Transitions in Work and Learning: Implications for Assessment (1997)

Harry J. Holzer

Discussions in the popular press and in the policy-making community frequently imply that there is a "skills gap" in the U.S. economy—a gap between the skills needed by employers and the skills of workers in the labor force. Because of this alleged gap, a wide range of policies involving education and job training are frequently advocated by various groups and individuals. But what is the evidence that such a gap truly exists? And if it does, what skills are in particularly short supply in the work force? The latter question has been especially difficult to answer, at least partly because of a lack of available quantitative data on employers and their needs.

If a gap truly exists between the skills that employers need and those held by the work force, can this be inferred? What would be the likely economic effects of such a gap? Why might the choices of employers and employees with respect to skill acquisition still result in such a gap?

In recent years there has been much discussion among various social scientists of a skills mismatch in the work force, with the implication that this helps to account for the low employment rates of such groups as inner-city minorities (Wilson, 1987; Kasarda, 1995). Indeed, mismatch is another term for what economists

have often called structural unemployment, a situation in which jobs are actually available but unemployed workers cannot be hired because they lack the necessary skills, reside in the wrong area, and the like. (For a very recent description of "structural unemployment" in an introductory text, see Case and Fair, 1996.)

But unemployment is not the only possible result of such a situation. To an economist, mismatch suggests a gap between the demand and supply sides of the labor market—that is, between employers and prospective employees—terms of some particular characteristic. Thus, a skills mismatch might arise because the demand for workers with certain skills (such as higher education) has increased more rapidly than the supply of such workers.

Figure 2.1A illustrates the results of such a shift in demand from less educated workers to more educated ones. As the figure shows, the demand shift should result in both lower wages and lower employment among the less educated in the short run (and the opposite for the more educated), with the exact effects depending on the elasticity of the labor supply—that is, the responsiveness of the supply of workers to wage levels.¹ The lower this is the smaller will be the loss of employment and the greater the loss of wages for this group.²

Thus, widening wage and/or employment gaps between education groups could be interpreted as a sign that relative demands have shifted between them and that some degree of mismatch exists. Widening gaps in earnings or employment between specific demographic groups with more or less education (such as blacks and whites) could also result, and, even within a group whose members have comparable amounts of education, widening gaps for other dimensions of skills (such as those measured by test scores) could also indicate relative shifts in labor demand.

This analysis is strictly short run in nature. The new and higher wage gap between education groups would imply a higher return to investment in that particular skill, which should lead to greater enrollment in higher education (Becker, 1975). In the long run the relative supply of educated workers should gradually shift out while the relative supply of less educated workers should diminish (Figure 2.1B). This will, in turn, reduce the wage gaps between the two groups.