Intrahousehold Resource Allocation in Rural Africa: Evidence from a Randomized Field Experiment

I use a novel experiment to study intra-household allocation of resources in rural northern Ghana. In an experiment that provided subsidies towards the purchase of health insurance, the total value of the subsidies were not enough to enroll all household members. 200 households were randomly assigned to receive no subsidy vouchers, subsidy vouchers that specified a subsidy amount for each household member or a subsidy voucher that allowed households to decide which members to enroll to allocate. Comparison of enrollment patterns under the two types of vouchers shows that when allowed to allocate subsidy amounts, households enroll more boys than girls by 11.7 percentage points. This gender differential is concentrated in children aged 7-17 years. I find suggestive evidence that differential labor market participation rates is one explanation for this allocation pattern: among children aged 7-17 years, boys are 3.7 percentages points more likely to participate in the labor market as farm workers. These results suggest that households may be investing more in boys because of their contributions to household farm output.