A General Framework for Studying Contests

We develop a general framework for studying contests, including the well-known models of Tullock (1980) and Lazear & Rosen (1981) as special cases. The contest outcome depends on players' efforts and skills, the latter being subject to symmetric uncertainty. The model is tractable, because a symmetric equilibrium exists under general assumptions regarding production technologies and skill distributions. We derive new comparative statics results regarding how the size and composition of contests affect equilibrium effort, showing how standard results can be overturned. We also discuss the robustness of our results to changes in the information structure and explore implications for optimal team design