Christianity and girl child health in India

•This paper finds that Christian children 0–59 months of age are less likely to be stunted as compared to non-Christian children.•The advantage is especially evident for girls, less so for boys.•Mechanisms include access to historical schools and hospitals/pharmacies that raised the human capital of women over a century ago.•Son preference in Hinduism, by reducing the relative value of women, is also a factor that explains Christian advantage.•Results are robust to the inclusion of a variety of controls and specification tests. This paper studies child health focusing on differences in anthropometric outcomes between Christians and non-Christians in India. The non-Christian group includes Hindus and Muslims. Estimates indicate that young Christian children (ages 0–59 months) are less likely to be stunted as compared to similar aged children of Hindu and Muslim identities. The Christian relative advantage is particularly pronounced for girls. Using representative data on child health outcomes and information on the location of Protestant and Catholic missions, differences in the relative timing of establishment of missions in the same area, political crises that mission-establishing countries were engaged in during India's colonial history, and historical information from the 1901 Census, we find that Christian girls are significantly less likely to be stunted as compared to similarly aged non-Christian girls. We find no relative stunting advantage for Christian boys, which we attribute to son preference and patriarchy among Hindus in particular. An analysis of explanatory mechanisms indicates that elementary and higher education schools, as well as hospitals, pharmacies and print shops associated with the advent of Christianity improved the relative human capital of women with subsequent long-term implications for young Christian girls in India today. Our results survive a series of robustness and specification checks.