Some thoughts on the legal issues involved.
In reading the Sixth Circuit opinion in Ohio’s early voting case now before the Supreme Court, I get the sense that the Sixth Circuit believes that the exact same set of early voting opportunities would be constitutionally permissible in other states, just not in Ohio. Indeed, the Sixth Circuit went so far as to say “the same law may impose a significant burden in one state and only a minimal burden in another.” (Slip. op. at 25.)
This view is perplexing. What is it about Ohio that makes its relatively generous provision of early voting opportunities unconstitutional, even though they would be constitutionally permissible elsewhere? In their Supreme Court brief, the civil rights plaintiffs stress Ohio’s horrible experience on Election Day in 2004. Because of the atrociously long lines at the polls on Election Day in Ohio in 2004, the plaintiffs assert that “the default Election Day-only system was no longer a constitutional option for Ohio.” (Page 32-33.)
I entirely accept the premise of plaintiffs’ argument. But I don’t think their conclusion follows. Yes, I’ll taken as given that because of what happened in 2004 in Ohio it would be unconstitutional for Ohio to limit voting solely to Election Day. But does that mean that the amount of early voting that Ohio now has provided does not suffice?
Plaintiffs complain that only one Sunday of early voting is available, and that the only evening hours available are on Election Day itself. As a matter of policy, I concur with their complaint. If it were up to me, I’d provide two weeks of early voting, including two Sundays (as well as two Saturdays) and ample evening hours throughout the fortnight. But I’m not Election Czar of Ohio, or anywhere else, and can I really say that the amount of early voting that Ohio provides is constitutionally inadequate to redress the disenfranchising effect of overcrowding at the polls on Election Day, as occurred in 2004? I think not. As it stands, all Ohio voters have the option of voting on a Sunday, or voting in the evening on Election Day, or voting on either of two Saturdays, or voting on a weekday during regular business hours throughout the four weeks of early voting, or voting by mail during that same four-week period. It’s not the optimal schedule in my view, but it’s hard to see how any Ohio voter is disenfranchised by this particular menu of voting options.
In their Supreme Court brief, plaintiffs place great emphasis on the fact that before this year Ohio provided an even more generous early voting schedule than the current schedule. But it is difficult to see how this is relevant to the constitutional analysis. If Ohio cut back from three Sundays to two, would that be unconstitutional? Or from four Sundays to three? Rather it would seem that the better mode of analysis under the U.S. Constitution is that as long as Ohio (or any state) ends up with a constitutionally adequate amount of early voting, it does not matter whether it gets there by adjusting upward or revising downward.
To appreciate this point, let’s consider how Ohio (or another state) might end up reducing its amount of early voting in a way that would not be nefarious. Ohio currently has a Constitutional Modernization Commission, which is designed to be inherently bipartisan. It has six Democrats and six Republicans from the state’s legislature, as well as twenty citizen members chosen by these twelve legislative members. The Commission is currently considering various reforms of Ohio’s electoral procedures. Suppose that in an effort to get beyond the “voting wars” of the past decade, this bipartisan Commission adopted the same early voting schedule that Ohio currently has. Would it be unconstitutional solely because it is less generous than what Ohio had previously?
Again, Ohio’s current early voting schedule is not what I would consider optimal from a policy perspective. Were I a member of the Commission, I would push for more Sunday and evening hours. But as a member of an inherently bipartisan Commission, I would recognize that I could not insist that my policy preferences prevail. Bipartisan compromise would be necessary. If the Commission’s bipartisan compromise ended up in the same place as Ohio’s current schedule, would that bipartisan compromise violate the U.S. Constitution just because it provided less early voting than previously—and even though it gave all Ohio’s voters enough opportunities to cast a ballot?
I raise this question because one could argue that something like this is in fact what happened in Ohio. In addition to its bipartisan Constitution Modernization Commission, Ohio has a bipartisan Association of Election Officials. This Association has an equal number of Democrats and Republicans on its governing board. This Association came up with a recommended early voting schedule that essentially was adopted as Ohio’s current schedule.
As a policy matter, one can criticize the Association’s recommended schedule. It caters too much to the interests of the election officials themselves, and not enough to the interests of the voters they serve. It should have been more generous in its provisions for Sunday and evening hours. Indeed, given its bipartisan nature, I’m surprised that the Association ended up with the recommendation that it did. The compromise seems more one-sided than what I thought would have developed.
But a suboptimal bipartisan compromise is not necessarily unconstitutional. The federal judiciary does not exist to give either party a victory it could not achieve in the normal give-and-take of democratic politics. Thus, if a bipartisan compromise on the amount of early voting in Ohio does not fall short of providing the amount necessary to give all the state’s eligible voters a genuine opportunity to cast a ballot, the federal judiciary should not invalidate that bipartisan compromise as unconstitutional just because it seems to some a bad bipartisan deal.
My colleague Dan Tokaji sees Ohio’s current early voting schedule as not a product of bipartisanship, but instead one party’s imposition of its own preferences through its control of the state’s legislature (and secretary of state’s office). But I think that analysis gives insufficient attention that the role that the bipartisan Ohio Association of Election Officials (OAEO) played in the development of Ohio’s current early voting schedule. Like Dan, I deplore manipulation of voting rules by a political party in power in an effort to perpetuate that power against the wishes of the electorate. But unless and until we adopt institutional reforms that remove the power to adopt voting rules from partisan legislatures and administrators and instead place that power in nonpartisan bodies (as I’ve discussed elsewhere), then federal constitutional law will need to be sensitive about when conventional partisan legislatures and administrators have crossed the line into excessive partisan manipulation of voting rules. Over the last decade, Ohio’s legislature and secretary of state undoubtedly have crossed that line on other occasions. But given OAEO’s role in the development of Ohio’s current early voting schedule, I have a hard time seeing that Ohio crossed that line of excessive partisanship in this particular context.
