interactions with social, cultural, and economic circumstances. It’s always these interactions—the combined impacts of the laws on the books and the broader world beyond—that determine how easy or hard it is to vote under particular electoral policies. True, the broader world beyond is rarely in the foreground of judicial opinions. But that’s just because it’s relatively invariable. It still plays a crucial role in shaping the voting burdens that people experience.
Accordingly, if the broader world beyond changes materially, then so might people’s voting burdens. And if people’s voting burdens change, then regulations that were previously constitutional might become unlawful, and rules that were previously impermissible might become valid. This toggling of legal status is the necessary consequence of a regime where (1) people’s voting burdens are constitutionally relevant and (2) people’s voting burdens stem from the interactions of electoral policies with external circumstances.
Following precisely this logic, several courts have recently held that regulations that are generally lawful are invalid under the specific conditions of the pandemic. In normal times, these rules interact with external circumstances to impose acceptable voting burdens. But in this decidedly irregular period, the rules interact with the pandemic to impose undue voting burdens. For example, the Massachusetts Supreme Court acknowledged that signature requirements for candidates had previously “withstood constitutional scrutiny.” However, “statutory requirements that in ordinary times impose only modest burdens . . . may significantly interfere with the fundamental right to run for political office in a time of pandemic.” A Virginia district court similarly commented that, “[i]n ordinary times, Virginia’s witness signature requirement may not be a significant burden on the right to vote. But these are not ordinary times. In our current era of social distancing . . . the burden is substantial for a substantial and discrete class of Virginia’s electorate.”
These decisions seem sound to me. Unfortunately, their reasoning was rejected by the Supreme Court in its one pandemic-era voting case to date. A Wisconsin district court had allowed absentee ballots to be postmarked after election day as long as clerks received the ballots by six days after the election. According to the court, “although once constitutional,” Wisconsin’s absentee ballot deadlines were unlawful in the throes of the pandemic. “Wisconsin here cannot enforce laws that, even due to circumstances out of its control, impose unconstitutional burdens on voters.”
But the Supreme Court disagreed, largely on the ground that Wisconsin’s absentee ballot deadlines were no more burdensome in the time of the pandemic than in other periods. “[I]n an ordinary election,” stated the Court, “voters who request an absentee ballot at the deadline for requesting ballots . . . will usually receive their ballots on the day before or day of the election”—that is, too late to have the ballots counted. And so too in this election. “[V]oters here would [not] be in a substantially different position from late-requesting voters in other Wisconsin elections with respect to the timing of their receipt of absentee ballots.” “[V]oters here,” just like voters there, would be unable to cast valid absentee ballots.
It fell to Justice Ginsburg to expose the flaw in the Court’s analysis in her dissent. “The Court’s suggestion that the current situation is not ‘substantially different’ from ‘an ordinary election,’” she wrote, “boggles the mind.” In general, few Wisconsinites vote using absentee ballots, and even fewer request these ballots close to the deadline. But in this election, hundreds of thousands more Wisconsinites applied for absentee ballots (to avoid having to vote in person during the pandemic), and many of them submitted their requests as the deadline neared (because that’s when the pandemic hit in earnest). Also, in general, Wisconsin clerks are able to process absentee ballot applications quickly because their volume is small. But in this election, it took significantly longer for the clerks to send out absentee ballots because of the unprecedented flood of requests for them. The Court thus made a fundamental mistake in its application of sliding-scale scrutiny—an error few other courts have recently repeated. The Court treated the burdens imposed by Wisconsin’s absentee ballot deadlines as static when they’re actually dynamic. The Court ignored how the deadlines interacted with the pandemic to make compliance with them much more onerous than usual.
Turn next to the issue of plaintiff diligence. It makes sense that the punctuality or tardiness of a litigant’s activity would be relevant to a court’s evaluation of the burden imposed on the litigant. In particular, if a litigant twiddled her thumbs during a period when she could have satisfied a given requirement—and when many other parties did meet that condition—then a court might discount any alleged burden. In that case, the litigant’s inability ultimately to satisfy the requirement might be deemed her own fault, not that of the regulation. On the other hand, if a litigant didn’t meet a given condition, but would have met it had some unforeseeable obstacle not arisen, then a court might not be as harsh in its judgment. In that scenario, the court might conclude that, notwithstanding the litigant’s seeming negligence, she was substantially hindered by the combination of the unmet condition and the unforeseeable obstacle.
The pandemic, of course, is the quintessential unforeseeable obstacle. This discussion thus suggests that a court ask the following question when confronted with a plaintiff who is currently unable to satisfy a given electoral requirement: Would the litigant probably have managed to meet the condition in the absence of the pandemic? If so, then the litigant is burdened by the interaction of the condition with the pandemic—even if the litigant could have complied with the condition had she acted with more alacrity before the pandemic struck. But if the litigant would likely have failed to satisfy the requirement in the non-pandemic counterfactual, then the pandemic can’t be blamed for her noncompliance. It’s the product, instead, of her own delay.
