Research On Measurement And Prediction Of Banking Systemic Risk In China

The global financial crisis triggered by the 2007 US subprime crisis smashed the bubble of the capital market sharply decreased the amount of credit,resulting in insufficient market liquidity.The global financial crisis has aroused the attention of governments around the world on systemic risk.The systemic risk is overall,comprehensive and contagious,rapidly spreading to the world during the financial crisis.The systemic risk can lead to instability in the financial system,hinder economic growth and damage the social welfare of financial institutions with wide range of hazards.It has enormous negative externalities effect on both the global financial system and the real economy.In recent years,the international and domestic environment for China's economy and financial market are complicated.The development of China's economy presents short-and long-term,internal and external,cyclical and structural contradictions and problems.These challenges may affect the stability of China's financial system.To avoid the risk of a recession in China's economy in the medium and long term,and be wary of the new round of financial crisis,Chinese government has held many economic and financial work conferences in recent years,and repeatedly emphasized the importance of preventing and controlling the systemic risk.The report of the 19 th National Congress of the Communist Party of China also clearly stated that to prevent and resolve systemic financial risk will become the first of three major battles in China.It can be seen that to prevent and resolve the systemic risk has become the focus of China's current economic work.For a long time,China's banking industry has always dominated the financial system and played a pivotal role in the development and stability of China's economy and finance.To this end,this paper starting from China's specific economic and financial market,focuses on China's banking industry,and uses theoretical analysis and empirical analysis,as well as qualitative analysis and quantitative analysis,to measure banking systemic risk,study the influencing factors of banking systemic risks,and prediction of the systemic risk.The research in this paper mainly includes the following contents:Firstly,the paper introduces higher moments including skewness and kurtosis based on the generalized Edgeworth series expansion of the lognormal density function to extend the CCA model.Based on the extended CCA model,this paper measures the systemic risk of Chinese listed commercial banks.The research result shows that the CCA model with higher moments is highly sensitive and is able to accurately measure the systemic risk of the banking industry.Secondly,the paper uses ?CoVaR and MES to measure the contribution of China's isted commercial banks in banking systemic risk.The China's listed commercial banks are used as samples to measure their systemic risk contribution based on the stock market data.The results show that first of all,both ?CoVaR and MES are more accurate than VaR in measuring systemic risk contribution.The result of MES is close to that of ?CoVaR except for Bank of China,Ping An Bank,Shanghai Pudong Development Bank,China Merchants Bank,Bank of Nanjing,and Bank of Ningbo.Second,the systemic risk contribution of state-owned large commercial banks is higher than that of both joint-stock commercial banks and city commercial banks.Thirdly,the paper establishes the quantitative regression data model to study the influencing factors of banking systemic risk in China.First,the papaer establishes fixed-effect regression data model to study the impact of bank micro-characteristic variables including bank size,leverage,total return on assets,asset quality,and stock volatility on banking systemic risk.The empirical results show that the expansion of bank scale will significantly reduce the banking systemic risk,while the increase of bank's total return on assets,non-performing loan ratio and stock volatility will significantly increase the banking systemic risk.Although the increase in leverage will increase the banking systemic risk,the impact is not significant.Second,the paper uses the dynamic panel system GMM data model to study the relationship between off-balance sheet activity and banking systemic risk.The empirical results show that,overall,the off-balance sheet activity is negatively correlated with the banking systemic risk,that is,the increase of off-balance sheet activity by commercial banks can significantly reduce the banking systemic risk.Through different banks group regression,the increase of off-balance-sheet activity by state-owned commercial banks can significantly reduce its contribution to the banking systemic risk,while the increase of off-balance-sheet activity by joint-stock commercial banks or city commercial banks will increase their contribution to the banking systemic risk which will hedge some part of the contribution of state-owned commercial banks to the banking systemic risk to a certain extent,but the impact is not significant.Therefore,it is consistent with the empirical result when all banks are in empirical regression.Fourthly,the paper uses computer simulation and comparative analysis,mainly applying machine learning algorithms,to conduct the research on the prediction of China's systemic risk.The first part is based on the support vector machines(SVM)to build a prediction model of banking systemic risk in China,and compares the results with those of BP neural network model and Logit regression.The second part is based on least squares support vector machines(LSSVM)to build a prediction model of China's financial systemic risk.Compared with those of SVM model,BP neural network model and logit regression,the research result of the LSSVM model has a higher classification accuracy rate to predict systemic risk.Finally,this paper proposes a “double pillar” regulatory framework including both the monetary policy and the macroprudential policy to prevent and resolve systemic financial risk and maintain the stability of the economic and financial system guided by the national policy.In addition,this paper proposes to policies including to improve institutional measures,strengthen the ability of banks to prevent and control systemic risks,establish scientific and effective systemic risk prediction mechanisms,and prevent external shock risks.