Affirmative Action and Career Consequences: Evidence from India

• In Event: Gender and Opportunity in Higher Education and Work

Abstract

A small number of elite institutions disproportionately shape access to top paying careers, with highly selective universities in the United States channeling graduates into high paying occupations at rates far exceeding their enrollment share. In India, the Indian Institutes of Technology (IITs) occupy an analogous position in the production of economic elites, with IIT alumni leading some of the world's largest technology companies and commanding substantial labor market premiums. Yet access to these institutions has remained deeply unequal along gender lines: while women account for approximately 30% of STEM enrollment in India, female representation at IITs remained below 10% prior to 2018, severely constraining women's entry into the elite career pathways that IIT affiliation confers. This paper examines whether expanding women's access to elite technical education through affirmative action improves their labor market outcomes, specifically their likelihood of securing placements at elite firms. I leverage the Supernumerary Seats Scheme (SSS), a gender targeted affirmative action policy introduced by the Government of India in 2018 that created additional seats reserved exclusively for women within each program IIT cell , without displacing existing male admits, with the goal of raising female enrollment to 20% by 2020. In other words, programs with historically lower female representation received proportionally more supernumerary seats and thus experienced greater policy exposure, while programs closer to the target received fewer additional seats. I use two primary data sources: individual level LinkedIn profiles from Revelio Labs, containing information on institution, program, graduation cohort, gender, and employer identity; and IIT brochures supplemented with contemporaneous newspaper coverage to identify elite firms. The treatment intensity is measured as the ratio of supernumerary seats created in 2020 to baseline enrollment within each IIT and program cell, which varies based on how far each program's pre policy female enrollment was from the 20% target, generating exogenous variation in policy exposure. To identify causal effects, I employ a triple difference-in-differences framework comparing the gender gap in elite placements across IIT and program cells with high versus low policy exposure, before and after 2018, with fixed effects absorbing program level time trends and time invariant differences across IIT and program cells. The baseline results indicate that the gender gap in elite firm placements improved significantly more in programs with greater policy exposure, that is, historically male dominated programs where the most seats were added. The women in high policy exposure programs maintained same placement share despite a substantial increase in the applicant pool, while women in less exposed programs experienced declining rates. I do not find evidence that men's outcomes respond to the policy. Additionally, I find that the improvement is driven entirely by entry into firms that previously recruited from these programs but had hired only men, with no differential change at firms that already hired women, consistent with firms fulfilling previously constrained demand for women from these programs. This pattern is difficult to reconcile with peer effects and further, I do not find evidence of sorting on ability.

Author

• Lakshita Sharma, Cornell University
  Presenting Author