A Network Approach to the Study of the Dynamics of Risk Spillover in China’s Bond Market

Since 2018, the bond market has surpassed the stock market, becoming the biggest investment area in China's security market, and the systemic risks of China's bond market are of non-negligible importance. Based on daily interest rate data of representative bond categories, this study conducted a dynamic analysis based on generalized vector autoregressive volatility spillover variance decomposition, constructed a complex network, and adopted the minimum spanning tree method to clarify and analyze the risk propagation path between different bond types. It is found that the importance of each bond type is positively correlated with liquidity, transaction volume, and credit rating, and the inter-bank market is the most important market in the entire bond market, while interest rate bonds, bank bonds and urban investment bonds are important varieties with great systemic importance. In addition, the long-term trend of the dynamic spillover index of China's bond market falls in line with the pace of the interest rate adjustments. To hold the bottom line of preventing financial systemic risks of China's bond market, standard management, strict supervision, and timely regulation of the bond markets are required, and the structural entropy, as a useful indicator, also should be used in the risk management and monitoring.