Study On Interbank Network Evolution And Systemic Risk

The interbank market, providing convenience for banks to adjust liquidity, also provides a channel for the spread of crisis contagion. When one or a few banks default, through the interbank lending relationship, it will lead to other banks default and trigger a systemic risk. This paper introduces the risk aversion adjustment parameter, adjusting interbank network evolution process, to determine the structure of interbank network. Then we apply network model to study the relationship between the network structure and systemic risk.This paper takes the interbank market as the research object, to explore the formation and evolution of the interbank network. From the point of banking history development and behavior response decision, we give the formation foundation and change logic analysis of interbank network. That suggest, in the long process of historical development, the number of banking industry keep growing, and the new banks tend to seek safer banks to develop business relations. In view of this, the formation of interbank network obeys the principles of the growth and preferential loans. So we construct a network with growth and preferential attachment-scale-free network, as the benchmark network. Interbank business is not immutable and frozen. Interbank loans due to maturity need find another counterparty. Banks’choices of new and old counterparties abide by the principle of survival of the fittest. While banks seeking counterparties depends on its degree of risk aversion. This paper introduces a risk aversion adjustment parameter to indicate the degree of risk aversion. According to the principle of survival of the fittest, The priority reconnection and cut off of benchmark network, according to the different risk aversion adjustment parameter, generated3kinds of network:risk neutral, risk preference and risk aversion network. We found that the banking growth characteristics, preferential lending principles constitute the basis for the formation of the interbank network, while the trading principle of survival of the fittest determines the changes logic of interbank network structure. Finally, by comparison we find that network concentration keeps positive correlation with the degree of risk aversion.This paper also establishes a default contagion research framework. First of all, we need construct debt matrix of interbank network. According to existing research on balance sheet and node degree, we can get total interbank assets and liabilities of every bank, then using the maximum entropy method to estimate debt matrix of interbank network. Secondly, we give a comparative analysis for the consequences of risk contagion. Under stochastic shocks, the number of contagious default bank keeps positive correlation with network concentration, which is related to the degree of risk aversion. However there exists one threshold for shock scale, contagious default bank assets ratio rises with network concentration, after the threshold, moves in the opposite direction. Under the selective shock of order, that is from big to small, the higher the concentration, the more serious the contagion. We find that the security of single bank transactions may not be able to increase the stability of the system. Under the selective shock of reverse order, that is from small to big, when the bankruptcy of small banks accumulated to a certain amount, it can cause large bank contagious default, then trigger the collapse of the system, and the contagion once occurred, the consequence of risk contagion under reverse order is the most serious. So the counterparties of small banks excessively concentrate on some big bank or a few big banks, that also can induce systemic crisis. Finally, this paper uses Monte Carlo simulation method to compute respectively the average loss and the maximum possible loss of interbank network. Simulation of each bank assets are subject to different degrees of shocks, we find that when the bank asset volatility is more than0.05, the probability of contagion occurrence will increase dramatically, and the number of basic default and contagion default both rises sharply. The average loss of interbank network and network concentration has no obvious correlation. While the bank asset volatility is less than0.05, the maximum possible loss of interbank network and network concentration keep negative correlation. However, once contagion occurs, with the bank asset volatility rising, the maximum possible loss of interbank network and network concentration ie. the degree of risk aversion, have a positive correlation. Because contagion occurs often accompanying big banks failing, so two measure of systemic risk has micro difference.Therefore, to strengthen the traditional regulations of individual financial institution remains essential. Controlling the banking assets’ quality effectively can enhance the stability of the banking system. At the same time, strengthening the monitor of the interbank network structure, can supply an early warning for the systemic risk.