Working Papers

Natural Disaster, Economic Loss and Mental Health

with Prabal De (under review)

In this paper we estimate the causal effect of a natural disaster on mental health. Using longitudinal data from the Indonesian Family Life Survey, we investigate how the 2006 Yogyakarta earthquake impacted the mental health of individuals who were exposed to the earthquake to varying degrees. We also explore economic losses suffered from earthquake as a potential pathway to mental health deterioration. The results show that individuals exposed to the earthquake had a significantly higher likelihood of being screened for depression, and such effects were amplified by their experiences of economic loss. Differences in mental health outcomes between these two groups, however, were not significant during the pre-earthquake period. Finally, differences in the likelihood of being screened for depression decreased over time, demonstrating resilience among the individuals in our sample. Therefore, we can identify both the strongest adverse effects of a natural disaster on mental health in the immediate aftermath of such disasters, and their decline in strength over time.

The Educational and Fertility Effects of Sibling Deaths 

with Marc Rockmore and Willa Friedman (under review)

CINCH Working Paper

An emerging literature documents the long-term consequences of adverse events in childhood resulting in changes in adult outcomes and behavior. We extend this research to understand the influence of experiencing a sibling death as a child on subsequent educational and fertility outcomes in Indonesia. Using panel data and a sibling fixed effects model, we identify this relationship based on variation in the age of surviving children within the same family. Our findings strongly support the importance and persistence of adverse childhood experiences. In particular, for surviving brothers, experiencing a sibling death reduces the years of completed education and the likelihood of completing secondary schooling. The effect on surviving sisters is more muted. A potential channel for this result is that boys are pulled from school in the aftermath of a sibling death. Surviving the death of a sibling has little effect on desired fertility levels. Our findings overall suggest that interventions targeted at early-life outcomes may have important ripple effects and that the full impact of health interventions may not be visible until decades afterwards.

The Stability and Evolution of Risk Attitudes and Time Preferences After a Natural Disaster

with Marc Rockmore (please email for latest version)

By studying the short and medium run effects of the exposure to one of the largest recent earthquakes in Indonesia on risk and time preferences, we contribute to the literature on the stability of these parameters and the duration of the effects. We find a strong gender-specific effect one year after exposure which completely disappears within 8 years. Our analysis points to the importance of recovery efforts in mitigating the duration of the earthquakes effects on these parameters. We also rule out the mental health as a potential channel.