Wednesday, March 19, 2008

you could get hit by a bus

“Consumption Smoothing and Household Responses to Health Shocks: Evidence from Random Exogenous Health Shocks” CID Graduate Student and Postdoctoral Fellow Working Paper No. 23, January 2008

Endogeneity in the health-wealth relationship is one of the biggest challenges in studying the causal effect of health on household consumption, wealth and labor responses. In my dissertation research, I introduced an innovative study design that helped get around this endogeneity problem in the health-wealth relationship. My empirical strategy relies on health shocks that were suffered by passengers injured in a bus accident, with “controls” drawn from travelers on the same bus routes, matched on age, gender, area of residence. The matching strategy ensures that the exogenous health shocks from the bus accident are random, conditional upon matching. Using data from my household survey conducted one year after the accidents, I find evidence of imperfect consumption smoothing and large effects on household debt. Debt was the principal mechanism used by households to mitigate effects of the shock. The shock related expenses, equal on average to two months of household income, caused exposed households to reduce educational expenditures by roughly 20% and festival expenses by 9%. I find that the odds of having debt among exposed households were five times higher and the average size of debt was almost twice as much compared to the unexposed. These findings inform two key areas of research and policy. First, the large effect of health shocks on household debt in developing countries can exacerbate long-term financial insecurities through potential default of debt or depletion of assets as a result of the high debt burden. Second, these findings lend support to the argument that the conventional focus on consumption smoothing alone as a measure of household welfare ignores the fact that households resort to costly mechanisms to mitigate effects of health and income shocks. I presented an earlier version of this paper at the 2008 American Economics Association Conference in New Orleans (available here).

Download latest version of paper here

No comments: