
New VantageScore Credit Model Aims To Boost Predictive Performance

Also, company’s pilot program gives nonprofit lenders access to modern credit scoring while helping them maintain sound lending practices
VantageScore has launched its newest tri-bureau credit model, aiming to enhance the predictive power of the bureaus’ scoring capabilities. At the same time, the company expanded its pilot program with non-profit lenders.
VantageScore is an independently managed joint venture of the nation's three nationwide consumer reporting agencies — Equifax, Experian, and TransUnion.
The new VantageScore 5.0 model utilizes the company’s latest patent-pending attributes designed to offer insight into a customer’s creditworthiness. In part, it is trained on consumer loan data following the pandemic, at which point consumer credit behaviors changed significantly, according to the company.
The scoring attributes are said to capture real-time credit behaviors as well as historical trends to improve predictive performance, particularly for consumers with new or thin credit histories. “The credit landscape is evolving rapidly,” commented Andrada Pacheco, executive vice president and chief data scientist at VantageScore.
The testing results for the latest version are said to include a predictive lift of up to 9% on unsecured loans.
“The substantial rise in predictive lift for those with thin, inactive, or young credit files helps lenders like Patelco better identify credit risk and support the responsible expansion of access to credit to a wider population,” Yazel Pardo, vice president and head of credit risk at the Patelco Credit Union in Merced, Calif., said in a statement.
Pilot program
Meanwhile, VantageScore has expanded its pilot program with Washington, D.C.-headquartered Credit Builders Alliance, a national nonprofit network that helps individuals with poor or no credit build their credit. The experiment leverages the VantageScore 4.0 credit score model with open banking data to create a new VantageScore 4plus scoring program.
The collaboration provides nonprofit lenders with access to modern credit scores while allowing them to maintain sound lending practices. “When you add real-time, ‘permissioned’ data to the credit decision process, you not only improve predictive performance, you also create more opportunities for expanding responsible access to credit,” CBA’s CEO Dara Duguay said in a statement announcing the expansion.
Initial data from the pilot show meaningful improvements in the ways nonprofit lenders can assess and serve potential borrowers, including improved predictive insights, clearer risk segmentation and expanded lending opportunities.