The Study On Credit Risk Macro Stress Test Of Commercial Banks Based On Industry GVAR Model

Macro-prudentia l Regulation reflects the future development trend of financ ia lrisk mana geme nt, and macro stress test is an important tool to achieve it. Carrying outmacro stress test of the banking credit risk based on industries, on the one hand, needsto take into account the direct influence macro economic shocks on the industry creditrisk of banking system; on the other hand, owing to the industry correlatio n willgenerate risk contagion between the industries, requires to cons ider the ind irectinfluence macro econo mic shocks on the industry credit risk.In order to prevent systemic risk effectively, to improve the stability of thefina ncia l syste m, and to better imple ment the macro-prudential regulation, we shouldcarry out systematic study on the industry credit risk change tr ends and the overfloweffects caused by contagion. The research is based on this background.From theoretica l research to empirical research, this paper elaborates the basicidea of ind ustry GVAR model, and calculates grey correlation matrix betweenindustries through grey relationa l analysis method. Next, we bring the industrycharacteristics, industry re levance and macro economic factor into macro stress testmodel of credit risk, which establishes the conduction model based industry GVARpattern. On the basis, the industry macro stress test is carried out, and we analys is theimpact of industry relevance and the macro economic factors on industry defaultprobability·s dynamic change. Last, we also describe the factor contribution ofindustry default probability change with variance analysis.The conc lus ion as the fo llowing:(1) The grey correlation matrix shows that thereis a close relationship between industries. Industry credit risk·s spillover effect ispositive and significant, and industry influe nce elastic is varied to differentindustry.(2)The GDP growth rate has a negative effect to industry credit risk, whilethe benchmark one-year loan interest rate and broad money supply growth rate toindustry credit risk effect is positive. Otherwise, the stoc k price index·s influence isnot obvious.(3) When the macro economic factor shocks, agriculture industry,build ing industry and informatio n industry has a bigger increase about expected loss,meanwhile, these industries generates higer influence on the expected loss of overallcredit asset.(4) The impact factor contribution ana lys is shows that: industrycorrelation factor has a higher contribution, which the average contribution is8%or so. The benchmark one-year loan interest rate and broad mone y supply growth rate hasa bigger contribution than the GDP growth rate and stock price index.Based on the above the empirical results, some countermeasures and suggestionsare putted forward: conducting the industry credit risk·s macro economic stress onaccount of the ind ustry characteristics; to optimize the industry structure of creditassets; establis h the dyna mic risk early warning s ystem in view of systemic fina ncia lrisk prevention, etc.