Effort and selection effects of performance pay in knowledge creation

The effects of performance pay in routine, easy to measure tasks are well-documented, but they are much less understood in knowledge creation. This paper studies the effects of explicit and implicit, career concerns incentives common in knowledge work in a multitasking model, and estimates their causal effort and selection effects in knowledge creation by exploiting the introduction of performance pay in German academia as a natural experiment. Using data encompassing the universe of German academics, I find that performance incentives attract more productive academics, and research quantity increases by 14 to 18%, but without increasing output of the highest quality. The latter is explained by response heterogeneity. The quantity effort response is strongest for low productivity academics, who do not produce high quality work. High ability academics also produce more publications, but not more of the highest quality. Medium ability academics do not increase quantity but produce fewer high-quality papers.