Economists disagree about the factors driving the substantial increase in residual wage inequality in the U.S. over the past few decades. We identify and estimate a general model of log wage residuals that incorporates: (i) changing returns to unobserved skills, (ii) a changing distribution of unobserved skills, and (iii) changing volatility in wages due to factors unrelated to skills. Using data from the PSID, we estimate that the returns to unobserved skills have declined by as much as 50% since the mid-1980s despite a sizable increase in residual inequality. Instead, the variance of skills rose over this period due to increasing variability in life cycle skill growth. Finally, we develop an assignment model of the labor market and show that both demand and supply factors contributed to the downward trend in the returns to skills over time, with demand factors dominating for non-college-educated men.

"A large portion of the variation in wages and wage growth rates among individuals is due to "unobserved" heterogeneity, and the source of individual heterogeneity is typically attributed to data limitations and/or the unobservability of certain productivity related factors. In this paper we develop a test that discriminates between two inherently unobservable sources of heterogeneity (both of which can clearly account for the variation in wages and wage growth rates): learning ability and workers' inter-temporal preferences (discounting). We apply this test to the large observed differences in wages and wage growth rates between smokers and non-smokers. The evidence supports the discounting hypothesis"--National Bureau of Economic Research web site.

We consider a general framework to study the evolution of wage and earnings residuals that incorporates features highlighted by two influential but distinct literatures in economics: (i) unobserved skills with changing non-linear pricing functions and (ii) idiosyncratic shocks with both permanent and transitory components. We first provide nonparametric identification conditions for the distribution of unobserved skills, all unobserved skill pricing functions, and (nearly) all distributions for both permanent and MA(q) transitory shocks. We then discuss identification and estimation using a moment-based approach, restricting unobserved skill pricing functions to be polynomials. Using data on log earnings for men ages 30-59 in the PSID, we estimate the evolution of unobserved skill pricing functions and the distributions of unobserved skills, transitory, and permanent shocks from 1970 to 2008. We highlight five main findings: (i) The returns to unobserved skill rose over the 1970s and early 1980s, fell over the late 1980s and early 1990s, and then remained quite stable through the end of our sample period. Since the mid-1990s, we observe some evidence of polarization: the returns to unobserved skill declined at the bottom of the distribution while they remained relatively constant over the top half. (ii) The variance of unobserved skill changed very little across most cohorts in our sample (those born between 1925 and 1955). (iii) The variance of transitory shocks jumped up considerably in the early 1980s but shows little long-run trend otherwise over the more than thirty year period we study. (iv) The variance of permanent shocks declined very slightly over the 1970s, then rose systematically through the end of our sample by 15 to 20 log points. The increase in this variance over the 1980s and 1990s was strongest for workers with low unobserved ability. (v) In most years, the distribution of unobserved skill pricing is positively skewed, while the distributions of permanent and (especially) transitory shocks are negatively skewed.

Despite several decades of research there is still widespread disagreement over the interpretation of the wage differences between black and white workers. Do the differences reflect productivity differences, discrimination, or both? If lower black earnings reflect a productivity difference, then an economy-wide increase in the relative wages of more highly-skilled workers should lead to a parallel increase in the black-white earnings gap. We evaluate this hypothesis using longitudinal data for men and women from the Panel Study of Income Dynamics. Our findings suggest that returns to observed and unobserved skills of male workers rose by 5-10 percent between 1979 and 1985. For female workers, the return to observed skills was relatively constant while the return to unobserved skills increased by 15 percent. The evidence that black-white wage differentials rise with the return to skill is mixed. Among female workers the black-white wage gap widened in the early 1980s -- consistent with the premise that racial wage differences reflect a productivity difference. For men in our sample the black-white wage gap declined between 1979 and 1985 -- a change that is inconsistent with the rise in the return for skills

We estimate the effects of unobserved skills on labor market outcomes by investigating a change in the distribution of unobserved skills. Among people with the same levels of observed skills such as education and work experience, there are still disparities in labor market outcomes. since employers cannot observe all applicantsâ skills and productivity, they rely on the average skills of different groups. We exploit a discontinuity generated by the 2012 education policy in Iran. This policy restricting female students in specific college majors changes the size and skill distribution of high school graduates. We find three main findings. First, the education quota lowers women's college attendance. Second, young high-school graduate women are more likely to participate in the labor market and have a job. Third, the gender wage gap decreases among high-school workers due to both within and between occupation changes: treated women are paid more and they take up higher-paying middle-skilled positions that used to be non-traditional occupations for them.