| Sharing credit data - the concepts |
 |
 |
 |
| Tuesday, 11 May 2010 02:00 |
|
This article discusses the concept of credit grantors sharing data and in particular, the sharing of positive credit data. This article is illustrated with data sharing examples specifi c to South Africa, but applicable to many markets around the world. Credit Data - Common Perceptions A fairly common perception is that credit bureaux specialise in holding negative information. Typically, people refer to the credit bureau ‘blacklist’ as some sort of ominous tally of negative remarks against individuals and businesses. In reality, this view tells only one side of the story, as the following statistic illustrates: 85% of the consumer profiles held by Information Trust Corporation (ITC), a major Southern African supplier of consumer credit information, contain no judgments or adverse information. In other words, only 15% of the consumer profiles held by this credit bureau qualify for ‘blacklisting’ based on the presence of adverse information. The History of Credit Data The practice of blacklisting described a large part of what credit bureaux concerned themselves with in the past. R.G. Dun & Co., formed in 1841, was the first commercial reporting agency in the United States, and probably the world. The company employed a fleet of investigators to go out and compile handwritten reports on various trading entities and business people. This information, gathered and submitted bi-annually, was collated in ledgers housed at the R.G. Dun & Co head office in New York. Apart from standard business descriptors relating to turnover, assets, liabilities and so forth, this would have included comments that were negative in nature. The dubious task of blacklisting became indelibly associated with credit agencies. During the first half of the twentieth century the main function of credit reporting agencies was to confirm whether or not a person or entity had a record of negative information. However, there were dramatic changes following the Second World War. Around this time, the concept of operations research became a serious academic and business pursuit. Simply put, operations research sought to apply mathematic and scientific terms and solutions to business problems. The field of consumer credit proved no exception and as early as 1950, the possibility of applying statistical methodologies to the industry were being investigated. Coupled with this development, the industrial boom in the United States and Western Europe after the end of the Second World War brought increasing computerization and, in later years, an enormous growth in demand for access to consumer credit. As part of this evolution, credit operations became increasingly complex, as did the sophistication of credit bureaux functioning. Driven by the application of analytical tools to stored data in the late 1970’s and early 1980’s, the credit industry was beginning to understand the inherent value of positive credit information. Central to this was the idea that technology applied to positive repayment behaviour could enhance interpretation and prediction of future payment performance. Examples of Shared Credit Data South Africa followed the trend in the late 1980’s with the realisation by several larger retailers that they jointly had at their disposal an enormous pool of untapped consumer information. Shared amongst themselves, this resource could assist in making better-informed credit and risk decisions. The result, early in 1989, was the formation of the Consumer Credit Association (CCA) in South Africa. This non-profit body was founded by 11 retailers and two local credit bureaux, the latter involved to process and store the data. In addition to upholding a members’ code of conduct, the CCA was given mandate to control the security, privacy and integrity of all data held. The CCA has since grown from the initial 11 founders to a current membership of 110 contributors, jointly providing more than 23 million lines of account payment information to the credit bureaux on a monthly basis. The membership list is no longer confined to the retail sector and has diversified to include, amongst others, major local banks, private label cards, telecommunications providers, mail order companies, finance houses and collections agencies. Contained in this section of the credit bureaux databases is a payment behaviour synopsis for the preceding 24 months, available only to members who contribute regularly to the repository. An additional motivation to contribute and share data is that only CCA members may access the credit bureau scoring products that have been developed, as these draw significantly on payment behaviour information in the score calculation. The following data elements are typically displayed for each payment profile line: •Date the account was opened •Organisation where the account is held •Type of account i.e. revolving, installment •Account opening balance •Account current balance •The total value of all outstanding amounts – the ‘balance versus burden’ •Calendar month the account was last updated at the credit bureaux •24 monthly entries from the most recent to the least recent month - a series of alpha and numeric characters indicating the account status for a given month By referencing payment profile history, the credit grantor can ask specific questions during the application review, especially when processing ‘marginal’ accounts: •Has this person disclosed all relevant account information on their application form? Are there other accounts not mentioned? If so, why? •Have outstanding amounts on other accounts been accurately disclosed? Answering this enables the credit grantor to better determine the applicant’s current debt burden and provide guidelines in limit assignment/decline •Is the applicant only providing information on those accounts that are conducted well, i.e. the ‘best three’ scenario? Trends ascertained provide insight on historical account behaviour and performance In addition, credit bureaux databases, by their very nature, tend to be more dynamic and comprehensive than the databases held by any one credit grantor. This is due to the number of enquiries into the bureaux database from multiple sources. In the course of monthly matching of account information to the credit bureaux database, CCA members can get an indication as to the data quality of their particular portfolio. A lower match rate to the credit bureaux database tends to suggest data issues or poor data management of a portfolio. For example, a high number of telephone or address errors