The role of private benefits in the efficient transfer of control in a firm and equilibrium selection in a link formation game

This thesis presents three papers dealing with two important and independent issues in corporate finance and network theory. A brief description of the papers follows.; In the first paper, I examine a central question in finance regarding when the control of a firm may be efficiently transferred from one shareholder to another. A key issue arises when the controller may obtain private benefits from the administration of the firm. As opposed to the standard approach in finance literature, I consider this question in an environment in which the value of the firm under each potential controller is private information. Using a mechanism design approach, I show that an exogenous increase in the level of private benefits extracted by the controller can help the central planner reach an ex post efficient transfer of control.; In the second paper private benefits are endogenous and optimally selected by the controller as a function of the shares assigned to him through the mechanism. I show that now an endogenous increase in private benefits can enhance the set of firms where an ex post efficient transfer of control is achievable. I also show that the central planner can introduce pseudo private benefits as an object to be designed in order to facilitate an efficient transfer of control with a concentrated final ownership structure.; In the third paper,1 we study a link formation game in strategic form where multiple networks are supported by multiple Nash equilibria. In the literature, different equilibrium selection techniques have been proposed to refine this set, all of them based on cooperative refinements. In this paper we introduce a non-cooperative equilibrium selection approach. Specifically, we follow the global games approach to study the robustness of the set of Nash equilibria to incomplete information in a class of link formation games in strategic form with supermodular payoff functions. Interestingly, we are able to select a unique equilibrium, which is in conflict with those predicted by the traditional cooperative refinements. Moreover, we get a conflict between stability and efficiency even when the cooperative refinements do not.; 1This paper is a joint work with Rodrigo Harrison, Ph.D.(c) from Georgetown University.