UC Berkeley

Economic factors affect nearly every human decision, even those we consider most personal. This body of work demonstrates the presence of economic reasoning in the unexpected realms of sexuality and reproduction. Focusing on Sub-Saharan Africa, where financial concerns are often pressing, I show evidence for the influence of economics on child-rearing, sexual behavior, and reproduction, respectively.

On child-rearing, I model the parents' decision regarding investments in their children, in particular, investments in child health. The question of interest is how a child's cohort within the household affects the amount and type of investment she receives. Partitioning investments into private goods and club goods for children reveals that the size and gender composition of one's cohort affect these types of investments in opposite ways. I use data from Senegal to test the prediction that, ceteris paribus, children with larger (and more male) cohorts will receive more club investments. Employing a new method to deal with the endogeneity of siblings in this type of analysis, I exploit a unique characteristic of this data. The predominance of multi-family, cooperative households in this setting offers the existence of non-sibling cohorts that can instrument for a child's full cohort within the household. Using a 2SLS within-household estimation, I find that club investments are increasing in cohort size and male composition. This finding is particularly relevant to child health in Africa, where club investments such as water purification, bed nets, and immunizations could prevent sixty percent of child death.

In addition, this work builds on existing theories regarding sexual behavior responses to low-income shocks. Social scientists have suggested that African women use transactional sex for both income smoothing and insurance. In an environment of epidemic HIV, increases in casual partnerships, or increases in risks taken within partnerships can increase HIV-risk to a woman and her community. This work shows evidence of this dynamic. I employ individual serostatus data and overlay it with historical, village-level weather data across 19 countries in Africa. I find that when droughts cause economic hardship in rural Africa, women are significantly more likely to become infected with HIV. Concentration of this effect among women of little means, and the presence of a counterpart effect in men of great means, suggest a behavioral pathway. These findings indicate that crop insurance and social safety nets could significantly stem the spread of HIV in Sub-Saharan Africa.

This work also highlights the impact that economic policies can have on reproductive decisions, even taken at a great distance. Several times in the past quarter-century, the US has employed an economic policy to achieve a social objective: reduce the use of abortion abroad. By withholding funding to certain foreign NGOs, the policy rather had the effect of reducing the availability of contraceptives in poor countries. I estimate the impact of the policy on the use of abortion in Ghana, creating a woman-by-month panel over 25 years and exploiting the off-on-off-on history of the policy. Findings suggest that the policy did not reduce the use of abortion. Further, the reduced contraceptive availability resulted in increased pregnancy rates for rural women &mdash the explaining factor for why the policy increased use of abortion among these women. In Ghana alone, the policy resulted in nearly 100,000 additional abortions and up to 500,000 additional unplanned births.