Incentives for Parallel Innovation
Yale University - School of Management; Yale University - Cowles Foundation
August 29, 2013
Learning from the experiences of other innovators is a crucial aspect of the innovation process when several workers or teams explore new research avenues in parallel. In such a setting, under-exploration may result as workers attempt to free-ride on the new ideas generated by co-workers. This paper studies optimal incentive schemes for innovation and shows that when workers can learn from each other's experience, incentives for innovation fundamentally differ from incentives for routine activities. Optimal incentives for routine activities take the form of standard pay-for-performance where only individual success determines compensation. In contrast, the optimal incentive scheme for parallel innovation tolerates early failure and provides workers with a combination of long-term individual and group incentives for joint success. This result is in line with the empirical regularity documented in previous work that team compensation, profit sharing, employee ownership, and stock option plans are positively correlated with innovative activity. I further show that there exists a causal link between such incentives and innovation. When subjects in a controlled laboratory experiment are asked to perform a task that requires creativity and exploration, they attempt to free-ride on the exploration of others if given standard pay-for-performance contracts. Innovation success and performance is highest when subjects receive a mix of individual and group incentives that reward long-term joint success.
Number of Pages in PDF File: 52
Keywords: innovation, team incentives, exploration, group compensation, tolerance for failure, experiments
JEL Classification: M52, O31, C91, M13
Date posted: August 15, 2013 ; Last revised: August 30, 2013