A Study of the Credit Risk Evaluation Model and Loan Pricing Strategy on Personal Loan-an example of a domestic Bank

• Main Authors: Chi-Hung Chien, 簡啟鴻
• Other Authors: Chih-Peng Chu
• Format: Others
• Language: zh-TW
• Published: 2006
• Online Access: http://ndltd.ncl.edu.tw/handle/29717955498816050079

碩士 === 國立東華大學 === 高階經營管理碩士在職專班 === 95 === Abstract Due to the deregulation of the new bank licenses, the competition of financial banks is getting fierce. Conseguently, the profit margin of saving and loans is shrinking. Besides, due to the overflow of new competitors, the fierce competition has lead to bad quality of loans and to increasing debt rate. In the fierce competition of the banking business, the banks have to increase the amount of loans and increase the period of loans, and reduce verification time. Therefore, it is necessary to use automatic examination system to reduce the risks of having bad loans. Through the case study of a selected bank, this research discuss the factors which influence the risks of personal credit loans, so that we can evaluate the level of the risk of personal loan fast and objectively and take the evaluation as the base of giving loans. Furthermore, we develop a pricing model based on the risk segmentation so as to enhance the profit of the bank. This research adopted the model of Logistic Regression, discriminate analysis model and the linear programming model. The test results indicates that the factors which lead to different level of risks are academy background, job position, average annual income, the loans of other banks, and the banks checked within 3 months. The functions of the Logistic Regression Model are as follows. Z＝-0.4464＋0.1908X3+0.0646X7－0.0013X9＋0.1549X10＋0.0400X17 The forecast accuracy may reach 73.3%. This research also uses the cluster analysis to divide the customers into 3 risk groups by their possibilities of breaking the contract. The result shows that the customers are significantly divided into 3 types, including "low credit risk group", "gray credit risk group" and "high credit risk group". Further we used the linear programming model to develop the pricing strategy and asset allocation strategy to maximize bank profit. The result indicates that in 3 different risk groups, all the models suggest we use the highest interest rate under Complete Competition. Also in the asset allocation, it is suggested to lower the amount of loans. This result reflects the real direction of current banking industry, which is to ask higher profit margin and higher quality of loans instead of focusing on the amount of loans.