Workforce Composition and Earnings Inequality

As the United States passed through another election phase last fall, we again heard about the increasingly unequal earnings prospects of America’s workforce. While there are many suggested remedies, making educational opportuniies more available to all is the most common.1 Proponents of this approach believe that rising returns to education can be attenuated by increasing the supply of highly educated workers and reducing the supply of less skilled workers. This follows from analyses indicating that education is the primary factor contributing to earnings inequality.2 Existing research in this area has typically focused on a single demographic group rather than on how demographic groups’ earnings relate.3 Juhn and Murphy (1997) extend their earlier analysis on white males to both sexes by considering the effects of marriage and family structure on family inequality. They find that like workers tend to marry one another, increasing the earnings gap between families. This approach returns the focus to general workforce inequality, but includes four major demographic groups in a generalized inequality decomposition based on estimating their human capital returns independently. The most notable results of this analysis that could not be ascertained in previous research are 1) the increasing share of women in the workforce and their increasing realized tenures have reduced earnings inequality, and 2) a larger portion of the variation in earnings is associated with the changing composition of the workforce, rather than with changing returns to human capital investments. The three main qualitative results of the existing research are confirmed in Juhn and Murphy’s study, although the levels of these factors are altered somewhat by either the different population or the procedures used in their analysis: 1) educational differences are the primary variable associated with rising inequality, 2) industry affiliation and wage differentials are associated with rising inequality, and 3) wage