A Citizen Candidate Model with Private Information and Unique Equilibrium

Abstract

[Introduction] We study a citizen candidate model with private information about the candidates' preferred policies (or, ideal points). By contrast, in the seminal models of Osborne and Slivinski (OS 1996) and Besley and Coate (BC 1997), and most citizen candidate models that have followed, the candidates'ideal points are assumed to be common knowledge. In the baseline model, a community is about to elect a new leader to implement a policy decision. Each citizen may enter the electoral competition as a candidate at some commonly known cost. Because each candidate's preferred policy is public information, she cannot credible promise any other than this policy in case of being elected. Anticipating this, citizens prefer the candidate whose ideal point is closest to their own ideal point, possibly themselves. OS assume a continuum of citizens (i.e., potential candidates) and sincere voting. That is, citizens vote for the most preferred candidate. BC assume a finite number of citizens and strategic voting (i.e., a Nash equilibrium in undominated strategies for the voting game). They identify a variety of different kinds of equilibria supporting different numbers of entrants, and show how the set of equilibria depends on the distribution of ideal points as well as the entry costs and benefits from holding office. For most environments, there are multiple equilibria. Both median and non-median policy outcomes can be supported in equilibrium.