Gender Bias in Breastfeeding and Missing Girls in Africa: The Role of Fertility Choice∗

This paper investigates whether there is gender bias in the duration that children are breastfed in Africa. Given evidence that breastfeeding is negatively related to future fertility, we further investigate whether any part of the gender bias is due to son preference in fertility choice. We use identical methodology to Jayachandran and Kuziemko (2010), and compare our results on Africa to their findings on India. We present separate results for North and Sub-Saharan Africa to account for differing regional levels of gender discrimination, and use a sample of over 100,000 children from 32 waves of DHS surveys across 17 countries. We find that boys are breastfed for 0.657 months longer than girls in North African countries, which is nearly twice male advantage of 0.391 months found for India. For Sub-Saharan African countries, the male breastfeeding advantage is much smaller at 0.059 months. We also find evidence analogous to Jayachandran and Kuziemko (2010) linking son-biased fertility choice to breastfeeding duration, with children being breastfed longer as birth order increases, as mothers approach or exceed their ideal total fertility, and if mothers already have a male child. Having older male siblings reduces the male advantage in breastfeeding in North Africa, but not in Sub-Saharan Africa. We estimate that annually approximately 43,000-45,000 girls in Sub-Saharan Africa are missing due to gender discrimination in breastfeeding. JEL classification: I14, J16, O15, O55