Village sanitation externalities and children's human capital: Evidence from a randomized experiment by the Maharashtra government

Open defecation is exceptionally widespread in India, a country which also suffers some of the world's worst rates of child stunting. We study a randomized controlled trial of a village-level sanitation program, implemented in one district by the government of Maharashtra. We find that the program caused an average increase in child height that was large but plausible given other estimates in the literature. In evidence of sanitation externalities, this effect is found even on children in households that did not adopt latrines. Unusually, we also have comparable data from other districts where the government planned but ultimately did not conduct an experiment, allowing a consideration of the importance of the population chosen to be eligible for experimentation. We demonstrate techniques that respond to a recent critique of the small samples of clusters in many cluster-randomized field experiments in development economics.