Community Structure and Systemic Risk of Bank Correlation Networks Based on the U.S. Financial Crisis in 2008

Abstract

This paper studies the community structure of the bank correlation network in the financial system and analyzes the systemic risk of the community sub-networks. Based on the balance sheet data of U.S. commercial banks from 2008, we establish a bank correlation network for each state according to the banks’ investment portfolio ratio. First, we analyze the community structure of each bank’s correlation network and verify the effectiveness of the community division from the point of view of the importance of nodes. Then, combining the data of failed banks after the 2008 financial crisis, we find that for small communities, the financial systemic risk will appear to have obvious volatility, and it is quite likely to reach an extremely high level. With the increase in the number of nodes in the community, systemic risk will tend towards a stable and low level. Furthermore, if only communities with failed banks are considered, the regression analysis shows that systemic risk and the size of the community almost follow a power law distribution trend. These results reveal the importance of supervising the banking system at the level of community sub-networks, which has certain guiding significance for the stability of the financial system.