Optimal large population Tullock contests

Abstract

Abstract We consider Tullock contests where contestants can be divided into a finite set of types according to their strategy cost function. Solving such contests is intractable if the number of players is finite but large and there are nonlinearities and asymmetries present. But by approximating the finite player contest with a large population model that can be solved in closed form, we can approximate equilibrium behavior in the finite player model. We then characterize the optimal bias parameters of the large population contest and interpret them as approximations of optimal bias parameters in finite player contests. We also identify conditions under which those parameters are increasing or decreasing according to the cost parameters. The parameters are biased in favor of high-cost agents if the cost functions are strictly convex and the likelihood of success is sufficiently responsive to strategy.