In this dissertation I establish some of the first evidence on the early career labor market experiences of young American men from the Millennial cohort. I also conduct a cross-cohort comparison of the early career outcomes of Millennials compared to their predecessors from the Baby Boomer cohort. The empirical analysis in this dissertation is facilitated by the 1997 and 1979 samples of the National Longitudinal Survey of Youth (NLSY).

First, I document the racial gaps in early career labor market trajectories of a cohort of early Millennial men (NLSY–97, born 1980–1984), and explore the driving forces behind them. Tracing the experiences of Black and white young men over their first eight years after school completion, I show that racial gaps in various labor market outcomes opened up immediately post-schooling, and largely persisted over the subsequent years. In particular, I find that measured Black-white disparities in accumulated education and skills, especially cognitive skills, play the central role in explaining the observed racial gaps in employment and earnings.

Second, I compare how the racial labor market gaps have changed between the Baby Boomers (NLSY–79, born 1957–1964) add these Millennials. Both Black and white men in the older cohort experienced upward-sloping trajectories in employment and earnings in the first four to five years post-schooling. In the younger cohort, the labor market trajectories, especially for employment, were comparatively flatter both for Black men and for white men. Relative to the older cohort, a larger share of the racial employment and earnings gaps in the younger cohort cannot be explained by measured racial differences in observable premarket characteristics. Yet education and skills remain the key explanatory factor among observable characteristics.

Third, in co-authored work, we examine how the wage returns to cognitive skills have evolved across cohorts of white men in the U.S. labor market. We show that the distribution of measured cognitive skills has diverged between the NLSY–79 and the NLSY–97. This

divergence has a meaningful impact on estimated returns to cognitive skills. We explore why this divergence has occurred, considering both economic and measurement explanations, and we conclude that the conventional wisdom of a declining return to cognitive skills may well be incorrect.