during the matching process would suggest issues with the member’s information capturing process or data housing methods. Whilst the CCA is a good example of a credit data sharing arrangement, the principle of pooling information has not gone unnoticed in other business sectors. In some cases it has taken considerable time for these initiatives to get under way, especially where potential contributors have competitive and strategic concerns about revealing their data to other users. Data confidentiality is one way of overcoming this objection, for example, by disguising customer account details. What often acts as a catalyst for data sharing is where potential sharing partners have a common and pressing problem such as experiencing a high incidence of bad debt or fraud. Some other examples of shared data initiatives in South Africa are: The National Loans Register (NLR) The Micro Finance Regulatory Council (MFRC), the governing body regulating the South African micro-loans industry, established the NLR in late 2000. (A micro loan in this case is defined as a loan below the value of R10,000 and charged an annual interest rate in excess of 27%). As with payment profile information, this data is externally maintained by credit bureaux. At broadest, the NLR aims to assist both lenders and the market they serve by encouraging more responsible lending habits. Specifically, the NLR will collect information from participating micro-lenders regarding loans extended. This includes descriptions as to how many loans an individual has opened with other lenders; who these loans have been taken with; loan repayment behaviour; loan interest charges etc. Participating micro-lenders can then better assess an applicant given the information at hand. For example, the database may be referenced when determining if an applicant can afford another loan, and if they qualify, what the loan amount should be and an acceptable interest rate. Once the NLR database is sufficiently populated the MFRC also plans to provide other industry services such as comparative analysis by portfolio or region. The Equipment Identity Register (EIR) Similar to the operating methodology for the CCA and NLR, the EIR is a database maintained by bureaux on behalf of local mobile telephone operators. The database is a compilation of both positive and negative data relating to the unique identifying numbers on all handsets sold. In the event a telephone is stolen or misplaced, an alert is placed with EIR and the unique equipment number flagged on the database. Within hours the handset is rendered inoperable across all local telephone networks. Should the handset be recovered the block can be removed, making the telephone operable again. Hire Purchase Information (HPI) This is a privately managed repository of original equipment manufacturer (OEM) details for the automotive industry. The database includes vehicle chassis identification numbers, engine numbers, year of registration, vehicle finance history and other relevant information. Matches to the database can be variously positive, for example; verifying vehicle details at the time of sale, or negative, for example; the stated mileage at the time of sale does not correspond with previous odometer reading for the vehicle. The South African Fraud Prevention Service (SAFPS) The SAFPS is the latest collaborative initiative undertaken in South Africa and was launched to combat the incidence of fraud increasingly experienced by South African credit grantors. A not-for-profit company, the SAFPS is modeled on the highly successful UK scheme, the Credit Industry Fraud Avoidance System (CIFAS). In its 13 year existence, CIFAS has saved an estimated £650 million through fraud detection and prevention. The SAFPS, founded by twelve major financial institutions and credit grantors, operates on a shared database - in this instance data that is inappropriate and suspected to be fraudulent. The database, managed by South African credit bureaux, is only available to participating members. Key membership benefits include: •Improved fraud write-offs and associated cost savings •Internet access to a Protective Registration database to verify SA ID numbers •Categorised daily e-mail fraud alerts automatically sent to participants •Access to a dynamic and expanding fraud-specific data source •Improved customer service and response times •Better staff management and business resource allocation •Enhanced interpretation and meaningful risk evaluation of application data •Access to state-of-the-art data analysis software and database management •Regular SAFPS member interaction to address common business problems •Initiatives supported by major SA financial institutions and credit grantors In all instances, information stored on the SAFPS database is not taken as irrefutable evidence of fraud. Rather, SAFPS participants are bound by a code of conduct to further verify those particular account application details that appear to be fraudulent. As part of this emphasis on protecting individual rights and privacy, the SAFPS has made an additional service available so that individuals may list their own personal details on the SAFPS database, for example, in the event a passport is stolen. Summary The opportunities to share data are limited only by the imagination and willingness of potential members to create a joint user forum. Such opportunities will increasingly become available as advances in technology generate more comprehensive and manageable levels of information. History has shown that the nature of the credit industry and the general credit function has continuously evolved and there is little doubt this trend will continue in the coming years. Future evolution will almost certainly centre on a deeper understanding and multi-dimensional interpretation of credit behaviour and profiling of consumers, a process in large part driven by positive data. As it stands, both positive and negative information has a latent value, which can be used to supplement credit and risk management. All things considered, the benefits of data sharing far outweigh the potential drawbacks. About the Contributor: PIC Solutions is the leading specialist credit risk solutions company in the EMEA region. With offi ces in Cape Town, Dubai, Johannesburg, Manama and Nairobi, we deliver integrated analytics, consulting and software solutions to over 150 companies in 30+ countries. We work worldwide with organizations to improve performance, drive strategies and enhance profi tability. Analytics | Consulting | Software |
| Last Updated on Wednesday, 12 May 2010 11:47 |