Old scoring models don't take into account major changes in American earning and spending habits and the gig economy.
The major credit-scoring agencies say they're getting with the times, and are urging mortgage lenders to come with them.
A study commissioned by Experian earlier this summer showed 1 in 3 prospective homebuyers would walk away from a lender who relied only on older credit-reporting models that don't factor in a history of on-time rent and utilities payments.
At a meeting of mortgage business leaders on Tuesday outside of Washington, DC, experts from the credit-reporting agencies said the industry needs to evolve to better account for customers' modern financial habits.
This year the government encouraged wider use of FICO's 10T and VantageScore 4.0, which take into account rental history, utilities payments, and other factors.
It's a change from longstanding government preference—which banks have followed—for more traditional payment history, which exclude them. Instead, traditional credit scores reflect payments of things like cars, credit cards, and mortgages.
At a conference held by the Mortgage Industry Standards Maintenance Organization, which is a part of the Mortgage Bankers Association dealing with mortgage information and business standards, leaders encouraged banks to evolve.
Modernizing credit reporting, experts predict, could help up to 7.7 million people improve their credit scores. Old scoring models don't take into account major changes in American earning and spending habits and the gig economy.
"Seeing risk more effectively is not necessarily taking on more risk,” says Susan Allen, chief product officer of Experian Housing. People in their teens and 20s have very different financial lives than their parents.
That includes longer histories of rent payment, and less of homes. It includes gig economy work, high incomes, and limited traditional credit metrics. Venmo, buy-now-pay-later, and other financial tools even further upend traditional credit.
"It's a completely different world out there," Allen says. "A consumer with a thin traditional credit file is not necessarily a consumer with a thin financial life."
Changing times
The credit-reporting agencies developed more modern reporting systems several years ago, but the industry was slow to adopt them. But the changes are broadly popular with homebuyers, as they often expand credit access for younger borrowers with shorter credit histories.
In a panel, leaders at major credit-reporting agencies—Experian, FICO, VantageScore, TransUnion, and Equifax—all said they're finding the changed models don't equate to riskier borrowers.
VantageScore argues that including on-time rental history and utility payments is a better predictor of mortgage payment than the traditional model allows. Including it would boost millions of credit scores above 620, potentially making them eligible for traditional mortgages.
And it also helps better identify people who might miss those payments, too, says Tony Hutchinson, head of public affairs at VantageScore.
In reality, the modern system presents more relevant data to reporting agencies, not less, Allen says.
"Consumers' financial lives are not going to get any less complex going forward," she explains. "It's going to be incumbent on the five organizations we represent to stay in front of the changing financial lives of consumers."
Artificial intelligence and mortgages
At the same time, credit leaders expect artificial intelligence tools to continue to change the business.
While those agencies don't use AI directly in the information-gathering process, it can sometimes be used to collate the collected data, thus cutting down on the time it takes to analyze a person, says Eric Lapin, head of market strategy and intelligence for FICO's scoring business.
And customers are quick to jump into AI to prepare themselves, too, says Matias Petersen, senior vice president and head of U.S. credit risk solutions at TransUnion.
Consumers "are going to show you much more preparation in the sense of what do they need to do leading up to" getting a mortgage, such as improving their credit scores, Petersen says.
Experian polled 2,000 prospective homebuyers and found a third would walk away from a mortgage lender who won't use credit scoring that includes rental history. (JHVEPhoto - stock.adobe.com) Joyce Walsh, Fannie Mae's vice president of AI and cyber risk management, and Dan Miller, Freddie Mac's senior director of provider strategy and distribution, say both government-sponsored enterprises are trying to scale AI use for business impact. But they are already eyeing risks, such as prompt injection and security.
And Mike Hogan, chief information officer of PennyMac Financial Services, says the entity has encouraged AI adoption for its technologists. That's helped it stop experimenting and start implementing tools, even though it's "burning through" tokens in the process.
"It's really opened their minds to think about the possibilities of what they can do," Hogan says. "They're spending a lot of money building their programs, even if they're building them faster."