Relatedness and investment in children in South Africa

Abstract

Investment in children is examined using a nationally representative sample of 11,211 black (African) households in South Africa. I randomly selected one child from each household in the sample and calculated the average genetic relatedness of the other household members to the focal child. Using multivariate statistical analysis to control for background variables such as age and sex of child, house-hold size, and socioeconomic status, I examine whether the coefficient of relatedness predicts greater household expenditures on food, on health care, and on children's clothing. I also test whether relatedness is associated with health and schooling outcomes. The results are consistent with an inclusive fitness model: Households invest more in children who are more closely related. Two exceptions were found: in rural areas, genetic relatedness was negatively associated with money spent on food and on health care. Explanations for these results are discussed.