Professors Respond to Incentives

25 November 2010 at 3:03 pm Nicolai Foss 2 comments

| Nicolai Foss |

That’s the overall conclusion of a nice recent study, “Career Incentives and ‘Publish and Perish’ in German and US Universities,” by Uschi Backes-Gellner and Axel Schlinghof. Their theoretical basis is fairly standard personnel economics, but empirically they do something attractive, namely they compae intra-individual productivity differences and monetary incentives over a single researcher’s career. This means that they can avoid the biases introduced by inter-individual ability differences that plague cross-sectional comparisons of research productivity and incentives.

Briefly, Backes-Gellner and Schlinghof hypothesize that increases in research output will obtain prior to tenure in the US system as well as prior to lifetime employment in Germany (and a decline after tenure/lifetime employment). They expect productivity to rise more prior to promotion to full professor in the US than prior to equivalent career changes in Germany (because the wage structure is more compressed in German academia). Finally, for the US (but not for Germany), they expect research productivity to increase in the period before promotion to full professor, but decline afterwards. To test the hypotheses, the authors build a dataset from online CVs of US and German researchers. All hypotheses are borne out in the data.

Entry filed under: - Foss -, Institutions, Recommended Reading.

Miscellaneous Data and Measurement Links Does Research Productivity Decline with Age?

2 Comments Add your own

1. Peter Klein | 25 November 2010 at 3:42 pm

I’ve always thought the case of professorial incentives nicely illustrates the confusions in the intrinsic motivation literature. Intrinsic and extrinsic motives typically operate on different margins, making them complements, not substitutes. Few professors choose an academic career for the extrinsic rewards. On the margin, however, the choice to write one more article or one less, to give one more lecture or one less, to advise one more student or one less, etc., is often driven by extrinsic factors.
2. gabrielrossman | 26 November 2010 at 8:59 am

If people who are denied tenure leave academia their CVs are unavailable. Thus this could be, at least in part, a regression to the mean effect.