The interesting question on turnout is not why so few of us vote, but why so manyby Jim Holt / November 21, 2004 / Leave a comment
Image: secretlondon123 via flickr
Why does voting in a US presidential election feel at the same time both terribly important and utterly pointless? On the one hand, casting a ballot on election day strikes us as a kind of civic obligation; neglecting to do so is not so serious as neglecting to file a tax return, but it is still something you feel guilty about. On the other hand, nearly half of those Americans who are eligible to vote don’t bother. And, in a sense, they are right.
Some non-voters, no doubt, couldn’t care which candidate wins (the ancient Greeks had a word for a person who is indifferent to public affairs in this way: idiotes, or idiot). Others may be passionately interested in who wins, but suspect that their own ballot is immaterial to the outcome.
What is the chance, after all, that a single vote will swing an election? That is a tricky question, depending as it does on how close the race is. Still, ball-park estimates have been proposed. The simplest of them is just 1 divided by the total number of voters – the chance that a given voter will cast the last necessary vote for the winner. Since there are 100m voters or so in US presidential elections these days, the probability that any one of them will decide the outcome is of the order of .00000001.
If that is the infinitesimal impact you can expect to have, is it rational to take the trouble to cast a ballot? Perhaps not. Suppose you are one of the proverbial voters who “vote their pocketbooks.” Let’s say that the benefit to you personally if candidate A beats candidate B would be $1m (because of tax cuts, not having your job outsourced, and so on). If you multiply this benefit by the probability that you could affect the election (.00000001) you end up with… one lousy cent. But wait – voting also has costs, both in time (getting to the polling place, waiting in line) and money (I had to use two 37-cent stamps last week to send my social security number to the local board of elections). To be conservative, let’s put the cost of voting at $10. The expected payoff is a cent. This is one lottery ticket you don’t want to buy.
The fact that more than half of the US electorate nevertheless does go through the effort of voting is something of a puzzle to political scientists. Some of them have speculated that people must greatly overestimate the likelihood that their vote will be decisive, perhaps because of hearing about past elections that turned on a couple of hundred votes. Others have wondered whether voters are not motivated primarily by a desire for self-expression, or by the “entertainment value” of going to the polls, or even by a fascination with voting machines.
Even though an individual voter is virtually powerless to affect an election, some voters are more powerless than others, thanks to the electoral college system. Here too, though, the reality is not what it seems. Most people appear to believe that the electoral college favours voters in less populous states, since each state gets an extra two electors corresponding to its two senators, regardless of how paltry its population is. Thus the state of Wyoming (0.5m people, 3 electors) has almost four times the representation per capita in the electoral college as California (36m people, 55 electors).
But there is another feature of the electoral college system that rewards large states: the winner takes all rule. In 48 of the 50 states, the candidate who wins the popular vote gets all the state’s electors. This means that a voter in California commands potential influence over a far larger bloc of electors than one in Wyoming. In fact, as the political scientist Steven J Brams has shown, an individual voter in a large state can have as much as three times the power of one in a small state. And presidential campaigns seem to know this: they spend considerably more money per elector in large states.
Of course, living in a big state does not mitigate your powerlessness if that state is one-sided, like New York or Texas. And even in the notoriously close 2000 contest in Florida, the chance of one voter tipping the election was only of the order of 1 in 10,000.
The moral, if there is one, is to vote out of duty, not self-interest. Why duty? For the simple reason that (as the Marquis de Condorcet once suggested) the more people who vote, the greater the chance of a happy result – provided that each person is more likely to vote for the superior candidate (if you fail to meet that proviso, stay home, for heaven’s sake). But if your goal is to swing the election, run for president on a third-party ticket to draw votes away from the candidate you want to see lose. Or get appointed to the supreme court.