Multilayer information spillover networks: measuring interconnectedness of financial institutions

We propose multilayer information spillover networks to measure the interconnectedness of financial institutions by comprehensively considering mean spillover layer, volatility spillover layer and extreme risk spillover layer based on the Granger causality tests in mean, volatility and risk. Using daily returns of 24 Chinese publicly listed financial institutions during 2008–2018, we construct static and dynamic multilayer information spillover networks and analyze different layers' similarity, uniqueness and overlap. Some unique features, which could not be detected in a particular single-layer, are found in multilayer networks. Dynamic topological features of multilayer networks show that significant changes in degrees or unique edges on extreme risk and volatility spillover layers generally occur in the period before a financial stress, e.g. the beginning of the European sovereign debt crisis and ‘the 2015–2016 Chinese stock market turbulence,’ which can provide early warning signals of the financial stress. The systemically important financial institutions change over time, but banks generally have a high interconnectedness.