Economic and financial crises are characterized by tail events. When they occur, tail correlations emerge, which have linear and non-linear origins. The former are due to the Pearson correlations, while the strength of the latter depends on the heavyness of the tails. We introduce TailCoR, a new metric for tail correlations that disentangles straightforwardly the linear and non-linear correlations. TailCoR is simple to compute, no optimizations are needed, and it performs well in small samples. When applied to a panel of eight major US banks, TailCoR increases during the financial crisis because of a surge in both the linear and non-linear correlations. The end of 2012 also shows an increase of TailCoR, which is solely driven by the non-linearity, reflecting the risks of tail events and their spillovers associated with the European sovereign debt crisis.
Homepage of David Veredas