| This dissertation includes both theoretical and empirical research in economic efficiency. The first chapter theoretically evaluates how to design procurement models when two agents may have better information than the principal about their production costs and about the similarity (i.e., correlation) of those costs between agents. The results indicate that if the uncertainty about the correlation is not severe, the principal should design a contract as if she knew that correlation is the lowest possible. The drawback with such mechanism is that the high-cost agent earns rent if correlation is high. In contrast, if the uncertainty about the correlation is more severe, the principal should not tailor the solution to any of the potential correlations. This contract reduces the rent of the high-cost agent if correlation is high by giving rent to the low-cost agent if correlation is low. Although the welfare level may be different, the characterization of the results holds when both agents observe the correlation and do not report it, as well as in the case where only one agent observes and reports the correlation to the principal. Finally, when the probability of the low-correlation setting is very small, the principal may find optimal to exclude the high-cost agent from the contract if correlation is low, allowing her to achieve the socially optimal situation if correlation is high.;The second chapter empirically evaluates the relative efficiency of different type of water and sewerage operators in Brazil. The analysis consists of a cost comparison between public and private operators, and between state-level and municipal-level operators. In a first stage, a cost function is estimated utilizing a fixed-effects panel data model. In a second-stage, the firm-specific costs from the first stage are explained by means of firm-type indicator variables. The results illustrate that water and sewerage provision in Brazil is characterized by substantial economies of scale, indicating that state-level provision is economically more efficient. The results also show that there is not an economically significant difference between the costs of private and public firms. |
