Robust Financial Networks

In “Robust Financial Networks,” F. Hu, D. Mitchell, and S. Tompaidis study networks of financial institutions where only aggregate information on liabilities is available. The authors introduce the robust liability network, that is, the network consistent with the available information that exhibits the worst expected losses. They provide an algorithm to identify the robust liability network and, using aggregate data provided by bank holding companies to the Federal Reserve in form FR Y-9C, determine robust liability networks for U.S. banks under various network configurations. They show that the robust liability network is sparse, with links between institutions that hold highly correlated portfolios. They illustrate the methodology in two applications. (1) They look at how robust liability networks changed around the onset of the COVID-19 pandemic. (2) They evaluate the impact of a potential regulation that limits risk-taking based on each institution’s conditional value-at-risk. Their results can be used by regulators to monitor systemic risk in financial networks.