A model-based index for systemic risk contribution measurement in financial networks

Measurement of the potential risk of an institution’s failure to other institutions is essential in detecting vulnerable institutions. Based on the famous model proposed by Eisenberg and Noe (2001), we introduce a novel model-based systemic risk contribution index to quantify the marginal negative externality from the default institution to other institutions. The proposed index clearly distinguishes the contributions of direct and indirect risk transmitted through immediate liability connections and mediate liability chains, respectively. More importantly, it can identify susceptible institutions conditional on a default, which is critical for regulators to take targeted steps to break the chains of risk contagion. As an illustration, we analyze the systemic risk contribution between country pairs using a dataset of 18 Eurozone countries during the European Sovereign Debt Crisis. We find that direct risk contagion dominates in general, but indirect risk contagion still plays a nonnegligible role.