Research

Working Papers

Minimum Wage, Informality, and Gender Gaps: Evidence from Morocco

Abstract: This paper examines how minimum wage policies affect gender gaps in employment and wages in a setting characterized by a large informal sector and wide initial gender disparities. Focusing on the Moroccan manufacturing sector, I leverage matched formal employer-employee data to analyze the impacts of a 24% real increase in the national minimum wage between 2009 and 2015. During that period, the raw gender pay gap in the formal sector narrows from 28% to 22%. Using difference-in-differences designs, I find that this decline is driven by direct wage increases for workers previously below the new minimum wage and to spillover effects for workers higher up in the wage distribution. Both direct and spillover effects are larger for women. Female workers who remain in formal employment are also more likely to transition to larger, higher-paying firms. However, I also document a displacement effect: women directly affected by the minimum wage increases are 22% more likely to exit formal employment, while men and women in higher wage brackets remain unaffected. Most displaced women transition to the informal sector, where wages are substantially lower. Firm closures, which disproportionately affect firms employing more women, account for 40% of this displacement effect. Examining heterogeneity across local labor markets, I further show that the displacement effect is larger when women have a reduced set of formal sector outside options. Despite this increased job vulnerability for low-wage women, bounding exercises suggest that, on net, the minimum wage increases still contribute to a reduction in the gender pay gap.

Can Female Directors Shrink the Gender Gap? Evidence from France, with Francois-Xavier Ladant.

Paper || Media coverage

Abstract: This paper investigates how the gender composition of corporate boards impacts gender gaps within the firm, and whether this leads to an equity-efficiency trade-off. Harnessing a comprehensive panel of French firms from 2006 to 2021, we exploit a 2010 gender quota as an exogenous shock to board composition. In response to the quota, publicly listed firms increase their female board share from 11% to 42%, while non-listed firms exhibit a more modest change. Using firms' pre-quota listing status in difference-in-differences and IV strategies, we show that greater female board representation improves labor outcomes for women: it increases female representation among top earners and the likelihood that a woman becomes CEO, while reducing gender wage gaps across the wage distribution. These gains do not come at the expense of the firm's financial performance: despite higher labor costs, profitability remains unchanged. Boards with higher female representation leverage executive compensation to incentivize CEOs to address gender disparities, and female directors' effectiveness appears to be enhanced by their prior top executive experience and access to key board committees.

Can a Website Bring Unemployment Down? Evidence from a French Online Platform, with Aïcha Ben Dhia, Bruno Crépon, Esther Mbih, Bertille Picard, and Vincent Pons. Revise and resubmit, Journal of Labor Economics.

Paper || Media coverage

Abstract: We evaluate the impact of an online platform giving job seekers tips to improve their search and recommendations of new occupations and locations to target, based on their personal data and labor market data. Our experiment used an encouragement design and was conducted in collaboration with the French public employment agency. It includes 212,277 individuals. We find modest effects on search methods: the platform’s users adopt some of its tips and are more likely to use resources provided by public employment services. However, following individual trajectories for 18 months after the intervention, we do not observe any impact on time spent looking for a job, search scope (occupational or geographical), or self-reported well-being. Most importantly, we do not find any effect on any employment outcome in the short or medium run. We conclude that the enthusiasm around the potential for job-search assistance platforms to help reduce unemployment should be toned down.

Work in Progress

Impact of Moroccan Active Labor Market Policies Directed at Firms, with Bruno Crépon, Florencia Devoto, and Rema Hanna.

Abstract: This project assesses whether and how offering hiring support to firms can increase employment, the quality of job matches, and the diversity of the hiring pool. We have partnered with the Moroccan Employment Public Agency — Agence Nationale de Promotion de l'Emploi et des Compétences (Anapec) — to experimentally test the impact of their e-prospection service. This service involves reaching out to firms with scripted messages to promote PES services, such as wage subsidies, training subsidies, and job placement assistance. When a firm shows interest, Anapec caseworkers offer personalized assistance to facilitate their access to these services. The main goal of this intervention is to bridge the gap in information between employers and job seekers, with the final aim of enhancing employment opportunities and matching processes. We are particularly interested in understanding how these e-prospection and intermediation services can benefit young people and women, who often lack access to traditional job networks.

