Systemic Risk and the Macroeconomy: An Empirical Evaluation,” working paper

We propose a criterion to evaluate the empirical relevance of systemic risk measures based on their ability to predict low quantiles of real macroeconomic aggregates. We also propose and evaluate methodologies for constructing systemic risk indices that capture the joint information content of a large cross-section of systemic risk measures. We construct over 20 measures of systemic risk in the US and Europe extending across several decades. We show that, taken individually, these measures reveal low predictive ability for macroeconomic downturns. However, an index that parsimoniously aggre-gates individual measures consistently performs well in forecasting downturns both in-sample and out-of-sample. ∗We thank Lars Hansen for many helpful conversations about this project. We also thank John Cochrane, Doug Diamond, Amir Sufi, Amit Seru, and seminar participants at University of Chicago (Booth) for helpful comments. We thank Christian Brownlees, Mark Flood, and Andrew Lo for sharing Matlab code.