Essays in Education and Labor

Abstract

In this dissertation, made up of three chapters, I look at several topics related to education and labor economics. The first two chapters focus on education and the last chapter focuses on labor economics.

In the first chapter I examine how local homicides affect both attendance and enrollment in elementary schools. I use publicly available data from Chicago Public Schools and the City of Chicago. To estimate treatment effects, I use a difference-in-differences estimator that allows for both heterogeneity in treatment-timing and intensity. I find no effect at lower homicide levels, but for two homicides per year in a school's catchment area I find an approximately 0.33 percentage point increase in average daily attendance in the initial year the homicides occurred. This is a relatively large effect, about 14.7% of the standard deviation for all elementary schools. For three or more homicides I find a relatively smaller positive but insignificant effect. There also appears to be a relatively small and insignificant decrease in enrollment at the bottom of the homicide distribution, but towards the top the decrease in enrollment becomes larger in magnitude and significant. These enrollment estimates are not robust to the inclusion of pre-treatment homicides as a control, however, but the point estimates do remain negative. With the inclusion of controls, the attendance results remain similar and the estimates for the very top of the homicide distribution become significant. Overall, these findings suggest that the effect of homicides on attendance may behave in a counter-intuitive way, particularly at the levels of homicides studied in this chapter.

In the second chapter I look at how local minimum wages may affect students' high school enrollment decisions. A new minimum wage may affect enrollment by changing students' perceived payoffs to staying in education and/or working. To assess this, I estimate the effect of sub-state minimum wage policy on twelfth grade enrollment in California using yearly school enrollment data disaggregated by grade. In my preferred specification, I estimate treatment effects using a triple difference-in-differences estimator that accounts for the staggered introduction of local minimum wages. To do so, I exploit variation in local minimum wage policy across grade levels, regions, and time. I find that, except for two cohorts, enrollment typically increased in response to the introduction of a local minimum wage. For the remaining two cohorts, any changes in enrollment after adopting a minimum wage are relatively small and insignificant. Within this specification, I also allow for spillover effects, which are generally small in magnitude and insignificant.

In the third and final chapter I look at how labor market beliefs and job search behavior are affected by the UK National Minimum Wage introduction. To do so, I use a triple difference-in-differences to estimate the effect of the National Minimum Wage in low wage, high dose, areas relative to high wage, low dose areas. My preferred estimation strategy estimates treatment effects on a subset of the data that excludes people with children as they would also be subject to the Working Families' Tax Credit, a policy aimed at incentivizing people with children into work, which began around the same time. I find no evidence that the minimum wage introduction affected reservation wages or expected wages in high dose areas relative to low dose areas. I do find some evidence, however, that it affected other search behavior, such as the probability of looking for work and search effort.

Date of Award	4 Feb 2025
Original language	English
Awarding Institution	University of Bristol
Supervisor	Simon M Burgess (Supervisor) & Patrick P Arni (Supervisor)