Monday, November 30, 2009
Daniel S. Hamermesh (University of Texas, Department of Economics) & Gerard A. Pfann (Maastricht University, Faculty of Quantitative Economics) have posted Markets for Reputation: Evidence on Quality and Quantity in Academe on NBER. Here is the abstract:
We develop a theory of the market for individual reputation, an indicator of regard by one’s peers and others. The central questions are: 1) Does the quantity of exposures raise reputation independent of their quality? and 2) Assuming that overall quality matters for reputation, does the quality of an individual’s most important exposure have an extra effect on reputation? Using evidence for academic economists, we find that, conditional on its impact, the quantity of output has no or even a negative effect on each of a number of proxies for reputation, and very little evidence that a scholar’s most influential work provides any extra enhancement of reputation. Quality ranking matters more than absolute quality. Data on mobility and salaries show, on the contrary, substantial positive effects of quantity, independent of quality. We test various explanations for the differences between the determinants of reputation and salary.
Professor Hamermesh asks today on the New York Times Freakonomics Blog, Why Don’t Reputations and Salaries Rise Together?:
Our new study poses a conundrum: in a professional market (for economists), having more scholars pay attention to your research raises your reputation and your salary. Conditional on that attention, though, writing more papers lowers your reputation — but it raises your salary! The question is why writing more (essentially ignored) papers has opposite effects on reputation and salary? Are university administrators ignorant, rewarding something visible that in fact reduces the scholar’s quality, as measured by his/her colleagues? We tested lots of explanations for the anomaly, but none described it well. The results suggest that there might be room for a Billy Beane (see Moneyball) of academe.
Update: Stephen Bainbridge (UCLA), Deans Really Can't Read but Can Count:
It is an age-old academic aphorism that your dean can't read but he can count. In other words, at least insofar as internal matters like promotions and raises are concerned, quantity matters more than quality.
Now we have some empirical evidence that the old saying is true. Paul Caron passes along an NBER paper, which finds that:
Using evidence for academic economists, we find that, conditional on its impact, the quantity of output has no or even a negative effect on each of a number of proxies for reputation .... Data on mobility and salaries show, on the contrary, substantial positive effects of quantity, independent of quality.
In other words, prolific publication--even of crap--has a positive effect on an academic's salary. Which means deans really can count but can't read.