Re-examining Voter Turnout in Large Elections∗

A well-known shortcoming of rational voter models is that the equilibrium probability that an individual votes converges to zero as the population of citizens tends to infinity. We show that this does not – as is often suggested – imply that equilibrium voter turnout is insignificant in the limit. We characterize limiting equilibrium turnout and show that it may actually be arbitrarily large. Indeed, expected equilibrium turnout is shown to be closely approximated by 1/ ( 2πc ) , where c is the lowest possible realization of an individual’s voting cost. ∗We thank John Aldrich and Mike Munger for valuable discussions. Taylor’s research was supported by NSF grant SES-0417737. †Department of Economics, Duke University, Box 90097, Durham, NC 27708-0097, email: [email protected] ‡Department of Economics, Duke University, Box 90097, Durham, NC 27708-0097, email: [email protected]