Learning and Expectations in Dynamic Spatial Economies

The impact of shocks in dynamic environments depends on how forward-looking agents anticipate the path of future fundamentals that shape their decisions. We incorporate flexible beliefs about future fundamentals into a general class of dynamic spatial models, allowing beliefs to be evolving, uncertain, and heterogeneous across groups of agents. We show how to implement our methodology to study both ex-ante and ex-post shocks to fundamentals. We apply our method to two settings: an ex-ante study of the economic impacts of climate change, and an ex-post evaluation of the China productivity shock on the U.S. economy. In both cases, we study the impact of deviations from perfect foresight on different outcomes.