In India, as in many developing countries, female sterilization is the main contraceptive method: 37% of women older than 25 are sterilized. However, no economic study assesses the effect of sterilization, providing guidance on efficient reproductive health policies. We analyze the consequences of sterilization for maternal health, considering the endogeneity of the decision. We exploit that Indian households face different infant mortality risks - driven by malaria prevalence - and have a son preference. Sterilization increases when women have a boy first-born, but less so when they live in a malarious area, as they fear losing the boy; this situation provides an instrument. We show that sterilization strongly increases the prevalence of various symptoms in the reproductive sphere while also reducing the risk of anemia, likely from avoiding pregnancy. This paper is the first to assess the effect of a specific contraceptive method with a clear identification strategy.