Essays on the Economics of Education and Labour Market

This thesis consists of three chapters on the economics of education and labor market. Chapter 1 focuses on discontinuation decisions at the Technical University of Denmark, and Chapter 2 and 3 study peer effects and social connections among Business Economics students at Copenhagen Business School.
Chapter 1, ”An Event-History Analysis of Students Discontinuation Decisions”, inves-tigates the discontinuation decision of university students in their first university study program in Denmark. Particularly, the chapter studies the decision for students enrolled at the Technical University of Denmark. The chapter has two main contributions: The first contribution is the evidence of time dependence in the discontinuation probability that is observed with a continuous-time model using a rich data structure. The chapter confirms previous findings in the literature that the first year of study is of particular importance, and adds the empirical observation of within year dependence related to the typical outside option of an alternative study program offer. It further adds the observation that decisions are made without much knowledge from the exam process to form academic expectations.
The second contribution of the chapter is methodological. The chapter suggests a model that extends the current competing risk model in the educational literature to allow for i) stochastic decision times, ii) separation of the destination state depending on the outside option, and iii) separation of the enrollment state to allow for stochastic arrival of exam information. The exam information brings study progression when an exam is passed and performance information when an exam is graded.
The suggested model treats the discontinuation decision in sequential competing steps with the exam result arrivals and finds that information from grades only affect discontinuation decisions to a limited extent. I use this model to simulate the effect of exam information on the discontinuation decision of the students, and I find that 19%of discontinuation decisions in the first academic year can be explained by the students accumulated GPA. This estimate is in the low end compared to the literature and can be explained by a high level of early discontinuation decisions.
Chapter 2, ”University Peers and Labour Market Gender Gaps”, which is joint work with Alexander Fischer, Andrei Gorshkov, and Tróndur M. Sandoy, investigates how university peers affect the divergence in career trajectories of male and female students to top-earning jobs in the Danish labour market.
We use university records covering 21 cohorts of incoming students randomly assigned to peer groups in a large Business Economics program in Denmark and merge it with high-quality administrative register data on students’ careers. This setup allows us to estimate the effect of peer ability on students’ labour market outcomes many years post matriculation.
We find that male students assigned to higher-achieving peers are largely unaffected. On the other hand, we find that female students assigned to peers of higher ability suffer severe earnings losses, have weaker labour market attachment, are less likely to work in positions with management responsibilities, and are less likely to reach the top of the earnings distribution.
The negative effect on female careers is driven by exposure to higher-achieving male peers. To explore potential reasons for the negative effect, we investigate how peer ability af-fects educational outcomes and family decisions. Female students who are assigned to higher-achieving peers are more likely to discontinue from their study, and are less likely to have either a Bachelor or a Master degree 10 years post matriculation. We show that higher-achieving peers have limited effect on family decisions.
In Chapter 3, ”Peers and Careers: Labour Market Effects of Alumni Networks” which is also joint work with Alexander Fischer, Andrei Gorshkov, and Tróndur M. Sandoy, we study the importance of social connections formed among university peers in terms of shaping their future careers. We find that students randomly assigned to the same peer group tend to have more similar careers than students from the same cohort but a different peer group: they tend to work in the same occupations and industries and are more likely to be hired by the same employer. The strongest ”excess” similarity of group peers over cohort peers are observed at the most disaggregated level, the workplace. This effect is strong, persistent (although decreasing over time), characterized by homophily, and pronounced the most for students from the wealthiest family backgrounds.
By comparing the transitions of students to firms with incumbent group peers to firms with incumbent cohort peers, we find that students benefit from their connections by gaining access to more stable and higher-paying jobs.