The Long-Run Socio-Economic Consequences of a Large Disaster: The 1995 earthquake in Kobe

Author Name William DUPONT IV  (Colby College) /Ilan NOY  (Victoria University of Wellington) /OKUYAMA Yoko  (Yale University) /SAWADA Yasuyuki  (Faculty Fellow, RIETI)
Creation Date/NO. March 2015 15-E-035
Research Project Post-disaster Recovery Policies and Insurance Mechanisms against Disasters: Case studies on earthquakes in Japan and floods in Thailand
Download / Links


In this study, we aim at quantifying the permanent socio-economic impacts of the Great Hanshin-Awaji (Kobe) Earthquake in 1995. We employ a large scale panel data of 1,719 wards (shi, ku, cho, son) from Japan over almost three decades. In order to overcome a fundamental difficulty of obtaining a clean control group, i.e., the Kobe economy without the earthquake, we adopt the synthetic control method of Abadie et al. (2010). Three important empirical findings emerged from our empirical analyses. First, the income level and the population size of the Kobe economy have been significantly lower than the counterfactual level without the earthquake over 15 years, indicating a significant permanent negative effect of the earthquake. Such a negative impact can be found especially in the central areas such as Chuo, Hyogo, and Nagata wards in Kobe, which are close to the epicenter. Second, the surrounding areas such as the city of Nishinomiya encountered positive permanent impacts with short-run negative effects of the earthquake. Third, the relatively outside areas such as the north (kita) wards of Kobe, the city of Akashi, and the city of Himeji seem to be insulated from the large direct and indirect impacts of the earthquake. In sum, there seem to be significant heterogeneities of the short-run and long-run losses caused by the earthquake even within the affected areas, suggesting that different market and non-market mechanisms function significantly to weather the impact of the earthquake spatially.