Abstract
Structural and social inequalities are determinants of poor human development outcomes in low- and middle-income countries. In a quasi-experimental, regression discontinuity design study, we examined associations of affirmative action in local governments for a historically marginalized group in India with health and education outcomes, using administrative health data, primary data in schools, and interviews. Exposure was 60 village clusters with, and nonexposure was 60 village clusters without, local leadership reserved for scheduled caste (SC) in 2021 local elections. Primary outcomes were infant mortality rate (IMR) and student test scores. Secondary outcomes were prenatal visits, mother tetanus vaccinations, prenatal supplement, infant vaccination, and teacher engagement in schools. Groups did not differ on socioeconomic, geographic, and service delivery indicators in 2020 (baseline). In 2023, 2.5 y after elections, exposure village clusters had lower IMR (0.46 IRR; [0.30 to 0.69]; P < 0.001), and exposure village clusters had significantly higher odds of two prenatal visits, two tetanus vaccinations, and prenatal supplement received by the second trimester. Exposure was not associated with maternal care in the first trimester or infant vaccination. Exposure was not associated with test scores, but teacher engagement was lower in exposure village clusters. Exposure group leaders reported higher discrimination than nonexposure group, and in follow-up interviews, they described more difficulty in working with teachers. Affirmative action in local governments in India predicted greater infant survival and improved prenatal health services, indicating benefits of addressing structural inequities. Lack of benefits for education outcomes might reflect social inequity of discrimination experienced by SC leaders.