CIIS system of Industrial and Commercial Bank of China
Pay special attention to the customer information system (hereinafter referred to as CIIS system)
Currently, Industrial and Commercial Bank of China has incorporated the CIIS system into the credit approval control process. Bank regulations: Banks at all levels handle credit business for individuals (including self-operated personal housing loans, entrusted personal housing loans, personal comprehensive consumption loans, personal short-term credit loans, personal automobile consumption loans, national student loans, general commercial When providing student loans, credit cards, providing guarantees for other legal entities and individuals, etc.), the special attention customer system must be queried. No new personal credit business shall be conducted with the individuals listed in the special attention customer system.
The CIIS system breaks the business-centered data integration and establishes a customer-centered information system. A man walked into a branch of the Industrial and Commercial Bank of China to apply for a loan. After checking the system, the staff said to him: "I'm sorry, sir, you still owe money.
We can't give you a loan." It turned out that this man's ICBC credit card was overdrafted. It has not been repaid, so it has entered the "blacklist" system of ICBC.
According to ICBC's credit procedures, banks at all levels must first check this "blacklist" system when handling all personal credit businesses. Anyone who enters this system's list will, in principle, no longer be able to obtain loans. This system is called the "Customer Information System of Special Interest" ("CIIS System" for short) at ICBC. It automatically collects all bad credit customer records within the bank from ICBC's production system, and also collects records from other banks or institutions. bad customer information. By querying this system, staff can quickly eliminate high-risk loan businesses.
"Blacklist" entry
As we all know, loan interest rate spreads are still the main source of income for state-owned commercial banks. Bad loans mean that the loans issued cannot be recovered. If banks want to maintain capital balance, they can only reduce the number of loans issued in the next period, which reduces the bank's profits. As non-performing loans accumulate, banks' profit margins become smaller and smaller, and they may eventually lose the ability to pay interest to depositors.
The high non-performing loan ratio is still a stubborn problem for state-owned commercial banks. Experts point out: "If a bank's non-performing loan ratio is 20%, then its new non-performing loan ratio must not exceed 0.94 %, otherwise it will be in trouble in just 5 years.” According to ICBC’s 2004 annual report, ICBC’s non-performing loan ratio has dropped to 18.99%. Jiang Jianqing, President of the Industrial and Commercial Bank of China, proposed to reduce the non-performing loan ratio of the Industrial and Commercial Bank of China to 10% by 2006. This is also a prerequisite for ICBC to complete its listing. In addition to actively completing the disposal of historical non-performing assets, reducing the rate of new non-performing loans is an inevitable choice for the sustainable development of not only ICBC, but also all commercial banks.
In the past, due to the inability to fully share information, cases of bad customers from one region or department of ICBC going to another region or department for refinancing frequently occurred. With limited real information, it is difficult to make a correct evaluation of invisible and intangible customer credit, which makes it impossible to avoid the formation of non-performing loans. Now ICBC collects, records, and queries the information of these customers through the CIIS system to control risks from the source of the loan, while also eliminating the need for a large amount of data collection and review work in pre-loan investigations.
It is understood that the data sources of the CIIS system include internal sources and external sources. The internal source data is mainly the customer data of various professional departments and all branches of ICBC. These data are obtained from the CIIS system. Collected by the production system, when customer information reaches the warehousing standards set by the CIIS system, the data will be automatically included in the system; external data includes blacklists of 19 commercial banks, public utilities, and telecommunications departments, and even the United Nations, the United States and the list notified by our government.
“Since the CIIS system was put into use in September 2004 until now, we have rejected 2,100 non-performing loans, involving an amount of 285 million yuan, and recovered debts of 12 million yuan.” Data from the Industrial and Commercial Bank of China Dr. Zheng Yun, deputy director of the Warehouse Business Management Division, reported a series of figures to reporters. Without the CIIS system, this 285 million yuan is likely to become non-performing loans.
Not only that, when the system shows that the customer still has a deposit balance in the bank, ICBC can directly deduct the deposit through certain legal procedures to offset the customer's loan debt. When some customers discovered that ICBC had a very clear grasp of their credit history, they took the initiative to repay their debts, not only to ICBC but also to other banks. When credit is closely related to people's social lives, They truly understand the importance of credit.
