Contests with sequential entry and incomplete information

This paper provides a general study of a contest modeled as a multiplayer incomplete-information, all-pay auction with sequential entry. The contest consists of multiple periods. Players arrive and exert efforts sequentially to compete for a prize. They observe the efforts made by their earlier opponents, but not those of their contemporaneous or future rivals. We establish the existence and uniqueness of a symmetric perfect Bayesian equilibrium (PBE) and fully characterize the equilibrium. Based on the equilibrium result, we show that a later mover always secures a larger ex ante expected payoff. Further, we endogenize the timing of moves and show that all players choose to move in the last period in the unique equilibrium that survives iterated elimination of strictly dominated strategies (IESDS).