Wednesday, October 15, 2008

More on voting

Another friend writes in:
My god, more trees have been killed by political scientists trying to figure out how voting could possibly be rational. It boils down to four points.
1) The straight math --- and you don't need any equations to get there --- says it isn't rational, unless the costs are incredibly low. And having to spend 10 minutes walking to a polling place probably destroys the incredibly low argument for any mass election.
2) Nevertheless, people vote. Meaning they are either irrational or we are missing something in our equations. We've already ruled out the possibility that the costs are so low that the longshot possibility of breaking a tie is worth it. An related argument is that the massive individual effect of breaking a tie if it happened would justify the occasional trivial costs. This argument is probably weak. Alternative hypotheses, however, are not in short supply
3) Most of the alternative hypotheses reduce to an argument that particularized benefits exist for voting, we just don't see them clearly the way we do with large campaign donations. (One might wonder if it rational to donate small amounts to a campaign; the particularized benefit clearly appears for huge donations, but what about $100 donors?) The most commonly reference benefit is the psychological benefit of civic virtue. Basically, people feel good about themselves when they vote, in the Ben Barber "participatory democracy" sense (even if the election is a huge blowout, there are important individual and group benefits to having people push the lever). Another particularized benefit is the feeling of belonging or partisan attachment, in the "I voted for Obama so I'm part of his democratic coalition that is going to change america" or "I stopped the evil fascist-socialist-neo-neoconservaitve-radical-conservative GOP" sense.
4) A second group of hypotheses assert that voting is perfectly rational and, in fact, in equilibrium: if everyone believes that everyone else will act individually rationally, they must assume that no one is going to show up at the polls. But if no one shows up at the polls, then any individual vote becomes extremely powerful, so it becomes rational to vote. But this cannot be universalized, since we'd be back at square one. Therefore, individuals use a mixed strategy of voting or not voting.