Fighting for Lemons: The Encouragement Effect in Dynamic Contests with Private Information
Juan Beccuti and Marc Möller
This paper identifies the encouragement effect as a generic feature of dynamic competition in a common value environment. Losing conveys positive news about a rival’s estimation of a contested prize, capable of balancing the discouraging effect of falling behind. We show that, due to the encouragement effect, aggregate incentives under private information must be higher than under public information and can even exceed the benchmark associated with static competition. By challenging the common wisdom that incentives become undermined by the dynamic nature of competition, our results have implications for our understanding of R&D races, promotion tournaments, or presidential primaries.
Selecting the Best in Dynamic Contests
Mikhail Drugov, Meg Meyer, and Marc Möller
Draft available on request
In most organizations, promotion of the most able individuals is a key factor of success. This paper studies the optimal use of bias in a two-stage promotion contest with the goal of maximizing selective efficiency. It introduces a novel class of (biased) contest models which includes both a Tullock contest with a multiplicative bias and a tournament with an additive bias as special cases. For this class, we characterize the optimal second-stage bias granted to the first-stage winner (merit) and show that generically merit is neither monotonic in the players’ heterogeneity nor converges to zero when heterogeneity disappears. Favoritism (in form of a random allocation of a first-stage bias) can be optimal but only when the size of the second-stage bias may depend on whether or not the first-stage winner was favored in the past.