This article presents evidence for the view that physical capital is more complementary to educated labor than to less educated labor. For this reason previous estimates of elasticities of substitution between educated and less educated labor are too high. Using a two-level CES production function, we run international cross-sectional regressions at the level both of the economy and of individual sectors. The whole-economy model allows for the effects of wages on educational choices as well as vice versa. It predicts that, as total capitalper head rises, the share of physical capital in national income falls and that of human capital rises. The production function also shows that intercountry differences in output per head are due more to differences in physical capital than in human capital.

The ultimate concern of this paper is educational adaptation, that process by which the educational composition of the adult population is modified in response to changes in technology and economy. Focusing on the changing role of the college educated, the first section descriptively examines the inter- and intrasectoral components of change in educational composition of the labor force, 1929-69, and explicates the destabilizing consequences of demographic movements. The second section develops a more aggregative representation of the demand side of the labor market, utilizing a CES technology relating educated and uneducated labor and incorporating differential rates of factor augmenting technical progress, and conjoins this with an explicit model of the determinants of the educational composition of the adult population. In the context of future demographic movements, the model is employed in the third section to project a 33 percent decline in college enrollments between 1970 and 2000. The final section then places the simulated educational adaptation process in its larger social-demographic context.

This paper suggests a methodology by which the monetary value of an article can be calculated. Three types of benefits are discussed and quantified: direct salary increments, promotion-related salary increments, and career-related option effects. Using data from a national cross-section of university economists, the authors compute lifetime returns for selected article categories by academic rank. Their results suggest that the returns to publication of the first article are considerable, ranging from 12,340 for an assistant professor,to 10,256 for an associate professor, to $6,958 for a full professor, based on a 5 percent discount rate. The returns to additional publication diminish rapidly at first but at a lesser rate as the number of publications continues to increase. Elderly faculty who are well published can expect to earn a relatively small sum from the publication of an incremental article. This may help to explain the decline in output as faculty age observed by some scholars.