Since lending first began, banks have always used information from customers’ accounts to determine whether they were creditworthy. But most online lenders of consumer loans and cards don’t have a well of customer data to tap into. A credit card issuer is drawing on bank account data to make credit decisions — which will help underserved borrowers and those with no or low credit scores, its executives said.
Like many lenders, Genesis Financial Solutions was looking for a way to lend to people who are creditworthy, but have a thin or nonexistent credit file. Some are young, others recent immigrants and some have simply not applied for credit in the past. Most traditional loan models would discard such applicants.
What Genesis found, and plans to deploy by the end of the year, is technology Nova Credit is rolling out Tuesday called Cash Atlas. It analyzes a consumer’s bank account information (with their permission) to assess cashflow, debt repayment habits and overall creditworthiness.
Genesis Financial Solutions expects to see “a meaningful percentage increase” in the number of credit applicants it can approve, said Jeff Giacomi, senior vice president of credit risk and pricing at Genesis Financial Solutions (before that he was at Wells Fargo for 12 years, then at Deserve).
Genesis has 5 million customers of near-prime credit cards, some its own and others private-labeled for 600 merchants including Home Depot, Wayfair and Ashley Furniture. When a retailer partner declines a potential credit card holder, Genesis gives that borrower a second look. Interest rates on the cards range from 14.9% to 34.9%. The median rate of interest across all credit cards is 19.63%, according to Investopedia.
Genesis plans to use Nova Credit’s technology to extend credit to customers on the margin of approval and to thin-file and no-file applicants.
“If you don't have a long tenure or you are brand new to credit, you're not going to have a robust score,” Giacomi noted. “And the bureaus aren't going to be able to give you much data to help you make an informed decision. Looking at a customer's deposit account and understanding how well they manage their finances can give you great insight into how likely they are to repay their debt. We're looking to leverage this data to say yes more often to applicants.”
Before he joined Genesis, Giacomi used bank account data in underwriting decisions at Deserve and at Wells Fargo, which like many big banks uses its own customer data when making underwriting decisions.
“What's changed in this environment is that this information is now available for the fintechs and other lenders out there to augment the traditional credit data that everyone has access to,” Giacomi said. “So I think this is going to be a competitive advantage for those that leverage this data set.”
For several years, Nova Credit has
“If you think about what we've managed to do with the Credit Passport, we're taking someone who has no financial identity here in the U.S. or very little identity,” said Misha Esipov, Nova Credit’s CEO. “And while they are applying for a financial product that they need, whether that be a credit card or an auto loan or an apartment lease, we give them the tools to insert more information about themselves, to paint a more complete picture of who they are. And when we do that, we allow a lender to make a more fair and responsible underwriting decision.”
Cash Atlas does the same thing, but using bank transaction data instead of foreign credit bureau data. It also can underwrite loans — Esipov thinks of it as a “cashflow underwriting as a service” product.
Nova Credit categorizes and analyzes transactions to identify paychecks, debt payments and other relevant attributes. It also provides adverse action codes to borrowers its system declines.
Esipov noted that other attempts have been made to recognize credit applicants’ recent good behavior in their credit score, such as Experian Boost and Ultra FICO.
"The problem with those approaches is that they don't serve the consumer at the point of need,” he said. "And require the consumer to know to go to their website prior to applying for a financial product. And perhaps that lender will accept that data. Maybe."
Nova Credit is used at the point of need, when the consumer is applying for a financial product.
“That fundamentally changes the game,” Esipov said. “The consumers that can benefit from this are not only the super users who know that they can go and improve their score, but those that are actually applying for products and in doing so can actually increase their likelihood of approval.”
Cashflow underwriting is the way banks made loan decisions before there were credit bureaus, and it’s still the way many mortgages are underwritten, Esipov pointed out.
“A hundred years ago, you would use your existing cash balance or your existing proof of income,” he said. “You would walk into a bank branch, bring your bank statements with you.”
Esipov acknowledges that bank account data by itself may not be enough to assess creditworthiness.
“We haven't seen the proof points yet to suggest that bank transaction data alone can outperform bureau data,” he said. “The way we think about it is it's an incredible complement for people that don't have bureau data.”
Banks can use Cash Atlas to feed Nova Credit’s data and analysis of creditworthiness attributes to their existing lending systems. This is what Genesis plans to do.
The reason banks typically don’t just pull transaction data into their loan decision systems themselves is that it's hard, Esipov said.
“If you go and speak with a credit risk officer, the vast majority of them will say yes, I believe that the bank transaction data has an incredible amount of insight to help more people be eligible for approval,” he said. “But to actually develop the systems to do that is incredibly difficult.”
Giacomi backs this up.
“There are providers out there that will give you raw transaction data, and that's very difficult to work with,” he said. “What you want are the key attributes on someone's checking account. And that is what Nova Credit will provide. It'll be a robust set of attributes that then you can use to develop models on. That's the game changer.”
A credit card applicant might have a thin credit bureau file but no overdrafts in the last 12 months, consistent inflows of income and a healthy average balance, for instance.
“That really indicates that that applicant is managing their finances in a responsible manner,” Esipov said. “And we'd be more likely to approve that applicant than if we didn't have that information or that information showed some negative marks.”
He believes using this technology will help Genesis meet its financial inclusion goals and help the underbanked.
“What we are doing is saying yes more often,” he said. “So we're looking at people who we decline today and finding a way to approve them.”
