[séminaire CREM] Centralized vs. Decentralized Incentives in Teams: Experimental Evidence (co-écrit avec David Masclet)

Abstract:
This paper presents evidence from a laboratory experiment designed to compare the relative effectiveness and efficiency of centralized and decentralized incentive schemes for firm performance, when the workforce is organized in teams: target-based, team tournament and peer-pressure mechanisms. We provide a theoretical model and test its predictions via a controlled laboratory experiment. Our experiments confirm the existence of free-riding in the baseline treatment without incentive scheme, albeit not as severe as theoretically-predicted. Although peer pressure is effective in increasing individual effort, there remains substantial free-riding. Target-based schemes lead to near-Pareto levels of effort, but to lowest worker payoffs. Team tournaments are second-best for effort and close to Pareto for average worker payoffs, but involve considerable payoff inequalities between individuals. The two centralized incentive schemes dominate the baseline and peer-pressure treatments for firm profits and social welfare, which are maximized under target-based schemes and minimized in the baseline treatment.

Co-écrit avec David Masclet