The Washington PostDemocracy Dies in Darkness

From early voting to pre-voting: a response to critics

Placeholder while article actions load

My criticism of early voting in POLITICO last week (authored with John McGinnis) has received a robust response. Election law professors like Rick Hasen have been quite negative.. Our objection that early voting turns the process into more of an exercise in raw preference aggregation is in Doug Chapin’s words “so yesterday,” “outdated” and “popular a decade ago.” (I remember these criticism from high school, but not for my early voting arguments.)

It seems we’ve stepped into a the midst of a very chummy academic consensus. Our argument may not be “relevant” to the what election lawyers like to talk about these days. Of course – if they were all talking about it, we wouldn’t have needed to make our points!

So what has changed so much in the past decade that our arguments are no longer “relevant”? Dough Chapin points to studies showing early voters are ideological than others, and thus less likely to be swayed by intervening information. We are of course aware of this research, and refer to it directly in the POLITICO piece. The problem is the empirical observation does not tell you how to evaluate it. One might conclude that this only proves that early voting encourages gut-reaction voting, and increase polarization. While the more ideological early voters are less likely to be swayed by new information, giving them more opportunity to be swayed may be even more important than giving more time to endlessly hand-wringing undecided.

Whether we should let the ideologically-committed pre-vote is a question that is not determined, but rather raised, by their interest in doing so.

Others have asked why if we care about civic participation, why we don’t favor a national election holiday. Or why we would tolerate absentee ballots. Well, it is all a matter of tradeoffs, costs and benefits, and there is no objective scale where these can be weighed.

Interestingly, the economic reason for low turnout is that votes cannot be alienated. When the value of a good is artificially set at zero, it will be not be optimally utilized or there will be suboptimal investment in its exploitation. Thus one could say if the election experts really wanted to make every vote count, they would allow for votes to be sold, indeed for someone to prospectively sell all their lifetime votes.

Take a committed Democrat. He knows he will always want to vote for every Democratic candidate for the rest of his life, no matter what. Making him go through the motions to actually do so in every election makes it harder for him to effectuate his preferences. If, however, he could make out a proxy to the Democratic party, or some group whose judgment he trusts (ACLU, etc), if would reduce the cost and difficulty of voting immensely. That would be a form of very early voting. Presumably Hasen and Chapin would say that goes too far, but that concedes the point that it is a question of balancing and trade-offs.

A voting holiday is a particularly expensive and regressive fix, in that it takes people’s freedom of contract away, including those who may have no interest in voting at all. Furthermore, I would not support federal solutions for a state-regulated issue, and did not suggest any in the POLITICO piece.

Again, much of the dispute turns on what the purpose of voting is, which is far from clear. Much of the rhetoric of modern voting is non-instrumental (ie, making every vote count) and ritualistic, so one needs an agreed metric for evaluating opposing rituals.