Several courts have approached the issue of plaintiff diligence in exactly these terms. A Michigan district court, for instance, declined to criticize a congressional candidate who had collected 700 of the necessary 1000 signatures when the State’s stay-at-home order went into effect. While other candidates had already “successfully met the current filing requirements,” the plaintiff “had campaign events planned for late March and April” that probably would have gotten him over the hump had they not been canceled due to the pandemic. Likewise, a Utah district court sympathized with a gubernatorial candidate who had obtained 21,000 of the requisite 28,000 signatures when that State’s quarantine began. While another “well-heeled candidate just cleared the threshold signature requirement,” the plaintiff, too, “easily would have obtained enough signatures to appear on the primary ballot” but for the pandemic.
Again, however, the Supreme Court has charted a different course. In the Wisconsin case, the Court unfavorably compared “late-requesting voters” with the “1.2 million Wisconsin voters [who] have requested and have been sent their absentee ballots.” But the Court failed to inquire whether these “late-requesting voters” would have been able to cast valid absentee ballots in the absence of the pandemic. Many of them likely would have been able to do so. In an ordinary election, Wisconsin clerks would have processed their absentee ballot applications more promptly, enabling many of them to return their ballots in time. The Court thus erred anew by asking the wrong question about the “late-requesting voters.” It considered whether they were as diligent as their 1.2 million earlier-requesting peers. But it should have probed whether the alleged laggards would probably have been able to vote absentee but for the pandemic.
The last issue with which courts have recently grappled is the Purcell principle, named for a 2006 case holding that judicial revisions to electoral procedures close to election day are disfavored. The Purcell principle is certainly sensible. All else being equal, it’s better if courts don’t amend electoral practices when an election is imminent. That way the risk of voter confusion is reduced, as is the potential unfairness of changing the rules in the middle of the game. Sometimes, though, all else isn’t equal. Sometimes things happen as an election nears—a legislature passing new laws, a hurricane making landfall, a pandemic turning the world upside down—that require a judicial response. In these circumstances, it would be bizarre to prioritize Purcell above every competing consideration. Purcell is just one equitable factor that courts should consider in granting relief in electoral cases. Other such factors include ensuring that elections run smoothly and, even more importantly, protecting the right to vote. In many cases, Purcell may trump these other factors when an election is nigh. But not always. Sometimes the other factors are particularly pressing, and Purcell should yield to them.
This is how the Wisconsin district court reasoned in a ruling that wasn’t appealed to the Supreme Court. That court extended the State’s deadline for registering to vote online by twelve days. The court thus enabled many citizens to register remotely rather than in person on election day (an option also permitted by Wisconsin law). Citing Purcell, the court affirmed its “sensitiv[ity] to the fact that last-minute changes to election laws may generate confusion.” “Nevertheless, it is apparent that some accommodation is necessary to preserve citizens’ right to vote amidst this unprecedented public health crisis.” The Utah district court similarly acknowledged that Purcell generally promotes “fair and orderly elections.” During the pandemic, though, this goal was better served by judicially revising the State’s electoral procedures, specifically, by relaxing the signature requirement for candidates.
Once more, the Supreme Court is the outlier in its approach to Purcell. In the Wisconsin case, the Court criticized the district court for “changing the election rules [for absentee ballots] so close to the election date.” In so doing, the district court supposedly “contravened this Court’s precedents,” which have “emphasized that lower federal courts should ordinarily not alter the election rules on the eve of an election.” Absent from this reprimand was any recognition that what courts should “ordinarily” do might not be what they should do when a pandemic is raging. Absent, that is, was a proper understanding of Purcell as one equitable factor among many, to be balanced against (and sometimes superseded by) other considerations. Instead, the Court wrongly treated Purcell as an absolute, a flat ban on judicial amendments to electoral procedures close to election day. This position is unsupportable in theory: Why on earth should one equitable factor outweigh every other consideration? And it deviates from the Court’s doctrine, too. In Purcell itself, the Justices simply noted that “[c]ourt orders affecting elections . . . can themselves result in voter confusion,” and that “[a]s an election draws closer, that risk will increase” (emphases added). This circumspect language is far from the ironclad rule into which it has been twisted by the current Court.
We’re left, then, with a decidedly mixed picture of election litigation in the time of the pandemic. The glass is half-full if we focus on the lower courts, a number of which have done admirable jobs under difficult conditions. Flexibly deploying the framework of sliding-scale scrutiny, they have held that various policies impose heavier burdens on voting during the pandemic, that burdens exist when litigants are prevented by the pandemic from complying with requirements they would otherwise have satisfied, and that Purcell is just a part of the remedial analysis. But things get considerably bleaker at the level of the Supreme Court, which erred as to all these issues in its Wisconsin decision. Extending its dismal record in election law cases under Chief Justice Roberts, the Court failed to grasp that laws’ burdens vary in tandem with external circumstances, that plaintiff diligence isn’t assessed relative to more punctual nonparties, and that Purcell is the beginning, not the end, of the discussion.
If there’s a saving grace here, it’s the ratio of lower court to Supreme Court activity: about twenty-to-one, so far, in the area of pandemic-related election litigation. The sheer volume of these suits guarantees that the vast majority of them will never be subjected to Supreme Court review. To be sure, the Court will probably decide additional pandemic-related cases—and decide them badly, misapplying sliding-scale scrutiny and furthering its ideological agenda. But as Judge Stephen Reinhardt once said of the Justices, their limited caseload means “[t]hey can’t catch ’em all.” What they don’t catch in the leadup to the November election is likely to be better, legally and democratically, than what they do.
* * *
Nicholas Stephanopoulos is a professor at Harvard Law School. He specializes in election law.