Research

Work in Progress

Recipient of the Susan Schmidt Bies Prize for Doctoral Student Research on Economics and Public Policy, 2019

Does diversity influence scientific research? This paper examines the spillover effects of undergraduate gender diversity on the production of research related to gender. We study the impact of switching from male-only to coeducation on research at 78 U.S. universities between 1965 and 1990. We find a 37% increase in total research publications related to gender. This effect persists for nearly a decade after the school has turned coed. We show that at least 28% of the total increase comes from within-researcher changes in their research topics of interest, with the remainder coming from compositional changes. One important mediator for the treatment effect is the interaction between students and researchers. In psychology, the increase in gender-related research is driven by experimental research, providing one example of how undergraduate diversity can affect faculty's research questions.


  • Effect of Female Peers on MBAs’ Career Outcomes (with Menaka Hampole and Francesca Truffa)

Recipient of the Susan Schmidt Bies Prize for Doctoral Student Research on Economics and Public Policy, 2019

In this paper, we investigate how female peers influence the likelihood of attaining senior corporate leadership positions for female MBA graduates. Using a combination of administrative data from a top 10 U.S. business school and CV data from public LinkedIn profiles, we exploit the exogenous assignment of students into business-school sections to identify the causal impact of female peers. Despite no difference in the likelihood of becoming a manager, women are 27% less likely to hold senior leadership positions in the 10 years after graduation. An increase of 1 SD in the proportion of female MBA peers raises the probability of holding a senior management position by 8% for women with no effect for men. This corresponds to a 32% reduction in the gender gap. We show that this increase is not driven by new entries into the managerial pipeline, increases in labor supply, or a change in industry mix. Instead, this effect is driven by an increase of female senior managers in male-dominated industries and a shift of women into more female-friendly firms within industries. An analysis of the rates at which job switchers move to the same firm of a same-section classmate reveals that peer ties formed between female MBA classmates are much stronger than those between male classmates, suggesting that one important mechanism for these results is gender differences in referrals.


  • Pension Caregiver Credits and the Gender Gap in Old-Age Income (with Francesca Truffa)

We study a 2001 pension insurance reform in Germany that introduced additional caregiver credits for working mothers with children between the ages of 3 and 10. Using administrative social security data from Germany combined with a difference-in-differences design, we find that the reform leads to a 65% increase in yearly retirement contributions during the eligibility period. 76% of the total effect can be explained by a change in the labor market outcomes of eligible mothers, while the remaining 24% is the mechanical effect of the reform. We find a significant increase in employment earnings, driven by both an increase in employment and a switch from marginal to employment subject to social security contributions. Finally, a simple life-cycle model predicts that the pension reform leads to a 9.8% increase in retirement income and a 12% reduction in the gender gap in old-age income.


  • The Spillover Effects of Maternity Leave Extensions on Unemployment Insurance (with Francesca Truffa)

This paper examines the fiscal externality of maternity leave extensions on unemployment insurance using German administrative data. We exploit a reform in Germany to show that extensions of maternity leave reduce mothers' UI takeup by 20% and total unemployment benefits by 22% in the first five years after childbirth. The timing of the reduction suggests the use of UI as a substitute for income replacement in the absence of paid leave. Importantly for welfare calculations, the reduction in UI benefits is substantial and represents almost half of the increase in maternity leave benefits. However, while this reduces the cost of extending maternity leave, it also reduces the mothers' willingness to pay for the policy as the additional maternity leave benefits are offset by the reduction in UI payments. Incorporating effects on UI substantively reduces the implied marginal value of public funds (MVPF) of the policy. We also document considerable heterogeneity of these estimates by pre-birth earnings.


  • Business Collaborations and Female Entrepreneurship (with Edward Asiedu, Monica Lambon-Quayefio, and Francesca Truffa)

Registered in the AEA RCT Registry, unique identifying number: AEARCTR-0006439 [Data collection in progress]

Search and contracting frictions can hinder the creation of business partnerships and the effectiveness of business collaborations. In developing countries, these relationships are often informal and conducted with family members and friends, suggesting that search costs and contracting frictions can be important barriers to firm growth.

We conduct an RCT in Ghana on a sample of 2000 female entrepreneurs to investigate the effect of an online matching service combined with access to legal information and advisory services on collaborations and firm performance. Specifically, we hypothesized that access to the matching service can alleviate search frictions. In addition, access to legal advisory may help mitigate contracting frictions by formalizing interfirm relationships and lowering risks of collaboration. The results of this study will shed light on the potential collaboration barriers faced by female entrepreneurs.


  • Effect of Beliefs and Gender Roles on Girls' Math Education (with Francesca Truffa)

Registered in the AEA RCT Registry, unique identifying number: AEARCTR-0003054 [Data collection completed]


Publications