WORKING PAPERS
LEARNING WHILE SEARCHING
This paper provides a model for analyzing the search behavior of agents that are initially uncertain about their own prospects and that look for a partner in a matching model. Positive experiences (many past offers) lead the agents to act more selectively in the market, and the incentives for selective behavior are determined in a stationary equilibrium of the entire search market. We establish conditions for the existence of equilibria with learning and we characterize features of optimal individual search behavior in such markets. We show that agents are optimally more selective inaccepting offers than in proposing matches.
DOWNLOAD HERE
REPUTATION AND COMPETITION
In this paper I consider a dynamic, continuous time model of repu-tation building introduced by Board and Meyer-Ter-Vehn [2010] and add Bertrand competition to the model. I am interested in whet-her competition enhances or abates the incentives for reputation building. This question has received very little attention so far. In the paper I first characterize the full solution to the baseline, single-player model. I, then, add competition to the model and find that with relatively frequent type changes and while competition lasts, it strictly reduces the level of reputation the firms find worth keeping. As a consequence average product quality is (temporarily) and prices are (constantly) strictly lower than in the one-player case. Eventually monopoly is restored and quality goes back up. Thus competition can reduce product quality.
DOWNLOAD HERE
OPTIMAL STOPPING WITH REGRET
This article studies an optimal stopping problem in which agents are subject to anticipated regret over future outcomes. After stopping, future outcomes are observed. The difference between the best future outcome and the actual outcome is the agent's regret. I propose a novel way to model regret in which the time during which future outcomes are observed is endogenous. Stopping occurs later than in a model without regret. However, surprisingly, the stopping decision is non-monotone in important parameters. This model can be interpreted as a game against nature with many potential applications beyond regret: one in political economy is provided.
DOWNLOAD HERE
OPTIMAL SEARCH AND THE MAXIMUM PROCESS
This paper studies the decision problem of an impatient agent who can either search and receive a a ow payoff in the form of the realized value of a Brownian motion or stop and get the running maximum of the process up to date. I show that a solution to this problem exists for discount rates that are not too low and even if the drift is positive and unrestricted. I also derive a closed form solution for the stopping time distribution. The dependency of the expected stopping time and the value of the solution on the underlying parameters of the Brownian motion is investigated. Areas of application include explorative, creative and innovative activities.