Essays on repeated games and reputations.

University of Minnesota Ph.D. dissertation. July 2010. Major: Economics. Advisor:Aldo Rustichini. 1 computer file (PDF); xi, 135 pages, appendix pages 130-135.This dissertation is comprised of two papers, which use and aim to extend the tools of repeated games and reputations to analyze strategic interactions between two parties that can explain various economic phenomena. The first paper, "Reputation Effects in Two-Sided Incomplete-Information Games," studies reputation effects in a class of games with imperfect public monitoring and two long-lived players, both of whom have private information about their own type and uncertainty over the types of the other player. In particular, players may be either a strategic type who maximizes expected utility or a (simple) commitment type who finds it optimal to play a prespecified action every period. The strategic players may gain from opponent player's uncertainty about their types, by trying to convince the opponent that they are non-strategic. As in standard models, the (false) reputation of a strategic type of player for being the commitment type is established by mimicking the behavior of the commitment type. The distinct feature of my model is that both strategic players aim to establish a (false) reputation for being the commitment type. The class of games I consider, namely one-sided binding moral hazard at the commitment profile, encompasses a wide range of economic interactions between two parties that involve hidden-information (e.g. between a regulator and a regulatee) or hidden-action (e.g between an employer and an employee), where the reputation concerns of both parties are apparent. In both games, one party (principal) prefers that the other party (agent) play in a specific way and use costly auditing to enforce this behavior. The principal aims to establish a reputation for being diligent; whereas the agent want to build a reputation for being virtuous. Long-run equilibrium analysis requires to examine the evolution of reputations, i.e. what happens to false reputations in the longrun. Extending the techniques of Cripps, Mailath, and Samuelson (2004), I find that neither strategic player can sustain a reputation for playing a noncredible behavior, i.e. a behavior which is not optimal given that the opponent is best responding in the stage game. Hence, in this class, the true types of both players will be revealed eventually in all Nash equilibria uniformly and the asymmetric information does not affect equilibrium analysis in the long-run. The second paper, "Strategic Communication vs. Strategic Auditing with Reputation Concerns on Both Sides," studies misrepresentation of information, by a privately-informed agent, to an authority figure. Misrepresenting private information by agents, even under the existence of a regulator or monitor, is a common feature of many economic interactions. For instance, a bank who has private information about its financial health may misreport this information to a regulator; or a tax payer may fill out false income statements; or an investor may be engaged in fraudulent behavior by misrepresenting its books or have false fillings to a regulatory agency. Moreover, in most of such situations, the regulator or the monitor, who is supposed to detect deviations from the desirable behavior, may himself have an incentive to be engaged in moral hazard because of costly or timely monitoring. Moreover, both parties may have some prior or established beliefs about the other party and concerns about their reputation (for good behavior) or lack of it. The goal of this paper is to understand how private information can be manipulated by a strategic Sender (through false messages), in the presence of a strategic Receiver, who aims to deter the manipulation of information using costly auditing, when their interactions are not contractable and both parties have reputation concerns. More specifically, the Sender (`she') has noisy private information about an underlying state of nature and is able to misrepresent this information by sending false messages. There is a strategic Receiver (`he') who aims to deter this manipulation using costly auditing. The magnitude of the Senders cost of lying is governed by the auditing strategy of the Receiver, which determines the probability of an audit and detecting an undesirable behavior of the Sender. The Receiver, on the other hand, may have an inventive not to audit intensively if he thinks that the Sender is going to give the accurate information (since auditing is costly). Receiver believes that the Sender could be an honest type with some strictly positive probability. An honest type always sends the true message, whereas a strategic Sender maximizes expected payoffs. Similarly, the Sender believes that the Receiver could be a tough type with some strictly positive probability. A tough Receiver always chooses high auditing whereas a strategic Receiver maximizes expected payoffs. The fact that the private information of the Sender is imperfect and the auditing by the Receiver is random prevent players from learning each others true types (strategic). To model this environment, I use a simultaneous move version of an inspection game, with incomplete-information about the types of players, when their actions are not observable. This paper aims to analyze how uncertainty about each other's types and the concerns for (false) reputation pay of for both parties; and characterize the equilibria in the (1) one-shot game; (2) two-period game. The equilibrium strategies are determined by the parameters of the model, as well as the discount factors (in the two-period game). The infinitely-repeated game of strategic communication vs strategic auditing fits into the class of games analyzed in the first paper; and thus, none of the parties can fool the other party indefinitely, i.e. their true types will be revealed eventually