Cash-flow data can predict creditworthiness, new study shows
The use of cash-flow data — the money flowing in and out of customer accounts every month — may be useful in predicting creditworthiness, according to a study released Thursday.
FinRegLab, a nonprofit that tests new technologies to foster an "inclusive financial marketplace" analyzed the data of six nonbank lenders that use cash-flow data in their underwriting: Accion, Brigit, Kabbage, LendUp, Oportun and Petal.
While acknowledging the limitations of their analysis to date, including that the study doesn't reflect the full U.S. population and range of financial services products available, the nonprofit found encouraging indications that cash-flow analysis can be used as a predictor of creditworthiness on its own and in combination with credit scores.
The study gives nonbank lenders like Kabbage and Upstart additional ammunition in their efforts to prove that alternative data can be a better judge of creditworthiness than the traditional model that relies on credit bureau information and FICO scores. FinRegLab, which is the first to do an independent study on the idea, found that the use of cash-flow predicts loan performance, helps provide access to credit to borrowers who ordinarily wouldn’t be eligible for it (for instance, if they had no credit bureau file), and appears to meet the requirements of fair-lending rules.
“This is an encouraging set of results and it makes us think that it is worth investing some time and effort to keep going in this area,” said Kelly Cochran, deputy director at FinRegLab.
Cash-flow data is not as exotic as some other types of alternative data, such as education information.
“This is really going to core financial stability and financial habits, the kinds of things that underwriters have always looked at in lending,” Cochran said. “So it's much closer to the traditional questions even if it's using a slightly different data source to answer them.”
Some bankers are loath to change the underwriting process they've used for decades, due in part to fears about regulatory compliance. Studies like FinRegLab's could help provide comfort both for regulators and financial institution executives.
“We talk to a lot of different financial firms and bank and nonbank lenders, and there are lots of arguments made for trying out new data or trying out new methods for underwriting,” said Melissa Koide, founder and CEO of FinRegLab. “But when you're sitting on the government side, what you really want to know and have are fact-based insights to help then empirically evaluate the benefits and risks associated with new types of data. We undertook this research to generate those independent fact-based analyses so that regulators, policymakers and the banking and fintech industry have more facts to then think about where there may be opportunities and value of evolving regulations.”
Koide noted that even with confidence about the use of cash-flow data, regulators and bankers still have to grapple with questions about data sharing and consumer consent.