Working Papers

Informative Milestones in Experimentation
I study a continuous-time moral hazard problem with private learning about a project of unknown quality. There is ex ante symmetric information and full commitment for the principal. The project generates a profit if two consecutive stages are completed. The amount of experimentation required to complete the first stage (milestone) is informative but not conclusive about the quality of the project. The informativeness of the milestone yields an incentive to privately shirk in the first stage. Shirking increases the principal’s pessimism in the second stage and
thereby induces more favorable second-stage contract terms for the agent. In the optimal contract, the reward for a first-stage success decreases in its arrival time to prevent effort delays. The reward’s composition changes with the success time: early successes are rewarded with long second-stage deadlines and no bonus payments in the first stage, while later successes are rewarded with first-stage bonus payments and less continuation value from second-stage experimentation. Allowing for agent replacement between stages, I show that the principal wants to replace the agent in the second stage if the success arrives late.

Bidder Asymmetries in Procurement Auctions: Efficiency vs. Information
(joint with
Stefan Weiergraeber)
We develop a structural empirical model of procurement auctions with private and common value components and asymmetric bidders in both dimensions. While each asymmetry can explain the dominance of a firm, they have opposite welfare implications. We propose a novel empirical strategy to disentangle and to quantify the two asymmetries using detailed contract-level data on the German market for railway passenger services. Our results indicate that the incumbent is slightly more cost-efficient and has substantially more information about future ticket revenues than its competitors. If bidders’ common value asymmetry was eliminated, the average probability of selecting the efficient firm would increase by 40%-points.

Consumer Rating Dynamics 
(joint with André Stenzel and Peter Schmidt)
We consider dynamic price-setting by firms in the presence of rating systems and asymmetric information about product quality. We provide a general framework in which the price charged determines the characteristics of purchasing consumers. The price thus has two effects on future ratings: (i) a direct price effect on reviews if consumers evaluate a product relative to the purchase price, and (ii) an indirect selection effect through the determination of the set of reviewing consumers. Inference in our setup is characterized by a pair of inferred quality and purchasing consumers' tastes such that given the current price the aggregate rating is consistent and consumers' purchase decisions are individually rational. Sufficient conditions such that this inference is uniquely determined are obtained and the impact on the myopic and dynamic pricing incentives of a strategic firm are outlined. We provide empirical evidence for the relevance of our model and estimate the sign and magnitude of the effect of price on ratings using a unique dataset matching the price at which consumers purchased video games to their reviews. Finally, we show analytically for a class of additively separable utility and review functions that inference is correct in the long run provided that ratings aggregate all past reviews.

Work in Progress

Collaborating under Asymmetric Information
(joint with Sinem Hidir)
We analyze a model of dynamic collaboration in the presence of asymmetric information about a player’s ability. There is a team of two working to achieve a one time success on a project, and only the ability of one player is common knowledge (senior) while the ability of the other player (junior) is private information. This leads to gradual pessimism of the senior about the junior being of high ability as time passes without a success. The senior increases his effort over time in order to compensate for the junior’s (in expectation) lower ability. This is anticipated by the junior and therefore induces him to reduce his effort early on. We show that overall effort can increase when instead of adding a productive junior to the team with certainty, the junior’s ability is random and he is unproductive with positive probability. This uncertainty reduces the freeriding incentive of the senior.

Learning to Ride the Cycle 
(joint with David Lindequist)
We study how the business cycle impacts learning and termination within a lender-borrower relationship. Assuming that borrower ability and the business cycle are substitutes in determining a borrower’s performance, we analyze how the cycle interferes with the lender’s learning from the borrower’s performance. In our model, the lender takes into account that the borrower might be free riding on the business cycle. We find that the cycle’s persistence is crucial in understanding the lender’s willingness to terminate a credit relationship. We also find that a high current or future business cycle is a substitute for information generated from screening and thus, reduces the lender’s incentive to generate information about the borrower. Our theory gives a new understanding of banks’ reactions to loan covenant violations as well as a new perspective on bank screening cycles.

Experimentation and Project Choice
(joint with Johannes Schneider)

Old Working Papers