REPLICATION STUDY: Toya and Skidmore (Economics Letters, 2007

This study replicates the empirical findings of Toya and Skidmore (2007), hereinaf¬ter “TS”, and performs a variety of robustness checks. We are able to exactly replicate the find¬ings reported by TS. Our robustness checks consist of two parts. Firstly, we update TS’s orig¬in¬al data set, both with respect to variable values and years. We then address a number of estim-ation issues: (i) truncation bias, (ii) the effect of severe skewness in the disaster data, (iii) fixed effects, and (iv) the omission of a time trend. Our robustness checks produce two major results: We confirm TS’s finding that income is negatively related to both fatalities and econ¬omic losses from disasters. In fact, we estimate coefficients that are substantially larger than TS. On the other hand, we find no evidence to indicate that the other economic devel¬op¬ment variables (educational attainment, size of government, economic openness, financial sec¬tor development) are statistically related to either fatalities or economic damages.