Abstract:
This paper investigates whether early-life adversity (ELA) affects later-life health in India using nationally representative LASI data and quasi-experimental variation from cohort differences and the Bengal Famine. We estimate probit models, KHB mediation, and GSEM across seven outcomes: underweight, food insufficiency, poor self-rated health, low cognition, depression, ADL, and IADL. ELA shows a monotonic dose-response relationship with all seven outcomes. Education mediates most of the ELA effect on cognition and IADL, while life satisfaction dominates for depression. Direct effects persist, indicating biological scarring. The Bengal Famine produced lasting health deficits, except for IADL where mortality selection dominates. Heterogeneity reveals larger cognitive penalties for women, stronger physical/functional penalties in Central/Eastern India, and wealth buffers underweight but not depression or self-rated health. These findings survive extensive robustness checks (IV-PSM, inverse probability weighting, alternative ELA measures, birth cohort analysis, and machine learning). Results imply that early-life interventions, especially universal education, can substantially reduce later-life health inequalities in India.
