Yu Jiuhong，Lu Yue，Wang Zhibo

(Center for Financial Engineering，Soochow University，Suzhou Jiangsu 215006)

Ⅰ．Introduction

1．Connotation and status quo of credit risk

Credit risk，also called default risk，refers to the possibility of financial loss being caused by that the counterparty cannot fulfill obligations according to the contract．Credit risk comes into being with the deal on credit，which is quite common in financial activities such as borrowing and lending，transaction in assets and so on．

The credit risk of the commercial banks mainly refers to the possible loss caused by the contract－ breaking behavior of the credit side．Broadly speaking，credit risk also includes the potential loss caused by the falling of the value of the creditor’s rights．In detail，the main risk of the various loans of the commercial banks，such as loan on credit，guaranteed loan，notes discounted and various bonds and derivatives is credit risk．When the debit side，the issuer of the bonds or the counterparty of the derivatives breaks the contract，the commercial bank will have to bear the loss．

Since the 20th century，the academic circle has been positively exploring the management of the credit risk，accumulating much valuable experience．Generally speaking，risk management can be divided into the recognition，the measurement and the control of the risk and the measurement of risk keeps being the research focus of the academic circle both at home and abroad．As for practice，many financial organizations tend to measure and analyze the credit risk of their own from both the qualitative and the quantitative perspectives．According to the development history of the management of the credit risk，the management of the credit risk can be divided into three periods:the traditional qualitative analysis，the model－ analysis based on financial index and the modern credit risk model analysis．

2．Development and status quo of the credit risk management in China

With the reform and opening up of China，the economy becomes more and more marketized，and the marketing environment confronted by the commercial banks becomes more and more complicated．As for the managers of the commercial banks，it is much more important to choose proper method to measure and manage the credit risk．At present，commercial banks in China mainly adopt the internal rating－based approach．

The five－class classification method is widely applied in the risk controlling process．The five － class classification method attaches importance to micro implementation，relying on the normative procedure，the subjective experience of the loan officer and the information offered by the financial index．The five－class classification method has its own shortcomings．First，the measurement of the credit risk is limited on the qualitative layer，lacking quantitative measurement．Second，it is difficult to find out the potential credit risk．Finally，the correlation of various credit events is ignored．

Up till now，the commercial banks in China lack the precise quantitative analysis and management of the credit risk．There are mainly two reasons．Firstly，there is no historical database related to credit assets for the state－owned commercial banks．Secondly，the credit rating agencies in China still need development，and there is no rating standard and rating method;the internal rating system of the commercial banks in China is also immature．

Such shortcomings reflect the laggard situation of the commercial banks in China;in the long run，it may well influence the development of the banking industry in China．As a result，we shall and must learn from the advanced experience of the commercial banks in the West．

3．Research significance and innovation of the this paper

This paper tends to research upon the quantitative method of the credit risk of the commercial banks and the application basis and prospect of such a method in China．Based on the VaR method and the Credit Metrics model，this paper analyzes the advantages of the quantitative method and explores the feasibility of the quantitative method and puts forward some advice for the measurement of the credit risk and the credit rating system．

Focusing on the hot issues of the management of the commercial banks in China，this paper analyzes the new situation of the banking industry and the new challenges of the credit risk，exploring effective method of solving practical problems so as to improve the current situation，providing more powerful and more permanent guarantee for the healthy development of the banking industry in China．

This paper innovatively analyzes the advanced credit risk management model—the Credit Metrics model and based on the exploration of the details of the model，this paper points out the application of the model in the credit risk management of the commercial banks in China．At the same time，this paper creatively puts forward how to improve the current situation of the management of the credit risk of the commercial banks in China．

Ⅱ．Credit risk model based on VaR—the Credit Metrics model

VaR method was first put forward by J．P．Morgan in April，1997．It was the measurement method for the credit risk based on credit rating，put forward for precisely measuring market risk at first，and then it was applied in various risk measurement and developed into the standard of risk management．The VaR method，with the pressure test and the scenario analysis，forms the VaR system．This model forms the framework simulating the potential changes of the credit assets and the default fluctuations．It adopts the market－to －market method to calculate the credit VaR．

This model can not only recognize the credit risk of the traditional investment tool，but also recognize the credit risk of the modern financial derivatives，having been applied in the credit risk management of the banks in the developed countries．Research upon Credit Metrics is also helpful for the risk management of the commercial banks in China．

1．Foundation of the Credit Metrics model

The credit risk is not related to the market risk．In the Credit Metrics model，the future value of the property depends on the forward interest rate curve，and the credit risk caused by the change of the value is not influenced by the fluctuation of the market．In the model，the one and only variable is the credit ratings，without considering the existence of the market risk．

The distribution of credit is discrete，and under the same credit rating，all the enterprises have the same credit．In the model，the counterparty is classified according to the credit rating of the rating agency，and the counterparties of the same rating have the same transfer matrix，forward interest rate and default rate．And the recovery rate is got from the average of the net recovery of the enterprise．Al-though such a hypothesis is not proper，it is difficult to quantize more precisely in this model．

