The economics of voting

I’ve been thinking a lot this week about the standard, rational-decision-maker model of voting. Or, I suppose, more accurately, choosing not to vote.

The model is very simple, and can be described with four statements:

1. The marginal value (expected impact on the outcome of an election) is almost vanishingly small. (This is the benefit of voting.)

2. Voting requires time and effort, and, therefore, there are non-trivial costs to voting.

3. It is highly likely that the costs to an individual of voting are larger than the expected benefits of casting a vote.

4. Therefore, the rational (non-)voter will choose not to vote.

I think I’ve come to the conclusion that there are a couple of mistakes with this argument.

First, suppose more and more people accept the argument, so fewer and fewer people vote. Then, the marginal value of my vote rises, and it becomes increasingly rational for me to vote. And, since I have to decide to vote before knowing whether others are voting, I can take out insurance for my vote being meaningful only by voting.

Second, and more important, however, is this. It may well be in my own personal interests to create a situation in which no one else’s vote becomes the crucial vote in the election. By not voting, I raise the marginal value of the votes of others; the more people who choose not to vote, the greater the marginal value of the votes of those who do. (Note that this is an argument not that my vote becomes more meaningful if you do not vote, but that yours does if I do not vote.) Some outcomes of the election will be undesirable to me. How do I take out insurance against the crucial vote being cast by someone whose vote will lead to an undesirable outcome? I vote, and I vote for the purpose of making other people’s votes non-crucial.

So, I hope you all voted.