银行业同业业务网络与穿透式监管研究

The Network of Interbank Business and Look-through Supervisions

随着我国银行业金融机构同业业务的高速发展,同业业务网络的交易关系日益复杂。基于此,部分金融机构利用层层嵌套的方式规避监管,金融"脱实向虚"的现象日益突出。本文结合监管实践,针对目前同业业务网络演化难以预估、膨胀规模难以度量等问题,提出了基于复杂网络理论的演化模型。运用社会型网络的同配流策略,宏观上可反映同业业务的系统性及趋势性风险,微观上通过节点研判可监测单点风险。通过对同业业务网络的拓扑演化进行仿真模拟,本文发现:在外部监管条件不变的情况下,风险将随着同业业务网络节点数的增加而放大;而同业交易对手增加时,将使同业业务网络更加复杂,从而对监管提出更高要求。利用复杂同业业务网络模型可迅速发现同业业务网络中的关键节点、量化网络资金流规模,为监管部门及时抓住要害、理清头绪,以及实施穿透式监管提供依据。

With the rapid development of interbank business in China＇s banking industry, the transaction relationships based on interbank network have become increasingly complicated. Institutions using multiple networks to evade supervision, thus the phenomenon of the funds shifting from real economy to virtual economy is becoming more and more prominent. According to regulatory practices, this paper proposes an model based on complex network theory, to solve the problem of unpredictable evolution in interbank network scale. Using the strategy of social network, this model could capture the systematic risk and the trend of interbank business, and monitor the single-point risk. This paper uses Matlab to simulate the topology evolution of the interbank business network. The experiment shows that the risk will increase by more peer nodes with the same supervision conditions. Relaxing the control of counter party witl cause a higher speed in interbank business market and require much more effective regulatory. This model could discover key nodes and quantify the flow scale of network, which is helpful that regulatory authorities could implement the penetrating supervision of financial institutions.