The Growing Allocative Inefficiency of the U.S. Higher Education Sector

This paper presents new evidence on research and teaching productivity in universities. The findings are based on a panel that covers 1981-1999 and includes 102 top U.S. universities. Faculty size grows at 0.6 percent per year, compared with growth of 4.9 percent in the industrial science and engineering workforce. Measured by papers and citations per researcher, productivity grows at 1.4-6.7 percent per year and productivity and its rate of growth are higher in private than public universities. Measured by baccalaureate and graduate degrees per teacher, teaching productivity grows at 0.8-1.1 percent per year and growth is faster in public than private universities. A decomposition analysis shows that growth in research productivity within universities exceeds overall growth. This is because research shares grow more rapidly in universities whose productivity grows less rapidly. Likewise the research share of public universities increases even though productivity grows less rapidly in public universities. Together these findings suggest that allocative efficiency of U.S. higher education declined during the late 20th century. Regression analysis of individual universities finds that R&D stock, endowment, and postdoctoral students increase research productivity, that the effect of nonfederal R&D stock is less, and that research is subject to decreasing returns. Since the nonfederal R&D share grows and is much higher in public universities, this could account for some of the rising allocative inefficiency. The evidence for decreasing returns in research, which are greater than in teaching, suggests limits on the ability of more efficient institutions to expand and implies that differences in the scale of the teaching function are the primary reason for differences in university size. Besides all this the data strongly hint at growing financial pressures on U.S. public universities.