Same concerns, same responses? A Bayesian quantile regression analysis of the determinants for supporting nuclear power generation in Japan

Abstract

Using internet survey data from 6,500 individuals, this study examines the determinants for supporting the restart of nuclear power plants operation in Japan. As in previous studies, the variable of interest is a categorical and ordered variable that measures the level of support, for which ordered logit or ordered probit is commonly estimated. This study departs from the literature by using Bayesian ordinal quantile regression recently proposed by Rahman (2015). This approach allowed us to explore whether covariates have differential effects at various conditional quantiles of the latent response variable, which can be interpreted as the willingness to support the restart. The results show that for most of the covariates we examined, including concerns about meltdowns and the storage of nuclear material and concerns about global warming, the effects differ across conditional quantiles. In other words, the covariate effects depend on individuals’ unobserved preferences for the restart (conditional on observables). The results also show that for some covariates, the effects differ considerably across gender.