wages for workers in the United States during the 1980s. In this chapter we summarize what is known about the changing wage structure and some of the more prominent explanations of those changes. The principal findings for the earnings of white men over the past 25 years are that returns to schooling and returns to labor market experience have risen and that wage inequality, both overall and within schooling and experience groups, has risen. The timing and magnitudes of these changes are not always coincident. Inequality increased more or less continually over the same period, but schooling returns actually fell in the 1970s. However, inequality and returns to schooling and experience all rose concurrently in the 1980s. The 1980s were also an interesting period for changing relative wages across gender and racial groups. That period witnessed a pause in the convergence of wages between black and white men as well as the beginnings of rapid wage gains for white women relative to white men.
Validating explanations for these changes has proven difficult, and one would not expect any single factor to account for all of the dimensions of the changing wage structure. However, many of the shifts in relative wages suggest a general rise in the value of what is thought of as generalized ability or "skill." This being the case, researchers have looked to shifts in demand that favor more skilled workers as a possible explanation. Hypotheses about shifts in demand are especially attractive since relative prices and quantities have sometimes moved in the same direction. One possibility is that shifts in product demand—for example, as a result of changing international trade flows—cause some sectors to shrink or grow and result in labor demand shifts in favor of the types of workers employed in the growing sectors. One can assess this theory by determining if expanding industries are the ones that tend to hire more skilled workers. Our impression is that, at least for the education dimension, the data support this version of the demand shift story as a partial explanation. However, most of the change in employment for college graduates is attributable to shifts in industries rather than the changing importance of college-intensive industries.
A second possibility, nonneutral technical change, perhaps linked to the changing availability and power of computers, is often put forward to account for some within-industry employment shifts. This hypothesis is plausible but by its nature difficult to evaluate. One method of accumulating indirect evidence is to determine what types of jobs or occupations are growing. Employment of college-educated workers grew rapidly in the 1970s and 1980s in technical fields such as engineering and computer science and in financial and business-related occupations. In contrast, employment of people with less schooling has shifted toward a different set of occupations, where wages tend to be lower than average. It is not obvious whether this reflects changing labor demand conditions or other factors, such as changing quality of secondary schooling or different raw abilities among those in more recent birth cohorts who do not pursue college degrees. But none of these possibilities suggests that the trends documented below are likely to reverse themselves in the near future.