Bayesian

An Empirical Analysis of Health Shocks and Informal Risk Sharing Networks

This paper investigates informal risk sharing against health shocks in the presence of multiple risk sharing networks. We use a panel household survey data from rural Ethiopia that covers the period 1994--2004. We find that neither short-term nor long-term health shocks are insured through transfers from networks such as friends, neighbors, and members of informal associations. However, networks related along bloodline such as extended family members provide assistance when health shocks are long-term such as disabilities. The results show that these networks strategically complement planned component of their transfers which are made on a regular basis such as remittance, entitlement, or chop money (small cash sums for household expenses). Moreover, we find significant history dependence in transfers from not only genetically distant networks but also extended family members as well as formal institutions, which seems to discourage dependency. Finally, the findings suggest significant heterogeneity in transfers.

Evaluating Endogenous Program Interventions with Heterogeneous Treatment Intensity using Bayesian Potential Outcomes Approach

In this paper, we implement a Bayesian potential outcomes model to evaluate the impact of program interventions using non-randomized data. The approach jointly addresses selection bias in program placement, heterogeneous treatment intensity among the treated, and heterogeneity in treatment effects. Using data from a non-randomized household survey, we evaluate the impact of Ethiopia's Health Extension Program on fertility and child mortality outcomes. We find that there is significant selection bias in both program placement and intensity of exposure to the program among the treated. On average, the program has significant impact on reducing fertility and child mortality. However, there is notable heterogeneity in the treatment effects ranging from negative impacts for some individuals to positive impacts for the majority in the sample. We recover individual-level treatment effects and present the distributions graphically.