Studies of Supply and Demand in Higher Education.

• Author: Milkman, Martin

In 1990, expenditures by institutions of higher education accounted for 2.8 percent of GNP. Because of the economic significance of higher education, the NBER instituted a program on the economics of higher education. In 1991 Economic Challenges in Higher Education was published, focusing on three issues: undergraduate enrollments, academic labor markets, and operating costs of colleges and universities. Now the NBER has published a second collection of eight papers and the comments of paper discussants in Studies of the Supply and Demand in Higher Education. These papers were presented at an NBER conference held in 1991. The papers cover a wide variety of topics and many readers will only be interested in a small subsection of the papers.

After an introductory chapter, the first paper by Rothschild and White examines the market structure of higher education. The paper addresses a diversity of issues including the possibility of cross-subsidization between undergraduate education and graduate education and research. The authors conclude that this cross-subsidization can only exist if there are economies of scope in higher education. The paper also examines the allocation of students (based on quality) across universities and speculates about the efficiency of having students segregated according to ability.

The second paper, by Manski, argues that economists need to focus attention on how students infer expected returns to schooling. Manski claims that "without an understanding of expectations, it is not possible to interpret schooling behavior nor to measure the objective returns to schooling" and uses a formal model to illustrate his viewpoint. He argues that "progress is possible only if economists become more willing to entertain the use of subjective data in empirical analysis." This is something which the economics profession has been reluctant to attempt, perhaps due to a strong feeling that such data is unreliable. In his discussion of the Manski paper, Hanushek notes that it is the inclusion of heterogeneity of tastes for schooling in the Manski model which makes expectations of returns to schooling so central. If tastes do not enter the model, then the central results of the model seem to "evaporate." Despite this Hanushek agrees with Manski's contention that student expectations must be central to the college choice problem.

Hauser, using data from the CPS, examines college entry trends since 1972 by gender and race. Since, as...