Threshold effects of extreme heat on schooling and child labor in rural Bangladesh

Identifying threshold effects of extreme heat is key to understanding the true scale of climate-related risks to human capital development. This paper investigates how extreme heat shapes adolescent schooling and labor outcomes in rural Bangladesh, combining household survey data on adolescents with high-resolution temperature records to estimate the effects of prior-year, cumulative, and early-life heat exposure. We identify a precise temperature threshold at 36°C, above which each additional day reduces school attendance by 3.1 percentage points and increases child labor by 2.5 percentage points. Below this threshold, moderate heat (30-36°C) shows minimal single-year effects, though cumulative exposure over three years reveals significant negative impacts, indicating limited household adaptation. Effects are disproportionately concentrated among girls, who shift primarily toward household work rather than wage labor. Three interconnected channels drive these effects: heat-induced income shocks (11% reduction in household income), increased domestic labor demands from heat-related illness, and restrictive gender norms that amplify these impacts by magnifying girls’ household responsibilities. Extending the analysis to early-life conditions, exposure during the first 1,000 days also reduces adolescent schooling probability by 3.4-3.8 percentage points, with strongest effects at ages one and two. Boys show slightly larger early-life effects, contrasting with girls’ greater vulnerability to contemporaneous exposure, suggesting distinct mechanisms operating through biological development versus gendered household labor allocation. The findings point to both immediate income-mediated responses and long-term developmental pathways, with implications for temperature-triggered social protection, school infrastructure investments, and early-life health interventions.