Contests with endogenous entry

This paper studies the effort-maximizing design of a complete-information contest with endogenous entry. A fixed pool of homogenous potential players with identical marginal bidding cost must incur an entry cost to enter the contest before they bid for prize(s). The designer can flexibly adjust the impact function of a generalized nested lottery contest and use a fixed budget to fund single or multiple prizes. Applying Dasgupta and Maskin (Rev Econ Stud 53(1):1–26, 1986), we establish the existence of symmetric equilibrium for all contest mechanisms concerned. A uniform upper bound for expected overall bids is identified for any eligible contest, assuming that potential bidders play symmetric equilibria. We show that the upper bound can be achieved through a Tullock contest with a single contingent prize, which adopts compatible bundles of success function and entry fees/subsidies. In particular, we identify the conditions under which the optimum can be achieved by solely setting the right discriminatory power in a Tullock contest with a single fixed prize. Finally, our analysis characterizes the optimal shortlisting rule, which reveals that the contest designer generally should exclude potential bidders to elicit higher bids.