管理工程学报2012，Vol.26Issue(4)：71-76,99,7.
不同网络结构下银行间传染风险研究
Investigating Interbank Contagion Risks under Different Network Structures
摘要
Abstract
The interbank market is a significant part of the modern financial system, and an essential financial intermediary for the healthy and stable functioning of the whole economic system. With the development of the banking industry, the complicated interbank debtor-creditor relationship contributes to strong correlations among banks. Banking risks may spread through bank credit and information. Credit risks with a high degree of correlation are likely to cause a "domino" effect. Because of this effect, banks are facing market risk, operational risk, credit risk as well as contagion risk. Banking crises in real life show that contagion risk is often the key factor for determining the depth and breadth of a crisis. Contagion risk has received increased attention, and become an important topic on financial risks. Based on the Complex Network Theory, this paper investigates the characteristics of the interbank contagion risk and their difference in random network, small-world network and scale-free network of the interbank market. First, interbank credit lending relationships are given a network representation with reference to Complex Network Theory, where bank denotes network nodes and interbank credit lending relationships represent the edges between nodes. Based on empirical results of studying interbank market networks, this paper uses random network, small-world network and scale-free network to describe the structure of the credit lending relationships in the interbank market, where the determination of the edge direction in the networks is exogenous. Second, this paper describes a bank's balance sheet. The balance sheet of banks constructed by this paper consists of investments, liquidity assets and interbank loans as assets and interbank borrowing, deposits and equity as liabilities. Based on the balance sheet, the formation process for contagion risk is constructed, contagion risk is formatted through the interbank lending channel, and the effect of contagion risk can be measured. Finally, the paper runs simulation analysis for contagion risk charaeteristics under random shocks and selective shocks, where banks are homogeneous and heterogeneous. Simulation results show that when the number of shocked banks is small in the random network the effect of random shocks and selective shocks is basically identical. However, with the increase of the number of shocked banks the effect of selective shocks is larger than that of random shocks. In the small-world network, the effect of random shocks is larger than that of selective shocks. When the number of shocked banks is small in the scale-free network, the effect of random shocks is larger than that of selective shocks. After increasing the number of shocked banks up to certain levels, random shocks and selective shocks have identical effect. In the three network structure, the effect of contagion risk is the largest in the small-world network but smallest in the scale-free network. In summary, from the perspective of complex networks this paper studies the effect of contagion risk under different networks. The simulation analysis shows that contagion risk effect of random shocks and selective shocks is related to the interbank network structure and the number of shocked banks. In three kinds of network structure, the scale-free network of the interbank market has the largest stability. These findings can help understand how to promote the risk-withstanding ability of the interbank market, maintain the stability of the interbank market, and ensure the streamlined operation of the banking system.关键词
传染风险/复杂网络/选择性冲击/随机性冲击Key words
contagion risk/ complex network/ selective shocks/ random shocks分类
管理科学引用本文复制引用
李守伟,何建敏..不同网络结构下银行间传染风险研究[J].管理工程学报,2012,26(4):71-76,99,7.基金项目
国家自然科学基金项目(70671025 （）
71071034) （）
