Essays on Labor Economics: Education, Employment, and Gender

Abstract: Essay I, co-authored with Bertil Holmlund and Oskar Nordström Skans, offers evidence on how delayed transitions to higher education (“gap years”) affect subsequent earnings and lifetime earnings. We exploit rich Swedish register data and find that postponement of higher education is associated with a persistent and non-trivial earnings penalty. The main source of the earnings penalty associated with gap years is that the returns to post-university work experience are higher than the returns to gap years. Postponement of university education reduces time for post-university investment in skills and therefore entails lower earnings subsequent to university graduation. Essay II uses Swedish data to compute rates of returns to education that incorporate mechanisms operating through the marriage market. I define marriage income as the transfers one obtains from the spouse and the overall income as the sum of work income and marriage income. I find that the marriage market can work as a cushion to reduce the overall earnings gap among people in different education categories. Although women with graduate degrees earn on average more than twice as much in annual work income than women with less than two years of university, their marriage incomes are less than half of marriage incomes in the latter group. Essay III uses micro data to document and analyze patterns and trends in labor force participation and employment in urban China during the past couple of decades. Estimations of logit models show that age, education, communist-party membership and marital status are significantly associated with participation in the labor force and employment opportunities. Being a communist party member in 2002 is associated with 3.5 percentage points higher employment rate for men and 6 points higher employment rates for women. An extension of the Blinder-Oaxaca decomposition is used to analyze gender differences in participation and employment. Essay IV, co-authored with Oskar Nordström Skans, examines how parental leave policies influence children’s human capital accumulation. Using a three-month parental leave extension that covered all Swedish children born from August 1988, we evaluate children’s test scores and grades at the age of 16. We find that the duration of paid parental leave benefits have no impact on average scholastic performance, although heterogeneity analysis discovers positive effects for children to well-educated mothers. We find no effects on child health, mothers’ health and subsequent earnings or parents’ fertility and divorce rates.