Household Health Responses to the Introduction of Highly Active Antiretroviral Therapy in Kenya

There is little work on the complementarity effects of HIV treatment programs on health investments, despite the salience associated with mitigation of HIV-related mortality risk. Estimates of such spillover effects are vital for cost-benefit assessments of HIV programs and for resource allocation decisions. I exploit temporal and spatial variation in the rollout of HIV treatment programs using difference-in-differences and instrumental variables to assess for these effects. I find that HIV programs in Kenya were associated with an increase in birth weights of around 90 grams (40 -- 290 grams). The evidence for such effects on vaccinations is mixed, with significant effects noted across all vaccines with the exception of those under the third schedule. I find weaker evidence for complementarity effects in BCG and DPT vaccine uptake rates. These results are robust to a number of robustness checks, two-stage least squares, sample restrictions to HIV negative subpopulations and Monte Carlo simulation regressions.