Monday, May 25, 2009
From the June 2009 NBER Digest: The Governance and Performance of Research Universities: Evidence from Europe and the U.S. (NBER Working Paper No. 14851), by Philippe Aghion (Harvard University), Mathias Dewatripont (Université libre de Bruxelles), Caroline M. Hoxby (Stanford University), Andreu Mas-Colell (Empresa of Universitat Pompeu Fabra) & André Sapir (Université Libre de Bruxelles):
Highly productive universities both control their own destinies and face stiff external competition, according to a recent NBER Working Paper. [The authors] construct an index of research productivity that is based on the Shanghai Ranking of World Universities, which includes measures of patents, the number of alumni who have won Nobel Prizes in science, publications appearing in citation indices, or numbers of highly cited researchers. Combining the Shanghai Ranking -- which awards 500 points to the best university -- with the results from a survey of governance policies at 196 European universities, the authors find that "the average Shanghai ranking for a European university that must get its budget approved by the government is just above 200 while the average ranking for a European university that does not need budget approval is 316. In general, each percentage of a university’s budget that comes from core government funds reduces its rank by 3.2 points."
European universities required to pay the same amount to all faculty members with the same seniority and rank have an average Shanghai ranking of 213. Universities free to pay faculty as they see fit have an average ranking of 322. Universities free to select undergraduate students as they see fit have a Shanghai ranking 156 points higher than those in which the government determines who will attend. Competition also improves research quality. Each percentage of a university’s budget that comes from competitive research grants increases its ranking by 6.5 points.
The NBER researchers find that in Sweden and the United Kingdom universities with high autonomy have high Shanghai ranking scores, while in Spain and the United Kingdom universities with low autonomy have low rankings. The results for state universities in the United States are similar. Research productivity is highest for schools in states that allow more autonomy, such as independent purchasing systems, no state approval of the university budget, and complete control of personnel hiring and pay. States with high rankings and high autonomy include Washington, Colorado, California, Wisconsin, and Michigan. States with low rankings and low autonomy include Arkansas, South Carolina, Kansas, and Louisiana.
Perhaps autonomous universities respond to competition for research funds by developing more productive, inventive, or efficient research programs. The authors seek to show how autonomy and competition affect research productivity by exploiting survey data on the wide variations in those variables among colleges in the U.S. states. Their results confirm that competition increases research productivity. In New Jersey, a highly competitive environment, an increase in exogenous research university expenditure per person increases patenting by residents of that state. In Arkansas, New Mexico, and Maine, where autonomy is low and competition is lackluster, additional spending on research universities may be wasted and may even reduce over-all patenting. Private research universities, which by definition have more autonomy, produce the most patents for any exogenous spending increase. Additional exogenous spending on 2-year colleges generally added little to research productivity during their sample period, and, in some states, may have reduced it.