Under the threat of earthquakes, long-term policy makers need tools to optimally decide on the economic trajectories that will maximize the society welfare. Tools should be flexible and account for the consequences of earthquakes, incorporating the best estimate of their frequency and intensity. In this regard, this paper presents a modeling strategy that combines optimal control techniques and Bayesian learning: after an earthquake occurs, policy makers can improve their knowledge and adjust policies optimally. Some numerical examples illustrate the advantages of our modeling strategy along different dimensions. While Japan symbolizes the policy maker who has learned from earthquakes protecting the economy accordingly; Italy underlines the importance of prevention capital. China shows the hidden dangers of its extraordinary economic growth. Finally, the Chinese region of Yunan puts forward the roles of learning and of political independence.
