Association for Public Policy Analysis & Management

According to the U.S. Geological Survey (USGS), earthquake alone has led to more than 800,000 deaths worldwide between 2000 and 2015. More specifically, Asia comprised almost 40% of the global natural disaster damages and 90% of the victims worldwide between 2001 and 2010 (Guha-Sapir et al., 2012). The recent earthquake in Nepal in 2015 is a case in point, which affected approximately 8 million people, resulting to a total economic loss in the order of US$ 10
billion, equivalent to about a half of Nepal’s gross domestic product (Goda et al., 2015).

Given that there exists well-documented evidence on long-term repercussions of shocks in early childhood on cognitive development (Almond, 2006), welfare consequences of a massive earthquake in the context of a developing country are of a major public policy concern. Consistent with a large number of growing literature (Caruso, 2015; Caruso and Miller, 2015; Tan et al., 2015), we take advantage of the exogeneity of the earthquake in 1988 and evaluate the long-term repercussions of the natural disaster on human capital accumulation of the affected generation.

Nepal represents an excellent context to investigate a causal association between the incidence of a massive natural disaster and human capital accumulation. To evaluate the long-term repercussions of a major earthquake on years of educational attainment, we examine the educational outcomes of a sample of children who were 0-2 years old in age during the time of the earthquake in 1988. Specifically, we compare the educational outcomes of children from earthquake-affected and earthquake-free districts, and treat older cohort that has attained complete education from earthquake-affected and earthquake-free districts as a relevant control group.

The empirical results indicate that infants exposed to earthquake completed on average 0.6 years less schooling compared to their counterparts. Consistent with prior literature on potential gender bias (Neumayer and Plumper, 2007; Caruso, 2015; Tan et al., 2015), we are able to demonstrate that massive natural disasters turn out to be worse for females, who end up completing almost 0.9 years less schooling compared to male counterparts. We also examine two potential mechanisms that possibly explain the adverse impact of the earthquake, prevalence of poverty trap and negative health shock arising from the earthquake.

The study improves upon the existing literature in a number of ways. First, it employs a difference-in-differences study design to account for potential bias originating from time-invariant differences between infants from earthquake-affected and earthquake-free districts. Second, it performs different sets of placebo tests that strengthen the validity of the methodological approach used and provide substantial confidence in our primary findings. Finally, this is the first study that takes advantage of rich household data from a developing country in South Asia and employs a quasi-experimental technique to assess the causal impact of an earthquake on human capital accumulation.