« Most Important Election Law Question of the Decade: Local Rules on Voting Rights | Main | The Gig Economy and the Future of Employment and Labor Law »

Tuesday, October 11, 2016

In Defense of Early Voting

Coming from a state like Washington, where almost all voting is done by absentee ballot (and everyone has the option of voting in person, if desired, starting in late October), I may be favorably predisposed toward early voting. But I confess that I don’t entirely understand the concerns over the practice—and given the critical function that early voting serves, I think it warrants a robust defense.

The objections to early voting often center on the possibility that events happening in a narrow window (post-vote, pre-Election Day) will cause early voters to regret their decisions. There are two reasons why I have trouble understanding these objections. First, November 8 is the date we have set for the end of voting because a date needs to be set—not because events freeze in time at that point.  Imagine terrible news emerging about a winning candidate on the day after Election Day. That news very well might cause voters to change their minds about the candidate they had selected and wish they had voted otherwise. But no one would suggest that we therefore re-open voting. Why is this concern over buyer’s remorse so much greater if the news emerges prior to Election Day?  To me, it’s the same problem, with just a slightly increased chance that it will affect any given voter.

That said, the calculus is, indeed, different if the nominee withdraws (through death or otherwise) prior to Election Day. In that case, early voters could effectively be disenfranchised in a way that does not have a precise post-Election Day analogue. But this gets to the second point. Most of the concerns over early voting seem to stem from a desire to protect early voters. But no one is required to vote early. It is simply an option given to those who are (in my mind) fortunate enough to have the option. While some have questioned whether voters are able to understand the risks when they make the decision to vote before Election Day, this strikes me as a debater’s point: while surely this sometimes is true (that is, surely it is true that people sometimes vote early without considering the possibility that they would want to change their minds before Election Day), it’s hard for me to imagine that an information deficit of this sort really has much an effect on how or when people vote.

There are other criticisms of early voting—for example, the idea that it may make it harder for down-ballot candidates to challenge incumbents—that are not focused on the rights of individual voters, but rather on how the design of the election tends to affect outcomes. There are still other criticisms focused on abstract notions about, for example, how a “single Election Day creates a focal point that gives solemnity and relevance to the state of popular opinion at a particular moment in time.” But for most critics of early voting, these more abstract arguments do not appear to be central to their concerns. (And if they are, then the debate needs to shift, given that it is by no means settled that the nature of democracy should be defined in a given way or that certain electoral outcomes are better or worse than others.) Rather, most of the discussion about early voting is, as noted, about protecting the voters.

And this, for me, is precisely why early voting needs to be defended. Early voting is what protects voters. It allows voters—particularly those without the privilege of adequate flexibility in their schedules—greater access to the ballot. It also helps those voting on Election Day by shortening the length of polling place lines and reducing the burden on officials administering the elections. It assists all voters by facilitating the earlier identification and correction of errors.  (These advantages, among others, are discussed in this report by the Brennan Center.)  There’s a reason why so many jurisdictions, voters, and election experts are strongly in favor of the practice.

As for me, I mentioned I’m from Washington State, where virtually no one walks into a polling place on Election Day. Most of us receive our absentee ballot early, fill it out, then drop it in the mail when we have the chance. Most of the votes I’ve cast in my life have been through this same system. That being the case, it really does strike me as strange, and quite problematic, to require voters—who may have any number of commitments and complications that happen to fall on Election Day—to physically appear at a certain time, at a certain place, in order to access the ballot. And despite ample precedent across the country for long polling place lines, I continue to be shocked, and appalled, that Election Day voters are at times required to wait in line for hours in order to exercise their right to vote. Early voting directly responds to these problems, and, in so doing, it does a great deal to protect voters. Even as extraordinary events unfold in an extraordinary election, we should not lose sight of this basic fact.

Posted by Lisa Manheim on October 11, 2016 at 04:39 PM in Current Affairs, Law and Politics | Permalink


" I think Election Day voters are "basing their decisions on the same set of *available* information" -- the fact that they choose not to access, or to give weight to, certain available information simply reflects their personal decisionmaking process with that available informational set."

