How Much Is One Vote Worth?

Voting is an act of charity -- something we do for our fellow citizens, not for ourselves.
I'm just here for the sticker. Photographer: Ty Wright/Bloomberg

My local registrar of voters wants to kick me off the rolls. I got a letter the other day explaining that my town no longer believes I live within its borders, and therefore, unless I take prompt action, my registration will be canceled.

By interesting coincidence, the notice arrived just days after my column appeared on some plausible reasons one might choose not to vote. With so big a decision now forced upon me (albeit by administrative error), it seems a propitious moment to discuss plausible reasons to vote after all.

In last week's column, I pointed first to the economist Gordon Tullock's famous claim that voting is a poor investment of time, given that there is virtually no chance that your vote will be decisive. Second, I wrote about my father's view that a candidate can be right on the issues and yet not deserve to win because he runs a campaign full of emotional appeals and enemies lists.

So, what are the arguments to the contrary?

Let's begin with the position defended by Tullock (and many other economists) that a single vote cannot possibly make the difference. In a national election, involving tens of millions of votes, he is obviously right. No matter how strong my interest in the outcome, I cannot reasonably believe that one vote will be decisive.

That doesn't mean, however, that my vote is worthless. Shortly before the 2008 election, the economists Aaron Edlin, Andrew Gelman and Noah Kaplan published a useful essay arguing that voting might be rational (in a cost-benefit sense) if you can reliably estimate the benefits that each side will bestow on the citizenry at large. They consider the hypothetical of a winner-take-all election in a country with 300 million people and 100 million voters. They estimate the likelihood that a single vote will be pivotal at 1 in 5 million, and use a value of $1,000 per citizen as the benefit that your party, if victorious, will bestow. On these figures, the expected value of your vote is given by:

300,000,000 x $1,000 x (1/5,000,000) = $60,000.

Not at all a bad expected value for the investment of time involved in a vote.

Alas, the model works only if a voter can confidently estimate that one side will provide $300 billion more in value than the other. The committed partisan has no doubt; the rest of us may be less sure.

Moreover, their 1-in-5-million calculation that a single vote is likely to be pivotal is somewhat optimistic. Edlin is co-author of a later paper, published in 2012, which estimates (based on 10,000 simulated elections run by Nate Silver) that the probability of a single vote turning out to be pivotal is more like 1 in 60 million. (For residents of four swing states -- New Mexico, Virginia, New Hampshire and Colorado -- the probability is closer to 1 in 10 million.)

The 2008 essay contends that we should think of voting as an act of charity, something we do for our fellow citizens, not for ourselves. I think the authors have the right conclusion but the wrong theory. Voting at its best should indeed be construed as a charitable act -- but not for the tangible, measurable benefits we may hope to bestow. Voting is, rather, a symbol of our faith in the democratic process itself.

Six decades ago, in his splendid book "An Economic Theory of Democracy," Anthony Downs put the point this way: "The return from voting per se is not the same thing as the return from voting correctly" -- that is, the value of casting a ballot is independent of the question of who wins. Over the years, lots of observers have argued that voting possesses a mainly symbolic importance.

The question is what it symbolizes.

For Downs, the vote is itself a contribution to democracy: by casting ballots, win or lose, we stave off the "catastrophe" that would occur should the electoral system grind to a halt. This might explain why we reward those who participate. At election time, those "I voted!" lapel stickers are worn with pride. Here instinct evidently accords with reality: Some studies suggest that there are reputational benefits to voting. In other words, we enjoy being able to tell our friends and colleagues that we've voted.

But why? Let me suggest a paradoxical reason. Voting, at its best, symbolizes less the desire to win than the willingness to lose. Voting is a process through which we place our most cherished convictions on the altar of democracy and say to those on the other side, "I'm willing to take my chances if you are." Voting demonstrates respect, both for our fellow citizens (including those we oppose) and for an electoral system that can, and often does, sweep from office those whose politics we admire.

Of course, if voting is an act of sacrifice, then everybody has to be willing to give something up. One of the great challenges to democracy today is the tendency of committed partisans to want to take as many issues as possible off the table -- to establish, through statute or court ruling or administrative law, limits to what the other side can do should it happen to prevail.

It's easy to understand the temptation to set one's own beliefs in stone. And in a just society, some questions should not be put to a majority vote. But the number of such questions should be small. Because every time we remove another contested issue from the reach of electoral politics, we weaken the argument in favor of voting. After all, why should you lay your cherished convictions on the altar of politics if I won't do the same?

Or, to put the question in more practical terms: Should I let the town cancel my registration? Have your polite say in the comments.

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.