Essays in Labor EconomicsPublic Deposited
This dissertation contains three chapters on two broad topics in labor economics: the determinants of early career outcomes and the impact of an aging population (and related policies). The first chapter investigates how the retirement slowdown among older Americans has affected the labor market prospects of younger Americans in recent decades. Using an instrumental variables approach exploiting plausibly exogenous variation in the age composition of the old across U.S. commuting zones, I find that the retirement slowdown has had a negative impact on the composition of jobs among the young. In commuting zones where fewer older workers retire due to the initial age structure, youth employment in high-skill occupations declines while youth employment in low-skill occupations increases. The estimates imply that the retirement slowdown can account for up to 60 percent of the rise of youth employment in low-skill jobs between 1990 and 2007. This pattern of occupational downgrading is consistent with a model of the labor market featuring occupational choice, and the fact that older workers are increasingly concentrated in high-skill jobs. I also find evidence of declining youth wages and a shift towards part-time employment among the young. Together, the results suggest that retirement trends have contributed to stagnant youth labor market prospects in recent years. The second chapter, joint work with Daniel Fetter and Lee Lockwood, explores the relationship between government old-age support and transfers within the family by investigating the Old Age Assistance Program (OAA), a means-tested and state-administered pension program created by the Social Security Act of 1935. Using Census data on the entire U.S. population in 1930 and 1940, we exploit large differences in the generosity of OAA programs across state borders to estimate the effects of OAA on intergenerational living arrangements. Our results suggest that OAA reduced intergenerational co-residence among both elderly men and elderly women, enough to explain most or all of the aggregate decline between 1930 and 1940, and lay the foundation for future work using linked Census samples to investigate the impact of OAA on recipients' children and their families. The third chapter, joint work with Enrico Berkes and Bledi Taska, investigates how initial skill-specific labor market conditions affect early career outcomes of college graduates. Using data on the near-universe of online job postings in the U.S. between 2010 and 2016, we build a new measure of skill mismatch which captures how well an individual's college major matches the occupational composition of local labor demand around the time of graduation. Intuitively, a college graduate experiences skill mismatch when only a small fraction of online job postings in her city are suitable for her major in the year she graduates. Exploiting variation in skill mismatch across majors, cities and graduation cohorts, we find that a one standard deviation increase in our measure leads to a 3 percent decline in initial wages. Skill mismatch is also associated with a greater probability of being initially unemployed or employed in a part-time job, as well as a lower probability of being employed in a college occupation or one of the top occupations by college major. While the effects on unemployment, part-time employment and employment in college occupations gradually fade over time, the effects on wages and major-occupation fit persist up to 6 years after graduation. Our findings highlight the importance of having the right skills in the right place at the right time.