This thesis is a collection of three essays, each of which analyses an economic institution in one or more developing countries. A careful analysis of institutions is crucial for the understanding of economic performance and for the design of effective policy measures. In the first essay, "On the Structure of Tenancy Contracts" I analyse the effect of crop and tenant characteristics on the form and on the length of tenancy contracts. Using a principal-agent model I show that high-powered incentives are used when, due to the characteristics of the crop, their benefit is high and/or when, due to the characteristics of the tenant, their cost is low. The theoretical predictions are consistent with the empirical evidence from a unique data set of 705 contracts. The purpose of the second essay, "Competing for Protection: Land Fragmentation and the Rise of Mafia in 19th Century Sicily", is to identify the conditions that fostered the development of the mafia. I argue that in the context of 19 th century Sicily, land fragmentation was crucial for the rise of mafia. Using a menu-auction model I show that, by inducing landlords' competition for protection, land fragmentation increases the profits of mafia groups even if the assets in need of protection are unchanged. I show that the predictions of the theory are consistent with the available empirical evidence from a sample of 70 Sicilian villages. In the third essay, "Does Financial Reform Raise or Reduce Savings?", we analyse the effect of financial liberalisation on private savings in eight developing countries. To this purpose we construct an index which summarises the reform process and estimate an error correction model for savings. We find that the effect of financial reform on savings is ambiguous.
