Suchergebnisse

My dissertation utilizes tools from game theory to derive novel economic insights in a variety of settings, including social learning with biased information processing beliefs, repeated games with persistent actions and contract design in crowdsourcing labor markets. The first chapter explores how individuals learn from their predecessors when they are subject to biased beliefs about the information processing capabilities of others. I consider a social learning environment in which individuals observe private signals, and learning is asymptotically efficient in the absence of information processing biases. Either underestimating or overestimating others' information processing capabilities can have important implications for the asymptotic efficiency of learning. The second chapter studies a new class of stochastic games in which the actions of a long- run player have a persistent effect on payoffs. The setting is a continuous time game of imperfect monitoring between a long-run and a short-run player. The main result of this paper is to establish general conditions for the existence of Markovian equilibria and conditions for the uniqueness of a Markovian equilibrium in the class of all Perfect Public Equilibria. The existence proof is constructive and characterizes, for any discount rate, the explicit form of equilibrium payoffs, continuation values, and actions in Markovian equilibria. Action persistence creates a channel to provide intertemporal incentives, and offers a new and different framework for thinking about the reputations of firms, governments, and other long-run agents. The final chapter examines information and incentive issues in a novel labor market setting : crowdsourcing. Our research focuses on how to optimally design labor contracts and the marketplace in a crowdsourcing setting to facilitate efficient information transmission between workers and firms. The structure of such contracts will depend on the institutional features of the specific job, including how costly it is for a worker to acquire information or how likely it is for the firm to observe when a worker incorrectly completes a task

We model the dynamics of discrimination and show how its evolution can identify the underlying source. We test these theoretical predictions in a field experiment on a large online platform where users post content that is evaluated by other users on the platform. We assign posts to accounts that exogenously vary by gender and evaluation histories. With no prior evaluations, women face significant discrimination. However, following a sequence of positive evaluations, the direction of discrimination reverses: women's posts are favored over men's. Interpreting these results through the lens of our model, this dynamic reversal implies discrimination driven by biased beliefs. (JEL C93, D83, J16, J71)