CREDIT RISK MANAGEMENT OF THE CHINESE BANKS BASED ON THE KMV MODEL

Abstract: Due to the increasing need for advanced credit risk management and lacking quantitative credit risk measurement modeling at the Chinese banks, the purpose of this dissertation is to study the feasibility of applying the Western credit risk model, the Moody’s KMV (MKMV), to China. Because of the particular Chinese considerations, such as developing stage of financial market and lacking of default data, a modified MKMV model is suggested and tested. Firstly, the samples of Chinese public listed companies are classified into two groups: special-treated with high default risk and non-special treated with low risk. Then, the adjustments of various parameters are determined. After that, the findings’ validation test is carried out. Finally, the results are discussed from three approaches of analysis: horizontal, vertical and regression. In accordance with the results of regression analysis, the theoretical expected relationships among the model parameters are significantly found in the Chinese samples. From both the horizontal and vertical analysis, the positive signs showing that the model’s abilities in discriminating the good companies from bad ones and in predicting the default risk of the distress companies in China are found. However, it is hard to infer the practicability of the MKMV model in China as a consequence of the statistical limitations. Further studies on the particular Chinese factors; including (i) the pricing non-tradable ordinary share, (ii) the unclear definition of default, and (iii) the absent of extensive historical default database; are suggested.