Altruism and the Decision to Vote

A core problem at the foundation of rational-actor models for politics is the seeming irrationality of voting, i.e. that it is irrational for voters to vote and to be informed about politics. Myerson has shown that only low turnouts are feasible if we just assume citizens vote on the basis of self interest, and there is uncertainty as to the number of voters. I modify the standard analysis by arguing that individuals are motivated by weak altruistic considerations when deciding whether to vote. First, I present a formal analysis of voting that has a unique, symmetric Bayes–Nash equilibrium for the voting game. It is a mixedstrategy equilibrium which also specifies the conditions under which high turnouts (in excess of 50% of the potential voters) will result. Second, I present an empirical test of the hypothesis using a unique data set (the National Election Survey Pilot Study in 1995) which includes various measures of ‘humanitarianism.’ I am able to integrate that survey with the 1994 NES election survey to provide evidence that weak altruism is the single most important determinant of the decision to vote.