Dipartimento di Scienze economico-sociali e matematico-statistiche

Descrizione del progetto

Delegation is a widespread phenomenon which arises when there is a mismatch between formal authority and information: those with the legal right to make a decision do not possess the relevant information. They must then delegate the decision to experts with superior information. For instance, government agencies limit the prices that regulated firms can charge to consumers. Firms are better informed about their own technology and about market conditions. Similarly, Deans give recruitment committees some discretion on the salary and other characteristics of the offer to the new hire. Recruitment committees are typically better informed about the quality of candidates. In many cases, the principal (government agency or Dean) cannot rely on transfers to influence the decision of the agent (regulated firm or recruiting committee). For example, the recruitment committee does not get a bonus based on who they hire (their decision), or on the productivity of the hire (which ex-post reveals the state of the world). In such cases, the only instrument available to the principal is the choice of the delegation set, i.e. the set of actions which the agent can implement. The principal's optimal delegation set has to balance two countervailing forces. On the one hand, flexibility is desirable, as it allows the agent to react to his information. On the other hand, too much flexibility can hurt the principal as the agent can take advantage of discretion to pursue his own interests. An extensive body of the literature focuses on the properties of optimal delegation. A common assumption in the existing literature is that the distribution of the state of the world is exogenous. However, in many environments agents can affect the state by exerting costly effort. The recruitment committee can solicit applications and advertise the opening through their personal networks. In this way, they can increase the expected quality of the new hire. Similarly, a regulated firm can increase its productivity through costly investment. Furthermore, the agent's effort is unobservable, to some extent. The Dean cannot tell how intensely the committee searched and the regulatory agency cannot fully observe the firm's investment. When the state is endogenous the principal faces a new challenge: the delegation set must provide the agent with the right incentives to exert effort. The goal of this project is to develop a general theory of delegation with endogenous states. We aim at characterizing the optimal delegation set (options offered to the agent) when this form of moral hazard is present. Does the principal prevent the agent from choosing both low and high actions? Does the delegation set look like an interval or are some intermediate actions prohibited? Our project will advance the understanding of delegation, which is pervasive in the economy. We also aim to analyze optimal delegation to a team of agents, when agents misperceive each other's preferences and information.