A Theory of Elections and Voting Blocs

Abstract

I study how a group of agents with incomplete information about their conflicting preferences make a collective decision by means of voting. I present a model of representative democracy with citizen candidates in which the set of agents who runs for office is endogenously determined. I show that if the electorate is large enough and agents are not able to perfectly anticipate the electoral outcome, elections are always contested and an equilibrium with two candidates exists. In the last two chapters of this dissertation, I introduce a model of voting bloc formation in which groups of agents choose to coalesce to vote together in an assembly. Looking first at one coalition, then at two coalitions, and finally at an arbitrary number of coalitions emerging in a fully endogenous model of voting bloc formation, I analyze the incentives to join a voting bloc, the stability of different voting blocs and how the incentives and stable outcomes change with the size of the blocs, the internal voting rule that each bloc uses and the heterogeneity in the preferences of the agents. This model provides a new explanation of the formation of political parties in legislatures.