|dc.description.abstract||Credit scoring is broadly applied in consumer lending especially in credit cards and mortgages. Credit scoring has not been widely used in business lending because business loan differ substantially across household borrowers, making it more difficult to build up an accurate scoring method. However, this has changed. The complexity and flexibility of statistical models and advanced computing technology have made such credit scoring possible in business lending. Thus, many banks are using credit scoring model to evaluate business loan applications, which is a cost effective credit management tool (Mester, 1997).
Preceding credit scoring valuation is important against the increasing financial risk, and adopting credit risk analysis method to value the consumer credit is a necessary step to resist serious credit loss (Li and Zhang, 2003). It is important for banks to understand that every borrower is a potential lemon and the probability to default is high. This is evidenced in today’s U.S. subprime loan problems. In essence, a credit scoring model provides an objective estimate of a borrower’s credit risk. Since a lender can not observe the borrower’s probability to default, credit scoring models enable lending institutions to rank potential customers according to their default risk, and improve the allocation of loan resources.
Similarly, as China becomes more consumer credit culture and people start to borrow money for homes and other investments, an international risk standard management framework becomes more important for Chinese domestic banks. The efficiency advantage of credit scoring models certainly will help the banks to meet the growth in Chinese consumer loans and mitigate default risk.
The primary problem of any lender is to differentiate between "good" and "bad" debtors prior to granting credit. Such differentiation is possible by using a credit-scoring method. In this study, we examine how this inference problem impacts mortgage borrowers’ characteristics on the probability of loan default. This includes volume of loan granted and loan price which is the interest charged on loans. We use a credit scoring model to investigate the bank mortgage lending policy in China during the period 2004 to 2009.
The results show Age group, Education level, Occupation type, and Region positively impact the amount of credit granted. Gender, Age group, Marital status, Annual income, Bank rating, Occupation type, Loan duration and Region have negative impacted on loan pricing. The findings also suggest that it is necessary to review the borrowers’ creditworthiness periodically, as the changes in economic condition could affect the loan performance. The results are generally consistent with the existing literature.||en