AI and alternative data could help millions gain access to credit
LSU and Harvard research shows more borrowers could become eligible if lenders use artificial intelligence (AI) and alternative data, such as education and employment history
“Even though the vast majority of the U.S. population—80 percent—never defaulted on a loan, less than half have access to prime credit. With smarter credit models, lenders could approve almost twice as many borrowers, with fewer defaults.”
Lenders and borrowers would benefit from increased reliance on alternative data, instead of just credit scores, in evaluating a person’s default risk, according to the study.
Alternative data can include educational degrees, employment history, rental payments, asset ownership, spending behavior or social media.
While credit scores are reliable for assessing past creditworthiness, they’re not quite as good at predicting the future. This leaves traditional lenders, who primarily rely on credit scores, at a disadvantage compared to lenders that use advanced technology and alternative data to model risk.
Researchers looked at outcomes for more than three million personal loan applicants—some of whom were approved, some denied—and compared those to traditional lending models.
The LSU-Harvard study showed that, with alternative data and AI, lenders can segment borrowers with different levels of risk despite similar credit scores. The model was able to identify high-risk borrowers even among applicants with high credit scores.
Broader use of alternative data and AI in lending decisions could not only allow millions of borrowers access to credit, but those approved for loans could also get a better interest rate.