We offer a #datadriven theory of #belief formation that explains sudden surges in economic #uncertainty and their consequences. It argues that people, like #bayesian econometricians, estimate a distribution of macroeconomic outcomes but do not know the true distribution. The paper shows how real-time estimation of distributions with non-normal tails can result in large fluctuations in uncertainty, particularly related to tail events or "black swans." Using real-time GDP data, the authors find that revisions in estimated #blackswan #risk explain most of the fluctuations in uncertainty. These findings highlight the importance of #accounting for the effects of uncertainty and non-normality in economic decision-making and #policymaking.