Event study analysis on changes in residential property values between treated and control districts.

Note: This figure plots the point estimate and 95% confidence interval of the coefficients on TreatedXPost from the event study specification in Equation (4). The dependent variable in the model is a self-assessed residential property value (in logs) for an individual household. The empirical specification includes a full set of dummies extending from 15 months before the earthquake to 24 months after and their interaction with the treatment indicator, district-specific quadratic monthly time trends, year-by-village fixed effects, demographic and housing characteristics such as household size, location type (urban or rural), number of rooms, number of years since the construction of the house, water supply type and access to electricity. Standard errors are clustered at the district level. Source: Author’s calculations.