Performance Responses to Competition Across Skill-Levels in Rank Order Tournaments: Field Evidence and Implications for Tournament Design
Executive Summary — Tournaments and other rank-order incentive mechanisms have been used to model a wide range of settings: executive placement, elections, research and development and innovation contests, sports tournaments, and variable sales compensation: situations in which placing at the top of the performance rank-order leads to out-sized payoffs. This article analyzes how the level of competition and size of a tournament affects performance as a result of how strategic interactions affect contestants' incentives to exert high levels of effort. The authors estimate relationships between performance in these contests and competition levels across the full distribution of skill levels. They do this by studying data on software algorithm programming contests in which fine-grained data are available on contestant ability levels and performance over a large number of comparable contests. Findings show that while aggregate and average patterns of performance and effort may decline with increased competition, performance and effort may in fact increase among the highest-skilled contestants. The paper provides guidance to designers of innovation and crowdsourcing tournaments. Key concepts include:
- Tournaments and contests have a long history as a means of achieving technological advances in a range of industrial settings.
- For the strongest contestants, adding more contestants can produce effort-inducing rivalry.
- Increased competition beyond a minimum level may reduce the probability of winning to a level where incentives become depressed. However, the stimulating effect of rivalry may persist at least for highest-skilled contestants.
Tournaments are widely used in the economy to organize production and innovation. We study individual contestant-level data from 2,796 contestants in 774 software algorithm design contests with random assignment. Precisely conforming to theory predictions, the performance response to added contestants varies non-monotonically across contestants of different abilities, most respond negatively to competition, and highest-skilled contestants respond positively. In counterfactual simulations, we interpret a number of tournament design policies (number of competitors, prize allocation and structure, divisionalization, open entry) as a means of reconciling non-monotonic incentive responses to competition, effectively manipulating the number and skills distribution of contestants facing one another.