Not everyone actually has access to the same set of available information. A lot of voters don't even have internet access, many more have only dial-up access, some can't afford a Times subscription or basic cable, some are personally incapable of accessing information that's otherwise accessible to them - e.g., they're illiterate, their English is limited, they lack the cognitive skills to process information about things like the candidates' tax plans. So this picture of everyone being able to access the same information on Election Day if only they choose to access it seems rather idealized. I wouldn't rule out, though, the possibility that some valid argument can be grounded on that idealized picture. I do feel there's something symbolically valuable and legitimating, in maybe a civic republican sense, in having a collective moment of decisionmaking rather than aggregating voter preferences over a period of time.

Posted by: Asher Steinberg | Oct 12, 2016 2:28:48 PM

Lisa: I think Election Day voters are "basing their decisions on the same set of *available* information" -- the fact that they choose not to access, or to give weight to, certain available information simply reflects their personal decisionmaking process with that available informational set. That seems fundamentally different from voting at a time when certain information *does not yet exist* and thus is not *available* at all. That said, as I noted in my original comment, I agree that there's a balance of electoral legitimacy, in that decreased informational symmetry could be outweighed by increased exercise of franchise. But I'm incredibly skeptical that this could support early voting periods greater than a week or so. On the one hand, the informational asymmetry is relatively small for a week or less (because it's fairly improbable that truly and materially new events will occur in the week prior to the election) but grows exponentially larger as early voting extends beyond a week. And on the other hand, the disenfranchisement concern becomes negligible beyond a week, because it's incredibly unlikely that there are a significant number of resident voters for whom it's a significant burden to vote at any period in the week prior to the election; and the argument becomes laughable when the fight is between 35 days and 28 days, as if that extra 7 days a month ahead of time is the only magic period when certain people can cast a ballot. What is really going on is that the extra time is more valuable for groups who go round up people who otherwise aren't inclined to vote even though they very well could vote on Election Day or the week ahead of time; and suffice it to say that I don't think the so-called "disenfranchisment" of such people is something that the state must or should concern itself with.

Posted by: Hash | Oct 12, 2016 2:12:45 PM

It's not an absolute rule or in theory we would all have to vote at the same time since who knows something might occur later in day after you vote.

But, we do have an Election Day, and in part it is a matter of everyone having basically the same knowledge. Plus, how the system works, a special degree of reporting and debates occur near the end of the cycle. There is, e.g., still one debate, but early voting already occurred.

I personally like the basic idea of the people coming together at the polling place on the same date to vote. Seems a good civic practice. We have exceptions though for students etc. So, absentee voting is a thing. But, that only makes me wonder more the need for extended voting. If you can download a form, maybe get it notarized or something to validate it a bit more, and mail it. Or, pick it up at the post office or whatever. Drop it in the mail.

Why the need for 30 days or something?

Posted by: Joe | Oct 12, 2016 1:13:20 PM

Hash, I can't buy your argument because it discounts (and potentially disenfranchises) those "temporarily" away from where they're registered to vote. And that's a wider group of people than just students; it also includes (as for myself right now) those in postsurgical rehab and (as for myself in my first career) military personnel.

So long as we base "voting" and "ballots" on "legal residence," this is going to be a problem. The internet can help reduce informational inadequacy among these voters (presuming that the voters find sources they can rely upon for smaller/more-distant communities, let alone specific interests such as education policy), but cannot come even close to informational equality.

I served as a Voting Assistance Officer in military units for almost a decade. Admittedly, this was in the 1980s and 1990s... but I've seen little sign or acknowledgement that things are that much different now from then. But I think the least we can do is embrace absentee ballots, at least until we get away from this "legal residence is all" fiction.

Posted by: C.E. Petit | Oct 12, 2016 12:30:35 PM

Thanks, Hash. I was hoping someone would articulate an objection.

You suggest that the legitimacy of an election is based on the degree to which all voters are basing their decisions on the same set of available information. But voters are never basing their decisions on the same set of available information; everyone is exposed to (and receptive to) a different set of information in the run up to an election. A more precise way of articulating your principle may be to say that the legitimacy of an election is based on the degree to which all voters have the opportunity to base their decisions on the same set of available information. Stated that way, however, then I’m again not sure why early voting is such a problem. The early voters had that opportunity, just like everyone else; they elected not to pursue it. Just like the Election Day voters who limit their exposure to available information in any number of other ways.

More fundamentally, I’m not sure if your principle is the best way of gauging an election’s legitimacy. In my mind, it doesn’t focus enough on the degree to which each eligible voter has the opportunity to have a vote cast and counted. If this latter principle is what instead gauges legitimacy, then there is no mandate requiring everyone to cast that vote at the same time. If both principles are in play, then when they are in conflict (that is, when “getting all eligible voters to base their decisions on the same set of available information” runs up against “getting all eligible voters adequate access to the ballot”), I have a hard time giving the former more weight than the latter.

