Three essays on human capital

Date
2019
Journal Title
Journal ISSN
Volume Title
Publisher
University of Delaware
Abstract
I study three things that can affect human capital: adequate rest, female leadership and macroeconomic conditions. Chapter 1 uses a difference-in-difference methodology to measure the causal effects of changes in these hours of service regulations on safety. I compare intrastate and interstate trucking companies, and I compare a period with stricter regulations to pre- and post-periods with weaker regulations. I find that stronger requirements for minimum rest periods increased crashes per 1,000 trucks by 0.165 and increased crashes per 1,000,000 miles by 0.377. The effects on crashes per 1000 trucks persisted after the rules were subsequently relaxed, while effects on crashes per million miles did not persist. These results imply the intended regulatory goal of increased safety was not realized. Chapter 2 investigates the role-model effects of national female political leaders on male and female teenagers’ educational performance. I use global micro data from three major large-scale international assessments, from 2006 to 2016. I find no evidence for a country-level female leadership role model effect on teenagers’ math and science performance. Female leaders also do not close the gender gap in educational performance, but they do improve fourth-graders’ reading performance by 0.17 standard deviations. Alternative causal mechanisms like female empowerment are analyzed but seem to have little empirical support. Chapter 3 studies the relationship between parental household inputs and kids’ educational performance, with a focus on the role of unemployment. I find that the national unemployment rate is statistically significantly correlated with students’ test scores in math and science, but it is negatively correlated with students’ reading test scores. At the micro level, fathers’ job losses are negatively correlated with their children’s test scores in math, science, and reading, while mothers’ job losses do not show any significant relationship with their children’s test scores. I also find that kids’ educational performance is positively correlated with parental involvement, and fathers’ job loss is negatively correlated with parental involvement in reading while unemployment is positively related with parental involvement.
Description
Keywords
Human capital
Citation