A few days ago, I linked to a paper on Voter Ignorance by Ilya Somin. There are a few points in it that I would like to talk about:
1) Here is Somin talking about retrospective voting (p. 15)
The retrospective-voting argument does, however, possess a kernel of truth. As Fiorina puts it, retrospective voting can impose a kind of “rough justice” on political leaders who have failed badly. If a policy failure is large, highly visible, and easily attributable to a particular set of leaders, it is certainly likely that they will be voted out of office, as the elections of 1932, 1952, 1968, and 1980 suggest. Moreover, the bigger the failure, the less likely it is that the opposing party’s performance will be worse. The ability of voters to punish large and obvious policy failures by incumbents is one of the major advantages of democracy over dictatorship.
This is how I think about it too. Democracy in general gives politicians a weak but significant incentive to work for good policy, because while voters are quite ignorant about politics, they are not completely ignorant; they do generally know when politicians have done an especially poor job. Unfortunately democracy also gives politicians a host of other incentives that aren’t so great.
2) Somin argues that even altruistically motivated voters are rationally ignorant because the chance that they affect the election is vanishingly small. However, as Gelman and others show, altruistically motivated voters do have an incentive to show up to vote, and extending their logic, also to be informed. What Somin forgets is that while the chance that a particular voter changes the outcome of the election is minuscule when the voting population is large, when the voting population at large, the social welfare effects are also large. Moreover, the chance that a voter decides the election and the size of the welfare effect are inversely proportional to eachother, so that when we calculate expected altruistic utility of voting, it is constant.
However, I don’t think that Gelman’s analysis is quite right either because voters aren’t interested in being altruistic per se, they are interested in the psychological rewards of feeling altruistic, and that feeling doesn’t heavily depend on how informed the voter is. I should say that while it seems intuitively obvious to me that the rewarding altruistic feeling from behaving altruistically doesn’t heavily depend on how informed a voter is, I don’t have a good explanation for why that would be the case. I suspect there’s some sort of cognitive bias at work here.