Impact to VantageScore 3.0 Credit Score Model from Revisions to Public Record Reporting
As the industry expands and consumer needs become more and more complex, the three national credit reporting agencies (CRAs), — Equifax, Experian, and TransUnion—are constantly looking for new methods to ensure that the data they collect for their files are a comprehensive, accurate reflection of consumer’s’ financial history.
In accordance with the National Consumer Assistance Plan, the CRAs are, focusing on best practices and procedures pertaining to the selection, collection, reporting, and updating of public record data found in consumer credit reports.
More specifically, that effort has led the CRAs to a close examination of the manner in which manage public-record data is reported in consumer credit files. Public records that appear on credit files can include bankruptcies, property liens, and court-imposed civil judgments.
The CRAs have not completed their policy review, and no final decisions have been made, but preliminary discussions suggest that one outcome will be a reduction in the volume of tax lien and civil judgment public records recorded on consumer credit files.
In light of these possible changes, VantageScore Solutions conducted a study to assess their potential impact on the accuracy of its VantageScore 3.0 credit scoring model. VantageScore 3.0, like all generic credit scoring models, predicts risk of loan default through statistical analysis of credit file data. Any reduction in the availability of that data can therefore impact the model’s predictive performance and the accuracy of consumer scores.
Public records entries, which are present in only a minority of all consumer credit files, have a negative effect on consumer credit scores, so removing any of those entries from credit files would tend to raise the scores of the affected consumers.
In the event the CRAs revise their public-record policies, lenders will need to understand the impact on credit scores, and adjust their underwriting strategies accordingly. More specifically, lenders will need to understand the degree to which scores could increase, the volume of scores affected by the change, and ranges of scores where the impact is most significant.
To gauge the greatest possible impact of these revisions to models and scores, VantageScore studied a scenario in which all tax liens and civil judgments were omitted from credit bureau files of one CRA for a sample of four million consumers from the U.S. population.