The plaintiffs complain about not only the cutback in Sunday and evening hours, but also the elimination of so-called “Golden Week,” when at the beginning of early voting it had been possible to both register and vote at the same time. But it is difficult, too, to see the ending of Golden Week as unconstitutional. Eliminating Golden Week does not change Ohio’s voter registration deadline. One still needs to register 30 days before Election Day. It is just that now there is no longer a five-day window from 35 to 30 days before Election Day when one can register and vote simultaneously. Thus, the consequence is that if one registers during this five-day period, one must separately vote during one of the various times Ohio has made available for voting. No doubt separating registration from voting in this way puts a greater burden on voters than simultaneous registration-and-voting. But it has never been unconstitutional to require registration in advance of voting. This was true when voting was confined to a single Election Day, and it would seem equally true when voting has been expanded in the way that it has in Ohio’s early voting schedule.
Again, plaintiffs seem to depend on an “anti-rollback” theory in claiming that Ohio’s elimination of Golden Week is unconstitutional. In this view, simultaneous registration-and-voting is not constitutionally required, but once it is provided, it cannot be taken away, at least not in Ohio. But I do not see how this “anti-rollback” argument can be linked to Ohio’s mistakes of 2004. Whether or not Ohio provides five days at the beginning of early voting during which simultaneous registration-and-voting is possible does not determine whether Ohio has provided enough early voting opportunities to avoid the disenfranchising effect of overcrowding at the polls on Election Day. One could have a Golden Week, but still have overcrowding on Election Day (for example, if simultaneous registration-and-voting were available only during limited hours in Golden Week, and the rest of the early voting schedule were minimal, with no Sundays available). Likewise, one could eliminate simultaneous registration-and-voting during Golden Week and still avoid any risk of overcrowding on Election Day, because four Sundays of early voting are available, along with ample evening hours. Ohio’s decision to eliminate Golden Week, which was an accidental byproduct of moving to “no excuse” absentee voting, should not be considered unconstitutional as long as Ohio provides adequate opportunities to both register and vote, even if separately, for all its eligible citizens.
Although I have discussed only the constitutional issue in the case, plaintiffs also present a claim under section two of the Voting Rights Act. In their Supreme Court brief, they say that this section 2 claim does not depend on the kind of retrogression inquiry that existed under Section 5 before Shelby County. If this disavowal is correct, it would seem that their section 2 claim has far-reaching implications that they are not prepared to acknowledge.
Let’s assume for the moment that the kind of comparison that the Sixth Circuit undertook of how current law affects different racial groups is the correct analysis under section 2. Let’s stipulate further that the lack of Golden Week disproportionately makes it more difficult for blacks than for whites to cast ballots. If that is the case, it would seem that the lack of a comparable Golden Week in many other states would be disproportionately burdensome on the African-American voters in those states. It does not take sophisticated social science to know that African-Americans in Pennsylvania, Michigan, and elsewhere suffer disproportionately from poverty and other conditions that would make the ability to register and vote at the same time especially attractive. If not having the opportunity to register and vote at the same time violates section 2 of the Voting Rights Act in Ohio, it would seem also to violate section 2 in Pennsylvania and Michigan—unless one employs an anti-retrogression principle (which plaintiffs disavow) to distinguish Ohio from these neighboring states.
Pennsylvania and Michigan require registration in advance of voting. That’s a rule that can be analyzed under section 2. These states need not have to have that rule. (Other states that permit same day registration don’t.) If this rule imposes a disproportionate burden on minority voters, and that’s enough to trigger liability under section 2, then Pennsylvania and Michigan (among many other states) should be vulnerable to section 2 liability for requiring registration in advance of voting. But if the disproportionate burden of an advance registration requirement is not enough to trigger liability under section 2, then what is?
The plaintiffs talk of a “racialized” political environment in Ohio. Certainly, some of the comments individuals have made about early voting in Ohio have been deplorable and unconscionable. But there has never been a finding, or even a claim, that Ohio eliminated Golden Week in order to discriminate against African-Americans. And without additional evidence, it seems unlikely that Ohio’s politics are more racially divisive than Michigan’s or Pennsylvania’s.
Plaintiffs say that the “totality of the circumstances” inquiry under section 2 produces the “unique” conclusion that Ohio violates the Voting Rights Act for not having simultaneous registration-and-voting. But a “totality of the circumstances” inquiry still needs to depend on principled reasoning, so that like cases are treated alike under the rule of law. If plaintiffs can offer no principled basis for distinguishing Ohio from neighboring Pennsylvania and Michigan under the Voting Rights Act—or disavow an anti-retrogression reason for distinguishing Ohio from these neighbors—then it would be inappropriate to single out Ohio for liability.
Finally, a brief word about the “equities” in the context of the emergency stay application in the Supreme Court. Whatever happens now, Ohio’s voting process for this year has been disrupted. No doubt Ohio could comply with the district court’s preliminary injunction requiring more early voting than what Ohio’s current law would provide. But suppose that, for the reasons analyzed here, after the 2014 election is over, Ohio would prevail in the defense of its current law against both the constitutional and Voting Rights Act claims brought by plaintiffs. Then, the first federal election in which Ohio’s new law would take effect would be the presidential election of 2016. That seems highly undesirable from an administrative perspective. Thus, if Ohio is likely to prevail on the merits of the case, then the equities would seem in favor of giving Ohio a chance to implement its new law this year, rather than for the first time (in a federal election) in 2016.