Papers with abstracts
Publications
Returns to Education Quality for Low-Skilled Students: Evidence from a Discontinuity, with Pierre Mouganie. Journal of Labor Economics, 36(2): 395-436, 2018.
Abstract: This paper studies the labor market returns to higher-education quality for low-skilled students. Using a regression discontinuity design, we compare students who marginally pass and marginally fail the French high school exit exam on the first attempt. Threshold crossing leads to an improvement in quality but has no effect on quantity of higher education pursued. Specifically, students who marginally pass are more likely to enroll in STEM majors and postsecondary institutions with better peers. Marginally passing also increases earnings by 12.5% at the ages of 27–29.
The Long-Run Effects of Reducing Early School Tracking. Journal of Public Economics, 187: 104206, 2020.
Abstract: This paper studies the effects of a French detracking reform on individuals’ long-run education and labor market outcomes. The reform delayed the age at which students were separated into vocational and general education by two years, from age 11 to 13. The assignment of students to different types of education at age 11 was instead replaced by ability grouping, whereby students followed a common general education curriculum but were divided into different classrooms based on their academic achievement. Using a regression discontinuity design, I find that the reform raised individuals’ level of education and increased their wages by 6 percent at ages 40 to 45. These effects are concentrated among individuals from low socioeconomic backgrounds, who were the most likely to be placed in vocational education prior to the reform.
The Impact of Advisor Gender on Female Students' STEM Enrollment and Persistence, with Pierre Mouganie. Accepted, Journal of Human Resources
Available as: IZA DP No. 12415
Abstract: To reduce the gender gap in science fields, policymakers often propose providing women with mentoring by female scientists. However, there is no clear evidence on whether one-on-one mentor gender influences women’s STEM participation. We exploit a unique setting where students are randomly assigned to academic advisors—who are also faculty members—in their freshman year of college. Advisors help students select courses and decide on a major. We find that having a female rather than a male science advisor substantially increases the likelihood that women enroll and graduate with STEM degrees. A non-science advisor’s gender has no impact on students’ major choice.
Does Advisor Gender Affect Women's Persistence in Economics?, with Pierre Mouganie. AEA Papers and Proceedings, 2021.
Abstract: Women persist in the field of economics at significantly lower rates than men. We show that the gender of an academic advisor is an important determinant of female undergraduates' persistence in economics. We use data from a 4-year college where first-year economics majors are randomly assigned to advisors who are also faculty members in the economics department. Advisors help students choose courses and monitor their academic progress. We find that having a female rather than a male advisor reduces female students' first-year dropout rates and increases their likelihood of graduating with an economics degree.
Working papers
Advisor Value-Added and Student Outcomes: Evidence from Randomly Assigned College Advisors, with Antoine Deeb and Pierre Mouganie. Revise and resubmit, American Economic Journal: Economic Policy
Available as: IZA DP No. 12739
Abstract: This paper provides the first causal evidence on the impact of college advisor quality on student outcomes. To do so, we exploit a unique setting where students are randomly assigned to faculty advisors during their first year of college. We find that higher advisor value-added (VA) substantially improves freshman year GPA, time to complete freshman year and four-year graduation rates. Additionally, higher advisor VA increases high-ability students’ likelihood of enrolling and graduating with a STEM degree. Our results indicate that allocating resources towards improving the quality of academic advising may play a key role in promoting college success.
Parental Leave, Household Specialization and Children's Well-Being. Revise and resubmit, Labour Economics
Available as: IZA DP No. 12420. Media Coverage: IZA newsroom
Abstract: Many countries offer new parents long periods of paid leave. Proponents argue that parental leave programs can reduce gender gaps in the labor market, support marital stability and promote children’s well-being. In this paper, I show that lengthy leaves can instead work against several of these intended goals. Using a regression discontinuity design, I find that a 3-year expansion of paid leave in France increases household specialization by inducing mothers to exit the labor force and fathers to raise their work hours. The leave further decreases marriages among cohabiting couples and harms children’s verbal development.
The Impact of Religious Diversity on Students' Academic and Behavioral Outcomes, with Antoine Deeb and Pierre Mouganie
In many countries, religion is a source of deep societal divisions, violent conflicts and political polarization. In such contexts, religiously diverse educational institutions may help break down cultural barriers through facilitating interactions between religious groups. However, evidence on how intergroup contact affects individuals’ behavior is mixed, and studies focusing on contact between different religious groups are particularly sparse. This paper provides the first causal evidence on how exposure to college peers from different religious high schools affects students’ behavior and learning outcomes. We use data from the American University of Beirut (AUB), a secular 4-year college in Lebanon, where students are randomly assigned to different peer groups. Prior to enrolling in AUB, students attend either Secular, Christian or Islamic high schools. These schools differ in whether they provide religious education and the religious diversity of their student body—with secular schools being the most diverse and Islamic schools the most homogeneous. We show that for students from Islamic schools, exposure to dissimilar peers improves their academic performance but makes them more likely to take classes with Muslim teachers. In contrast, diversity decreases academic performance for students coming from secular schools, and has no impact on their instructor choice. Our results indicate that for individuals coming from religiously homogeneous backgrounds, intergroup contact strengthens within-group solidarity.
Is Parental Leave Costly for Firms and Coworkers?, with Anne A. Brenøe, Nikolaj Harmon and Heather Royer
Available as: NBER Working Paper No. 26622 and IZA DP No. 12870.
Abstract: How are firms and coworkers affected when employees take parental leave? We use Danish administrative data linking the universe of firms, workers and births to evaluate the impact on firms and co-workers of an additional woman giving birth in an environment with a generous leave policy. We utilize a dynamic difference-in-difference design that leverages the timing of a woman’s leave and matches women who give birth in a particular year with women who do not give birth in that year. We argue that while the timing of a woman’s childbearing may be endogenous to her own outcomes, the timing is likely less endogenous from the firm’s perspective because a firm is composed of many employees. We rule out this concern that the timing of birth is endogenous from the firm’s point of view via the examination of firm-level trends prior to the birth. Firms respond to leave-taking by increasing employment via hiring additional temporary employees and slightly increasing work hours of existing employees. Total hours are unchanged with only slight adjustments to workforce composition. The firm’s total wage bill increases but once we account for the reimbursements that firms receive for leave payments, the total wage bill is unaffected. We observe no effect on the unemployment risk of existing employees, firm output, gross profits or firm closures. Overall, our estimates suggest that the costs of parental leave on firms and coworkers are small at best.