Anatomy of a Sovereign Debt Crisis : Machine Learning, Real-time Macro Fundamentals, and CDS Spreads

We study the time-varying dependence of sovereign credit default swap (CDS) spreads on real-time, country-specific macro indicators during the eurozone sovereign debt crisis. Macro fundamentals explain 66% of the time-series variance of CDS spreads, but the time variation in macro sensitivities is also important, explaining close to 30% of the variance. Hence, while CDS spreads reflect macro fundamentals, they also display volatility unrelated to fundamentals. We identify distinct "regimes" of variation, consistent with a multiple-equilibrium view of the sovereign debt markets. Our estimated macro sensitivities predict the VSTOXX European equity volatility index out of sample