How to reduce the default contagion risk of intercorporate credit guarantee networks? Evidence from China

This study uses data from 2011 to 2018 for Chinese small and medium-sized enterprises to construct a weighted directed network to investigate the topology of intercorporate credit guarantee networks. Moreover, based on the DebtRank algorithm, it develops a novel GuaranteeRank model that includes three factors to comprehensively examine default risk contagion and systemic risk in various scenarios. The results demonstrate that (1) credit guarantee network has the topological characteristics of "scale-free" and "small world"; (2) default contagion and systemic risk increase significantly when the macro-external shock and company's off-balance-sheet debt exceed certain threshold values, while continuous bank credit support can notably reduce the risk; (3) credit guarantee network is "robust yet fragile", such that targeted shocks increase systemic risk much more than do random shocks; (4) in addition to the prevalent "too big to fail" and "too central to fail" phenomena, a "too connected with the central to fail" phenomenon is also identified for the first time. Therefore, this study provides an important reference for regulators and financial institutions to reduce the default contagion risk of intercorporate credit guarantee networks.