Balancing Selection Efficiency and Social Costs in Selective Contests

Selective contests can impair participants' overall welfare in overcompetitive environments, such as school admissions. This paper models the situation as an optimal contest design problem with binary actions, treating effort costs as societal costs incurred to achieve a desired level of selectivity. We provide a characterization for the feasible set of selection efficiency and societal cost in selective contests by establishing their relationship with feasible equilibrium strategies. We find that selection efficiency and contestants' welfare are complementary, i.e. it is almost impossible to improve one without sacrificing the other. We derive the optimal equilibrium outcome given the feasible set and characterize the corresponding optimal contest design. Our analysis demonstrates that it is always optimal for a contest designer who is sufficiently concerned with societal cost to intentionally introduce randomness into the contest. Furthermore, we show that the designer can optimize any linear payoff function by adjusting a single parameter related to the intensity of randomness, without altering the specific structure of the contest.