Designing Contests Between Heterogeneous Contestants: An Experimental Studie of Tie-Breaks and Bid-Caps in All-Pay Auctions

Open Access

Abstract

Contests are well-established mechanisms for political lobbying, innovation, rentseeking, incentivizing workers, and advancing R&D. A well-known theoretical result in the contest literature is that greater heterogeneity decreases investments of contestants because of the "discouragement effect." Leveling the playing field by favoring weaker contestants through strict bid-caps and favorable tie-breaking rules can reduce discouragement and increase the designer's revenue. We test these predictions in a laboratory experiment. Our data confirm that placing bid-caps and using favorable tie-breaking rules significantly diminishes discouragement in weaker contestants. The impact on revenue is more intricate. In contrast to theory, a strict bid-cap does not increase revenue, but a mild bid-cap can increase revenue even when not predicted by theory. Our data also show that tie-breaking rules seem to have little impact on the designer's revenue: the encouragement of weaker contestants is offset by stronger contestants competing less aggressively. We discuss deviations from the Nash predictions in light of different behavioral approaches.