Wiermanski 175Research results indicate that historical balance and credit limit amounts on consumer credit reports may significantly improve the performance of credit risk models. Initial analytic offerings involving this new information from U.S. Credit Reference Agencies (CRAs) appears to be focused on the “low hanging fruit” of improved credit scores and decision support rules. Hopefully, CRA resources are also investigating new and unproven innovations with this new information, which have the potential to radically improve the marketing, retention and risk management practices of their customers. One such concept that has been discussed for decades involves providing historical consumer credit scores on a current credit report.

Why aren’t historical credit scores available on U.S. consumer credit reports?

From a legal perspective, the U.S. Fair Credit Reporting Act requires CRAs to provide accurate information on a consumer credit report. Credit scores returned with a consumer credit report are calculated based upon information present at the time the credit report was requested and are not considered part of a consumer credit report. Because charged-off trade lines and derogatory public records must be removed from one’s credit report after a certain point in time, historical credit scores stored on one’s credit report which were originally based on information no longer present on a credit file could be considered inaccurate violating the Fair Credit Reporting Act. Imagine the difficulty explaining to regulators or a jury why a credit file without a derogatory trade line, public record a collection item contains a series of low score credit scores.

From a business perspective, CRAs have argued to government regulatory agencies that credit scores are an optional, value added, service and are not part of a consumer credit report. If the CRAs decided to store and return historical credit scores on a consumer credit report then their claim that the credit score is not part of a consumer credit report might become invalid. As a result, U.S. government regulation might require CRAs to provide consumers a free credit score when disputing the contents on their credit report or with free consumer reports. As a result, CRAs could jeopardize hundreds of millions of dollars in revenue from the sale of credit scores to consumers.

A potential approach to circumvent legal and revenue obstacles

Prior to the introduction of time series trade line balance and credit limit amounts it was impossible to calculate a time series of consumer credit scores from one consumer credit report. The addition of time series trade line balance and credit limit information now provides the missing component which allows the computation of a historical credit score(s) at any point(s) in time. By developing a dating process to establish the historical status of trade lines, collection accounts, public records and inquires at specific points in time one can establish which information was present on a consumer’s credit report at a specific point in time and whether or not that information qualified to be included within a particular credit score. For each time period of interest qualifying trade lines, collection accounts, public records and inquiries could be processed by the desired credit bureau based scoring system(s) to produce a time series of credit scores.

Because information used to compute each particular time series of credit scores is based upon information currently available on the consumer’s credit report the legal concern of returning historical scores from inaccurate information would be eliminated and because the credit score time series was calculated at the time the consumer credit report was requested the argument that credit scores are not part of the consumer’s credit report remains intact.

Is there proof that time series credit score information is predictive.

Published research on the value of time series credit scores does not exist. The limited research on the predictive value of changing credit scores is sparse. However, there is a fair amount of published research describing the predictive value of aggregate time series consumer credit report information to forecast consumer credit behavior http://www.forecastingsolutions.com/publications/RMA_2010.pdf. Based on this limited research it appears that there may be significant benefit, especially from credit marketing perspective, from a time series of series credit scores.

An opportunity for credit bureaus to demonstrate their leadership and analytic prowess exists

Because most lenders do not store a time series of monthly credit bureau scores obtained on their portfolios, only CRAs have the ability and incentive to research the potential benefits time series credit score information may offer. Although research of this nature is challenging, expensive and risky, if successful, the leader in this space could be in a unique position to radically redefine the nature of credit bureau information and the landscape of the services offered.

About the Author: Chet Wiermanski is one of BIIA’s contributing editors writing on the subjects of credit scoring and decision systems. He is a Visiting Scholar at the Federal Reserve Bank of Philadelphia researching new applications of consumer credit report information. Additionally, Chet is Managing Director of Aether Analytics which specializes on leveraging hidden data sequences and time series components within consumer credit information typically ignored by traditional credit bureau based solutions. Previously Chet was the Global Chief Scientist at TransUnion LLC. Holding a variety of positions within TransUnion, during his tenure, between July 1997 and February 2012, he was responsible for identifying, evaluating and developing new technology platforms involving alternative data sources, predictive modeling, econometric forecasting and related consulting services.