The Subgraph Analysis of Chinese Credit System

Recently, network subgraphs are widely used to capture basic interaction patterns and functional properties of complex networks. In our paper, we do a detailed sub graph analysis of real-world heterogeneous Chinese credit system. Based on the five years' real-world Chinese guarantee network data, we construct the dynamic Chinese guarantee network. Firstly, we identify the motifs in 2- and 3-node subgraphs, and find that firms have special preferences to form mutual and triangle guarantee relationships. Specifically, we consider the heterogeneity of firms and determine that firms with relatively large assets tend to form the sub-pattern of 2-out-stars and that small firms tend to form mutual guarantee relationships. Additionally, we characterize the functionally important subgraphs with high default risk or contagion ability. In particular, the empirical results provide evidence that (a) cycle and fully connected subgraphs of 3 -node demonstrate low default risk, but high contagion capability; (b) single-link and 2-star patterns are perfect choices to reduce or control the contagion probability. Our results give the policy makers an in-depth understanding of Chinese guarantee firms' preferential behavior from the perspective of network subgraphs, and provide structural design instructions for future policy making to reduce the potential default and contagion risk of guarantee network.