Take-up of Parental Leave Policies: The Role of Managers, with Francois-Xavier Ladant.

Abstract: Paid Family Leave (PFL) policies are designed to support new parents as they navigate the often-conflicting demands of work and family responsibilities. Beyond government mandates, firms are actively expanding parental leave benefits as a strategy to attract and retain talent. However, the determinants of employees' take-up are unclear. While previous literature has shown that coworkers' behaviors could have a significant impact on program participation, the role of managers remains unknown. Specifically, if information friction leads employees to hold incorrect beliefs about their manager's stance on parental leave, such firm programs could be underutilized. In this paper, we combine confidential HR data from a major French multinational corporation and employee surveys to examine how information friction influences the take-up and duration of parental leave.

Gendered Impacts of Group-Based Formal Lending: Evidence from Indian Villages, with Christine Binzel, Erica Field, and Rohini Pande.

Abstract: Theory and existing evidence support the idea that formal lending reduces informal lending in lower-income settings while disrupting established social networks. In South Asia, much of the formal lending to the rural poor is group-based and targeted at women. Moreover, this is a context in which individuals' social networks show gender-specific homophily. Does gender-specific homophily extend to informal borrowing networks, and if so, how does the gendered nature of formal lending impact the networks of husbands and wives? What are the implications of any gendered disruption of informal financial networks for gender gaps in informal borrowing capacity? Credible answers to these questions require an exhaustive and separate mapping of men's and women's informal financial relationships (with information on potential borrowing capacity from each network member), as well as significant exogenous exposure to formal credit at the village level. This paper combines a large-scale randomized bank branch rollout with comprehensive social network mapping spanning 205 villages and 19,000 spouse pairs in Tamil Nadu, India. At the outset, we see that both men and women have the same number of informal financial links, but men have twice the informal borrowing capacity. Consistent with gender homophily in social networks, men's networks include other men with high borrowing capacity. Women, on the other hand, rely on other women with lower borrowing capacity. Two years after the randomized branch rollout, administrative bank data shows that 36% of women and 3% of men had taken out a formal loan. Alongside, household surveys demonstrate that informal borrowing decreased for both genders, with men reporting a more pronounced decline. Treatment reduces gender gaps in informal borrowing capacity. When examining the financial networks of households in treatment villages, the average node degree and network density decreased while the average closeness and number of components increased. Ongoing work examines whether the within-village reduction in links across different risk-sharing groups varies by gender.

Does the Arrival of a Formal Financial Institution Alter Community-Level Cooperation? Experimental Evidence from Indian Villages, with Christine Binzel, Erica Field, and Rohini Pande.

Abstract: Processes of economic development are frequently associated with greater access to formal financial markets and lower reliance on informal loans. How does reduced financial interdependence among community members influence village cooperative arrangements and the capacity of social networks to provide insurance and public goods? Predictions from theoretical models of networks are ambiguous. We combine the randomized rollout of bank branches in rural India with full network mapping data from 205 villages to examine the spillover effects of access to bank loans on village-level risk-sharing capacity and public goods provision. The scale of our intervention implies that we can capture the general equilibrium effects of formal financial intermediation on network structure. Two years after the randomized rollout, 33% of households in treatment villages are clients of the rural bank, which is accompanied by a significant decline in informal borrowing. We show that this substitution away from informal financial intermediation is associated with a large reduction in the insurance capacity of village networks, as measured by the maximum amount that households state they could borrow in times of need from each of their links. Network structure is also impacted by the presence of a bank – average closeness falls suggesting that individuals are less connected to further away households, and households are associated with more tightly knit groups as reflected by a higher clustering coefficient and lower node degree. Finally, we see evidence of reduced cooperative capacity in the form of lower levels of public goods provision on networks in banked villages, in terms of individual time and money spent on village-level public goods.