159-2008: Identifying Potential Default Loan Applicants: A Case Study of Consumer Credit Decision for Chinese Commercial Bank
نویسندگان
چکیده
Consumer credit is a lucrative but risky business. In order to control risk and maximize profits, commercial banks around the world have made great efforts to develop various analytic models to identify potential default loan applicants. This is also critically important to China, as the non-performing individual loans of Chinese commercial banks have been fast growing because of the myopic business attitude and the overheated economic growth. To prevent the US sub-prime kind of crisis in China, Chinese commercial banks are adopting advanced analytic means to help make loan decisions. This paper reports a data mining application in the analysis of default loan applicants using a real dataset consisting of 641,988 observations obtained from a Chinese commercial bank, located in the southwest of China. An exploratory study of the dataset led to a number of interesting statistic figures that may characterize the applicants in the western region of China. In the analytic study, we constructed two types of models, unweighted and weighted, with SAS® Enterprise Miner. The models have revealed a number of useful findings that meet our expectation. They demonstrated good predictive power for loan decision making. The success of the study has consoled our concern in the data quality that is relevant to the efficiency of the data collection system rooted in the existing Chinese business system. INTRODUCTION Over the past decade, consumer credit business of commercial banks has been expanding rapidly in China. In 1998, the balance of consumer loans of commercial banks in China was about US$ 10 billion. By the end of 2005, the balance reached US$ 293 billion (China State Statistics Bureau, 2006) – increasing more than 29 times in seven years. On the other hand, the fierce competition of consumer credit in China seems to harbor potential risks for commercial banks. The real estate prices in China have been skyrocketing over the years and bubbles are forming. Traditionally, consumer credit in China seems to have less risk, especially mortgage loans—because it just postpones risks not eliminates them and Chinese commercial banks began to provide mortgage loan about ten years ago and the maturity of most loans is more than fifteen years. As the real estate market is booming, the loans generate profits to banks and the risk of customer default is less, the business seems promising. If the prices of real estates fall down, the risk will become significant and the default on loans would incur great loss – this is very possible as the warning of the overheated Chinese economy is getting louder. However, the myopic attitude and the pressure of competition are driving many Chinese commercial banks for more customers without carefully screening. As getting a mortgage loan from a Chinese commercial bank is relatively easy, the non-performing consumer loans are increasing with real estate market fluctuation of recent years in China (http://www.buyusa.gov/china/en/bank. html). According to Shanghai Banking Supervision Bureau’s statistics, the non-performing individual loans of Chinese commercial banks in Shanghai in the third season of 2006 totaled RMB 3.275 billion, that is, RMB 0.4 billion more than the last season and RMB 0.842 billion more than the beginning of the year. The non-performing rate of individual loans is 1.14%, an increase of about 10% than last season and about 20 % than the beginning of the year respectively. In particular, the non-performing consumer loans accounted for 90% of those non-performing individual loans (http://old.go24k.com/bank/news.asp?id=41470&anclassid=6). With recent US sub-prime mortgage crisis, the consumer credit decision of Chinese commercial banks becomes critical. The US sub-prime mortgages are those cases that banks gave cash to people with poor or no credit history 1 Upon the request from the bank, the bank’s name is not mentioned here. The findings and conclusions reported in this paper are solely the opinions of the author(s) and do not necessarily represent the viewpoints of their institutions. 2 In China, consumer credit includes both mortgage loans and others individual consumption loans, however, about 90% of consumer credit is mortgage. 1 Financial Services SAS Global Forum 2008
منابع مشابه
Investigating the missing data effect on credit scoring rule based models: The case of an Iranian bank
Credit risk management is a process in which banks estimate probability of default (PD) for each loan applicant. Data sets of previous loan applicants are built by gathering their data, and these internal data sets are usually completed using external credit bureau’s data and finally used for estimating PD in banks. There is also a continuous interest for bank to use rule based classifiers to b...
متن کاملMatrix Sequential Hybrid Credit Scorecard Based on Logistic Regression and Clustering
The Basel II Accord pointed out benefits of credit risk management through internal models to estimate Probability of Default (PD). Banks use default predictions to estimate the loan applicants’ PD. However, in practice, PD is not useful and banks applied credit scorecards for their decision making process. Also the competitive pressures in lending industry forced banks to use profit scorecards...
متن کاملThe Comparison of Credit Risk between Artificial Neural Network and Logistic Regression Models in Tose-Taavon Bank in Guilan
One of the most important issues always facing banks and financial institutes is the issue of credit risk or the possibility of failure in the fulfillment of obligations by applicants who are receiving credit facilities. The considerable number of banks’ delayed loan payments all around the world shows the importance of this issue and the necessary consideration of this topic. Accordingly...
متن کاملUsing the Hybrid Model for Credit Scoring (Case Study: Credit Clients of microloans, Bank Refah-Kargeran of Zanjan, Iran)
In any country, commercial banks lay the groundwork for economic growth by collecting national resources and capitals and allocating them to different economic sectors. Optimal allocation of resources is especially important in achieving this goal. Banks with an effective and dynamic system of customer assessment can efficiently allocate their resources to customers regardless of their geograph...
متن کاملExtracting Rules from Imbalanced Data: The Case of Credit Scoring
Credit scoring is an important topic, and banks collect different data from their loan applicant to make an appropriate and correct decision. Rule bases are of more attention in credit decision making because of their ability to explicitly distinguish between good and bad applicants. The credit scoring datasets are usually imbalanced. This is mainly because the number of good applicants in a po...
متن کامل