Opening up "information islands"
Before the CIIS system, ICBC also had a similar "blacklist" system, but these "blacklist" systems were compartmentalized. The various business departments and branches cannot realize complete sharing, just like information islands. This is still ineffective in preventing the credit risk of the entire bank.
There are many such cases. For example, an ICBC Peony Card customer who has had interest arrears for many years and has been written off may become an ICBC consumer credit customer again. Although the consumer credit department will conduct a pre-loan investigation, due to incomplete information, it is likely to issue consumer loans to them again. In the past, this kind of situation was difficult to avoid because bad information in the credit card system was not reflected in the loan department's system. Similarly, if a customer who has outstanding debts from the ICBC Beijing Branch goes to the Shanghai Branch to continue the loan, the Shanghai Branch will not be able to identify the customer's bad credit record. This gives many malicious loan fraudsters multiple opportunities to succeed. Especially when real estate speculation has been very popular in recent years, many real estate speculators have speculated through multiple fraudulent loans between different banks or different branches of the same bank.
According to Dr. Chen Daobin, deputy general manager of the Information Technology Department of the Industrial and Commercial Bank of China, 85% of the non-performing loans intercepted by the Industrial and Commercial Bank of China are cross-regional loans, and 33% are inter-provincial loans. Before the CIIS system, due to information barriers, the cost of mastering customer credit information was very high, and the risk of loan fraud was also difficult to guard against. In the past, Industrial and Commercial Bank of China also considered building such a system, but before realizing the national data centralization, information could not be fully shared, and the bank had very limited control over customer information; even if there was a large amount of customer data stored in the bank's system, the data would not be fully shared. Scattered and fragmented.
After completing the data centralization, ICBC integrated the data at the application level. The CIIS system breaks the business-centered data integration and establishes a customer-centered information system. In the past, each customer's information was thrown in every corner of the room like scattered clothes, pants, and socks. When such clothes, pants, and socks were piled up in the room, it was impossible to identify the relationship between them. Now, each customer's information is clear at a glance.
Zheng Yun said that when commercial banks carry out joint-stock reform and improve risk management levels, data management and mining are inevitable ways. Chen Daobin also said: "We do not lack data, what we lack is refined management of data. After the data is centralized, we still have a lot of work to do." The CIIS system is a successful application after the data is centralized.
Currently, the CIIS system of ICBC is undergoing the second phase of construction. The goal is to integrate this system into the credit management system and become a necessary link in the credit process, achieving a transition from institutional control to technical control. Zheng Yun said that although the system requires staff to query the CIIS system when handling credit business, due to interest relationships, sales staff may not necessarily strictly follow the regulations in order to complete the loan tasks, and risks still exist. Using technology for rigid control can reduce loopholes and risks in human control, while also reducing management costs.
Unified Credit Information System
The CIIS system of Industrial and Commercial Bank of China is a pioneer, but after all, it cannot achieve full sharing among various banks. Malicious loan fraudsters can no longer get loans from ICBC, but can also go to other banks for loans. Similarly, debtors from other banks may also go to ICBC for loans. For the entire banking industry, the ability to prevent credit risks is still relatively weak.
In March this year, the central bank launched a new mortgage loan policy and raised mortgage interest rates. Many banks announced the implementation of upper interest rate ceilings and increased down payment ratios for those who purchase non-first homes. However, this policy encountered "embarrassment" in actual implementation. Some real estate speculators applied for mortgages with different banks. The banks were unable to determine whether the customers purchased a second home, resulting in the failure of bank monitoring.
It is understood that currently only ICBC can achieve cross-professional, cross-regional and cross-branch bad customer information sharing across the bank. Relevant people from Minsheng Bank said that because of its small scale and few branches, Minsheng Bank has realized data centralization very early and adopted a national unified system. All data are centralized and processed at the headquarters, which can functionally implement customer "blacklisting" The maximum sharing of information, but because the amount of data is too small, it is not practical to build a CIIS-like system. Staff from the Science and Technology Department of the Zhengzhou Branch of the Bank of Communications told reporters that the Bank of Communications is actively carrying out national data logic centralized construction work, and currently has not yet achieved full sharing of bad credit customer information across the country.
