California Center for Population Research
Flood Exposure and Child Health in Bangladesh
- Author(s): Buttenheim, Alison M.
- et al.
In the summer of 1998, Bangladesh was inundated by significant flooding that covered two-thirds of the country and affected more than 30 million people. Although annual flooding is normal and expected in Bangladesh, the 1998 floods caused extraordinary devastation and were considered a “century” flood. Homestead flooding, crop loss, and infrastructure damage compromised household food security and increased disease prevalence in a population with already high rates of poverty and malnutrition.
Unfortunately, this type of scenario has become increasingly common around the world: a significant crisis—whether environmental, economic, or political—devastates a large population of densely-settled households who are already trapped in chronic poverty. How do households anticipate and respond to such crises in the context of ongoing livelihood struggles? Do shocks affect investments in human capital? More specifically, what happens to children in the wake of such shocks? In this paper I use longitudinal data from the post-flood period in rural Bangladesh to examine how children’s human capital, as measured by nutritional status, responds to severe flooding and its aftermath. I emphasize the importance of analyzing these responses in a dynamic context, linking exposure to shocks and nutritional outcomes to longer-term measures of household vulnerability and resilience.
I pose two related research questions. First, did flood exposure in 1998 cause marginal growth faltering in children? To isolate the effects of the flood and address the endogeneity of flood exposure, I use a difference-in-difference estimator and village fixed effects. I also exploit the fact that younger children are more vulnerable than older children to nutrition shocks. I next ask whether the effects of flooding on child growth faltering were mediated by household resources, hypothesizing that households with lower levels of pre-flood resources are less able to protect children from nutrition shocks.
These analysis help to answer several important policy questions related to crisis and recovery in vulnerable populations. The results reveal the extent to which children were nutritionally compromised by the flood, and which children fared worst. The analyses also contribute specifically to the design and implementation of livelihood interventions, and relief and recovery efforts. Can households use physical, financial and human capital to protect children’s nutritional status from significant shocks to income and food security? If so, is it more effective to focus on long-term asset-building strategies in vulnerable populations, or to facilitate asset recovery post-shock through access to credit and other forms of relief? Given the increasing exposure to shocks and the quantity of resources allocated post-disaster to relief and recovery, these questions are not trivial.