Two final responses: I think the earliest we could allow early voting would be the point at which the ballots could be prepared and finalized. Would it be an improvement to start voting that early? I’m not sure; I haven’t thought a lot a reform of that sort – but I certainly would be to be open to consideration of such an idea if it would aid in election administration. Finally, by describing New York as a “liberal” state, you suggest that early voting may be a partisan issue. But (my understanding is that) absentee ballot voting, which in many circumstances is a form of early voting, traditionally has favored Republicans. So in my mind, early voting is more about election administration than it is about partisan jostling. Which might help to explain why, to take one example, the director of elections in Transylvania County, NC (which on quick check appears to be a Republican stronghold) would provide the following endorsement of the practice (from page 6 of the PDF I linked to above): “I don’t know that we can do elections without early voting anymore. . . . I really can’t even begin to think of the disadvantages. From an administrative side, it’s just amazing how much more effective it is.”

Posted by: Lisa Manheim | Oct 12, 2016 12:21:15 PM

I wish every state had early voting for at least a reasonable (say 10-14 days) period of time before Election Day. I like going to the polls on election day. But sometimes it's hard to work out for various reasons (work, travel, family, etc.). Since I have tended to live in states that are unlikely to be close calls, it's tempting not to vote. Early voting can allow people to vote at a time that is convenient for them (or at least relatively more so). In an on-demand world, having to vote at in one particular place and at one particular time seems unnecessarily antiquated.

Posted by: Matthew Bruckner | Oct 12, 2016 11:10:52 AM

Lisa -- the objection to early voting isn't based on "buyer's remorse" or "protecting early voters." Rather, the objection is that an election is most legitimate when all voters are basing their decisions on the same set of available information; the election is less legitimate when different sub-sets of voters base their decisions on different sub-sets of available information, because then the election fails to tell us what the *entire* electorate would have done based on *any one available* set of information. In short, *asymmetric information* undermines the confidence that a majority vote truly reflects the actual views of a majority at any given point in time.

The concern that *asymmetric information* undermines the legitimacy of the election is why *post* Election Day news doesn't matter -- *no one* had the information, and so *everyone* is at equal risk of "buyer's remorse." That's inevitable given the need to have a cut-off date for voting. Likewise, the concern that asymmetric information undermines the legitimacy of the election explains why it doesn't matter that early voters have chosen to deprive themselves of available information -- the concern isn't with protecting early voters, but rather with protecting the integrity of the election's ability to reflect the actual views of a majority in a particular jurisdiction at a particular point in time.

Perhaps the easiest way of recognizing this is to consider the logical implications of your contrary position. Why not let people vote in the general election the day after the primaries are definitively resolved, even if that's several months before Election Day? If you don't care about asymmetry of information (or buyer's remorse, or protecting early voters from themselves), then allowing absurdly early voting provides at least some marginal benefit to voters who just want to get it done with, and no material marginal cost to the govt so long as the extra early voting doesn't entail additional administrative burdens (e.g., if it was done by mail, and the ballots just sat in an already-secured location until a week before Election Day). The reason why I assume neither you nor anyone else would support such a regime is because we all implicitly recognize the stark information asymmetry, and resulting lack of election integrity, when voters make their decisions several months apart. By contrast, a relatively short early-voting period means there's a relatively small amount of information asymmetry, and perhaps one whose negative effect on election integrity is outweighed by the corresponding benefit to election integrity from effectively broadening the franchise. (Though even quite liberal States like New York have quite reasonably decided otherwise, and stuck with Election Day alone.)

Posted by: Hash | Oct 12, 2016 1:37:26 AM

Great post, Lisa. You're spot-on about early voting protecting voters. There are simply too many voters and too antiquated voting procedures to handle everyone on a single day.

Posted by: Howard Wasserman | Oct 11, 2016 11:59:51 PM

There is from what I can tell a general understanding that early voting has to be within a certain reasonable time period. We don't have it starting on September 1st or something, even though that could expand the time when people can vote. So, it's a matter of degree. I'm unsure why voters are not protected enough if early voting was present for a week to ten days. This would especially be the case if one could vote absentee by let's say a downloadable form.

Posted by: Joe | Oct 11, 2016 7:25:11 PM

The comments to this entry are closed.