43 Pages Posted: 4 Mar 2012 Last revised: 16 Aug 2017
Date Written: August 15, 2017
This study provides a unified framework to compare three canonical types of contests: winner-take-all contests won by the best performer, winner-take-all lotteries where probability of success is proportional to performance, and proportional-prize contests in which rewards are shared in proportion to performance. Performance is affected by random noise, reflecting imperfect information. We derive equilibria and observe outcomes from each contest in a laboratory experiment. Equilibrium and observed efforts are highest in winner-take-all contests. Lotteries and proportional-prize contests have the same Nash equilibrium, but empirically, lotteries induce contestants to choose higher efforts and receive lower, more unequal payoffs. Behavioral deviations from theoretical benchmarks in different contests are caused by the same underlying attributes, such as risk-aversion and the utility of winning. Finally, we find that subjects exhibit consistent behavior across different types of contests, with subjects exerting higher effort in one contest also exerting higher effort in another contest.
Keywords: contests, rent-seeking, lotteries, incentives in experiments, risk aversion
JEL Classification: C72, D72, D74, J33
Suggested Citation: Suggested Citation
Sheremeta, Roman M. and Masters, William A. and Cason, Timothy N., Winner-Take-All and Proportional-Prize Contests: Theory and Experimental Results (August 15, 2017). Available at SSRN: https://ssrn.com/abstract=2014994 or http://dx.doi.org/10.2139/ssrn.2014994