In this paper, we examine how the gender of a sibling affects education, age at first birth and age at first marriage. Identification of the effect is hard because parental preferences for children’s sex composition can confound the analysis. We circumvent this problem by using a sample of dizygotic twins: in these cases, the two children are born at the same time, so parents cannot make decisions about the second-born twin based on the gender of the first-born twin. We find that women obtain less schooling, marry earlier and have children earlier if their cotwin is a sister instead of a brother. We examine several mechanisms and we find that the only way to explain the results on education is by taking the effects on family formation into account.
Having a sister changes preferences and/or abilities such that women form families earlier and they obtain lower education as a byproduct of these mechanisms.