In order to more effectively prevent credit risks and reduce bank non-performing loan rates, the People's Bank of China organized commercial banks to establish a "Basic Personal Credit Information Database" (also known as the "Credit Information System"). All commercial banks have passed The unified interface sends individual customers' credit information such as loans and repayments, credit cards, and guarantees in the bank to the system, as well as related identification information. On July 1, after a period of trial, this system has begun to be fully networked in commercial banks in 8 provinces and cities across the country, becoming the first credit data information sharing platform for various commercial banks.
Now, when a bank accepts a mortgage business, as long as it calls up the credit information of a loan customer, the customer's outstanding loans in all banks will be clear at a glance, making it impossible to get away with it. Some media even commented that the operation of the "Basic Personal Credit Information Database" will curb the lending and real estate speculation behavior of real estate speculators.
Wang Jianheng, general manager of the Computer Department of China Merchants Bank Beijing Branch, said that the national network of the central bank's "Basic Personal Credit Information Database" is valuable to all banks. Banks have a reliable basis for handling credit business. Reference standards, while eliminating a lot of data investigation work and improving work efficiency. Networking work has a positive significance in preventing non-performing loans. It is understood that during the trial operation phase, Minsheng Bank relied on the central bank's "Basic Personal Credit Information Database" to successfully reject several possible non-performing loans, and ICBC also rejected tens of millions of yuan in non-performing loans based on this.
However, Analysys International Consulting Company is not optimistic about the application prospects of the central bank's "credit reporting system". Analysts of the company believe that the data of the central bank's "Credit Information System" is mainly provided by commercial banks, and the quality of personal credit data currently provided by commercial banks is uneven, especially credit card data. in this way. At present, according to the number of credit cards issued by various commercial banks, the total number has exceeded 20 million. However, there are only about 8 million credit cards with the function of "spend now and pay later", and there are only two or three credit cards that are actually frequently used. Millions of photos. When credit cards are issued, the data collection is rough and incomplete, and a lot of valuable information is not entered into the system. The quality of the data is obviously not high, and a lot of effort is needed to improve the current situation.
Xu Qingsong, chief IT analyst at Analysys International Consulting Company, believes that the central bank's leadership in building a basic database of personal credit information is a very good starting point. But this database requires a process of continuous improvement and adjustment. Credit information involves all aspects of social life, and can be extended to securities, insurance, other financial industries, and other industries in the future, and through data mining technology, more functions can be realized, such as credit warnings, decision-making analysis, etc.
Credit evaluation has a long way to go
Mr. Tu from Standard Chartered Bank Beijing Branch has lived in the United States for many years. He believes that the personal credit system is a huge project that not only involves data collection issues , also involves how to identify, analyze, and evaluate this information. Even if a customer has a credit rating score, it does not mean that the bank is 100% free of credit risk, because personal credit is dynamic and requires time to accumulate. Some people can take advantage of the loopholes in their credit. If they are patient enough, they can continue to take out loans and repay them, and then when their credit has accumulated to a certain level, they can defraud a large loan.
Xu Qingsong of Analysys International Consulting Company pointed out in an interview with a reporter from Information Weekly that banks need to introduce evaluation models and establish a complete evaluation system based on complete data. Because only data cannot explain anything. The central bank's current credit reporting system, as a data collection platform, is more of an information collection channel.
Zheng Yun from Industrial and Commercial Bank of China said that the CIIS system includes a certain credit assessment model.
Based on the central bank's "Basic Personal Credit Information Database", CIIS will make credit ratings for customers based on customer information and a unified assessment model, and incorporate the list and information of bad credit customers into the system. At present, due to the limited amount of information, it is impossible to score the credit of customers without bad credit records. Loan officers often can only rely on some information provided by the customer and judge the customer's credit based on experience, which is risky. For those customers who provide false information, they cannot completely Identify.
In Zheng Yun’s view, what ICBC needs to do in the future is to establish a more complete assessment model and make credit scores for “white customers” who do not have bad credit information, and based on this score , the system determines the loan amount and interest rate, etc. This helps to better segment customers and find a balance between benefits and risks.