Land and Power: Theory and Evidence

In this paper we investigate the e�ect of the absence of a secret ballot on electoral outcomes and resource allocation. Once voting behavior is observable, votes can be bought and sold in a `market for votes'. We distinguish between direct vote buying, where individuals sell their own votes to political parties, and indirect vote buying, where people also sell the votes of others and we characterized the circumstances in which vote buying changes the electoral outcome. We then provide a microfoundation for indirect vote buying, which usually takes the form of employers selling the votes of their employees. This can oc- cur when the employment relationship involves rents since employers can use the threat of withdrawal of these rents to control the political behavior of their work- ers. This increases the demand for labor and generates an added incentive to own land, increasing the price of land. We test the predictions of the model by examining in detail the eff�ects of the introduction of the secret ballot in Chile in 1958. We show that this change in political institutions had implications for voting behavior and land prices which are consistent with the predictions of our model.