Abstract

Climate change mitigation and energy supply security are the most important issues worldwide. To attain carbon neutrality and restrict global warming to 1.5 °C, expediting investments in energy sectors and facilitating the expansion of clean energy is imperative. Nuclear power generation, as a base load, contributes to the energy transition because it is low-carbon and can be deployed on a large scale; however, particular challenges arise due to the risk of accidents, radioactive waste management problems, and public acceptance. Considering the similarity between macroeconomic data and nuclear power generation data from a time-series perspective, this study estimates the frequency of rare event occurrence in nuclear reactors using macro time-series methodology. We employ the Bayesian Markov-Chain Monte–Carlo method, developed by Barro and Jin (2021), to estimate the likelihood of rare event occurrences for the reactors using the panel data of electricity production (MWh) obtained from IAEA-PRIS. The findings indicate that the frequency of rare event occurrences significantly decreases as nuclear safety regulations strengthen and the regulatory workforce increases. Despite the limitations of being directly utilized as practical indicators for safety performance, observing its similar pattern with the World Association of Nuclear Operators performance indicators provides insights for policy implications in intermediate safety management aimed at preventing major disaster incidents.