Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
Urban Indian households with a male first child are approximately 2 percentage points more likely to use clean cooking fuel than comparable households with a female first child. Given Indian son preference, there are at least two mechanisms by which child sex could affect fuel choice: by improving the intrahousehold status of women, who bear more of the costs of traditional fuels, or by presenting an opportunity to invest in children's health, in the context of a preference for healthier boys. If child sex is not selected for by biased abortion or other processes, then the sex of a first child has an exogenous causal effect on household fuel choice. We show that the association between fuel choice and child sex is not driven by terminated pregnancies or by household wealth or family size. Among a range of outcomes we study, the effect of child sex is unique to fuel choice; our finding that there is no effect on other assets indicates that it is unlikely that the result is confounded by real or subjectively anticipated wealth. In addition to the National Family Health Survey NFHS-3, the main data source studied, we approximately replicate the result using the NFHS-2 and the District Level Health and Facilities Survey DLHS-3. Finally, we show evidence for a "first-stage" effect of having a first son on women's social status: such women have a greater body mass index, on average.