The Research On Identifying Chinese Systemically Important Banks

Systemically important banks are banks that are deemed systemically importantto the economy in the sense that the failure of one of them could trigger a financialcrisis because of their size, complexity, interconnectedness and the lack of readilyavailable substitutes of financial institution infrastructure. As a new concept, there isno consensual methodology of how to assess a systemically important bank is. Thesystemically important banks assume a key function in the financial markets, andscientific assessment methodology is critical to enhance the robustness of our bankingsystem and protect financial and national economic systems.At present, regulators in our country have adopted size indictors to judgesystemically important banks which are large banks requiring higher capital adequacyratio. This assessment method is easy to use, but its drawback is also very obvious.Firstly, size indictors are not equal to systemic importance indictors. Secondly, thismethod cannot dynamically judge systemic importance of banks.In view of drawbacks referenced, this thesis adopts indictors reflecting the size,interconnectedness and substitutes of banks, and builds WSII model which base onmultivariate extreme value theory to identify the systemically important of China’slisted banks and explore an proper assessment methodology.The research shows that state-owned commercial banks have higher systemicimportance. If set5%as the threshold of the systemically important banks, allstate-owned commercial banks, Shanghai Pudong Development Bank, ChinaMerchants Bank and Industrial Bank are systemically important banks. The study alsofinds that there is a certain difference between systemic importance rankings andassets ranking of banks. Therefore, despite systemic importance indictors are relatedto size indictors, size indictors can’t completely replace the systemic importance ones.To solve the problems of identifying systemic important banks in China and to liftthe restrictions faced by the existing identification methods, this thesis makes policyrecommendations from three aspects: developing an all-in-one assessmentmethodology, establishing a scientific and prudent dynamic judgment framework andimproving the environment of quantitative assessment methodology.