2．Direct measurement of the Credit Metrics model for the credit risk of commercial banks

The Credit Metrics model is mostly based on the credit default probability，the matrix of credit transition probability and the forward interest rate，all of which are the main carriers of the credit risk input of the model．The credit risks of various loans are different，because the credit ratings of various enterprises are different．The lower the credit rating is，the higher the credit defaults rating is．The data of the credit default probability and the matrix of credit transition probability come from the statistics provided by the authoritative credit rating agencies．When the credit risk increases，the default probability and the transition probability increase accordingly．

The Credit Metrics model provides the approach and standard for the quantization of credit risk，but it has short comings，such as being limited by the credit rating system，the hysteresis of the credit rating and so on．The credit rating system is the basis for the model，so if the system was not perfect，the result would be influenced．Even if there is a perfect database and much reasonable historical data，the model has its own hysteresis．

Ⅲ．Application of the Credit Metrics model in the credit risk management of commercial banks in China

1．Significance

With the development of the modern assets portfolio theory and the innovation of the statistical techniques，the modern credit risk management is widely applied in the various financial organizations all over the world．According to the current situation and the management of the credit risk in China，it is significant to actively carry forward various modern credit risk management tools，among which the Credit Metrics model takes up the dominant position．

Firstly，the Credit Metrics model provides more advanced thought and method for the credit risk management in China．Secondly，the Credit Metrics model provides a large platform and objective standard for the credit risk management in China．The commercial banks in China mainly adopt qualitative analysis，so the model can provide a more objective standard for the commercial banks and other financial organizations in China．Thirdly，the Credit Metrics model is more practical and provides the direction of improvement．

2．Conditions

As for the application of the Credit Metrics in the system of commercial banks，the following conditions are necessary．Firstly，there must be the database of the credit status of enterprises．In the Credit Metrics model，the value distribution of the assets is based on the credit rating，so the credit rating is in-dispensible．Secondly，the rating system must be authoritative and credible．The credit risk mainly comes from the decrease of the value caused by the change of credit rating．As a result，the authoritativeness and credibility of the credit system must meet the need．Finally，the credit rating system must be updated in time．

Besides，the interest rate market must be active，or the various forward interest rates are not effective．If there is no free interest rate mechanism，there would be no data of the forward interest rate．And there must be a sound，reasonable and effective institutional platform，which is the framework and governance mechanism of the model．Compared with the technical factors，the institutional factors are the essential，conditional and comprehensive．

3．Policy suggestions

Firstly，the interest rate market is the necessity．Through reform the liberalization of the interest rate is promoted．Secondly，the rating system must be perfected．Under the planning and arranging of the Central Bank，the database of the credit of enterprises must be established，and the internal rating system of the banking industry must be improved．Last but not least，the institutional supervision must be strengthened and the staff of the commercial banks must have the risk awareness．

〔1〕江成敏:《VaR主要計(jì)算方法介紹及實(shí)證分析》，《時(shí)代金融》2013年第21期。

〔2〕王舒:《中國(guó)對(duì)外貿(mào)易結(jié)構(gòu)與產(chǎn)業(yè)結(jié)構(gòu)的關(guān)系研究——基于VAR模型》，《當(dāng)代經(jīng)濟(jì)》2013年第24期。

〔3〕李偉杰:《淺談歷史模擬法VaR的設(shè)計(jì)和實(shí)現(xiàn)》，《時(shí)代金融》2009年第7期。

〔4〕黃潔卉:《我國(guó)商業(yè)銀行信用風(fēng)險(xiǎn)評(píng)估及管理》，《浙江大學(xué)》，2004年。

〔5〕艾丁:《XX商業(yè)銀行個(gè)人信貸業(yè)務(wù)風(fēng)險(xiǎn)及其防范研究》，《首都經(jīng)濟(jì)貿(mào)易大學(xué)》，2013年。

〔6〕龐增錄:《商業(yè)銀行信用風(fēng)險(xiǎn)研究綜述》，《中國(guó)證券期貨》2012年第4期。

〔7〕霍聰:《CreditMtrics模型在信用評(píng)級(jí)中的應(yīng)用》，《商場(chǎng)現(xiàn)代化》2012年第20期。

〔8〕姜新旺、黃勁松:《CreditMetrics模型及其對(duì)我國(guó)商業(yè)銀行的適用性》，《統(tǒng)計(jì)與決策》2005年第17期。

〔9〕申靖:《關(guān)于風(fēng)險(xiǎn)管理中VaR方法的文獻(xiàn)綜述》，《中國(guó)集體經(jīng)濟(jì)》2011年第24期。

〔10〕龍克維:《VaR方法及其在金融風(fēng)險(xiǎn)管理中的應(yīng)用》，《時(shí)代金融》2011年第3期。