FICO scores are flawed. These lenders say they’ve found a better way to judge your credit

Could FICO scores be going the way of the Dodo? Maybe not in the immediate future, but a growing number of lenders are ditching traditional credit score models to determine creditworthiness and instead using other metrics to make lending decisions.

Traditionally, someone looking for some type of loan—be it a mortgage, auto loan, etc.—could count on the lender pulling their credit report and looking at their credit score to help make a determination of whether to lend the money. But credit scores aren’t perfect, and their flaws, paired with advancements in data harvesting and analysis, have opened the door for a new generation of fintech companies and others to tap into populations who may have previously been unable to access lines of credit.

“We aim to expand access to folks who are not well-served by the FICO system,” says Jason Rosen, cofounder, and CEO of Petal, a company that utilizes sources of data other than credit scores to offer credit cards to individuals in need of credit. “It’s a lot of young people, people establishing or building credit, and students,” Rosen says of the company’s main customer base. “They’ll generally be turned away by banks and credit card issuers, or offered bad financial products.”

A more holistic approach

Instead of using FICO scores to judge a person’s creditworthiness, Rosen says Petal uses an approach called “cash scoring,” which looks at a credit card applicant’s larger financial picture—their income, their spending habits, what bills they pay on a monthly basis, and more. That process, Rosen argues, gives a much clearer idea to lenders of an individual’s creditworthiness compared to a FICO score, which can be slow to change, and doesn’t take numerous intangible factors into account. 

“If you were on an island during 2020, had no idea about the pandemic, and were just looking at people’s FICO scores, you’d have no idea that anything was wrong,” he says. But if you could see, in real-time, that people were losing their income streams and not paying their bills, it would be easy to tell that someone was struggling.

This is exactly what happened to several banks during the height of the pandemic. Many banks were caught “flying blind” as they couldn’t accurately tell who was creditworthy and who was not in mid-2020, according to The Wall Street Journal. And currently, as the economy has grown shakier with interest rate hikes and layoffs, banks are again tightening lending standards for credit cards to a level not seen since the 2008 financial crisis (with the exception of a period of time during the pandemic).

Petal also uses a two-pronged business approach as well: It issues credit cards (Petal Card), but also has a second company, Prism Data, under which it allows other financial institutions and fintech companies to use its underwriting technology and models. Rosen says there’s been “broad interest” in using that technology, and as such, he expects to see more and more lenders moving away from relying so heavily on credit scores to make lending decisions in the future. It seems to be catching on among consumers, too, as Petal has issued more than $1.5 billion in credit since its inception in 2016.

It’s not just credit cards

Another company in the fintech space, MPower Financing, uses “alternative sources of data” to issue student loans, mostly to international students studying in the United States. “International students don’t have social security numbers or credit scores,” says Sasha Ramani, director of strategy at MPower Financing, so the credit-issuing system can effectively shut them out. “There’s no room for nuance with credit scores,” he says, adding that “it doesn’t give a lender a good idea as to how a student will pay back their loans.” 

“If a student has a credit report, we’ll look at it, but we don’t boil them down to a single number [like a credit score],” Ramani adds. “We underwrite based on other factors, like work experience, the school they’re studying at, what they’re studying—it’s a look at the potential of a student’s success after graduation.”

And like Rosen, Ramani also thinks that utilizing other data sources besides the FICO system will prove fruitful for the financial system at large over time. “It’s a model that’s taking off, and it’s mostly opening up credit to immigrant populations, students, or others who have thin credit profiles—it can open up entire new pools of customers,” he says.

That includes segments of the population that may have suffered from inherent bias under the traditional credit score system, too. This is something that’s even caught the attention of members of the Federal Reserve. Speaking at Jackson State University in Mississippi on Tuesday, Michael Barr, the Federal Reserve’s vice chair for supervision, urged regulators to “engage with techniques, such as cash flow underwriting and alternative data, new credit models, and other technologies that hold out the promise to increase access and reduce bias in credit intermediation.”

“Ensuring that all would-be borrowers are treated equally means more customers, more loans, and better returns for banks,” Barr said. In short: adopting new models for judging creditworthiness could not only be a boon for underserved populations, but for lenders, too.

FICO still has its fans

While the tide appears to be shifting on the old credit score model, there are still those who think it has merit. “I’m actually a proponent of the [credit] scoring systems,” says John Ulzheimer, a credit expert and president of the Ulzheimer Group, who also formerly worked at FICO, Equifax, and Credit.com. The systems are “essentially re-engineered every few years so the ‘current’ versions are pretty cutting-edge,” he says. But he adds that he is supportive of utilizing anything that can be used as a “legal and effective” way to make better lending decisions, “whether that’s alternative data, cash-flow underwriting, or something else.”

Though traditional credit scores will likely remain the go-to creditworthiness measuring tool for many lenders in the immediate future, there are many organizations that are already creating hybrid models using other data sources. Perhaps one of the most well-known is Experian Boost, which looks at an individual’s recurring bill payment history and other factors to increase credit scores. And companies like Petal and MPower Financing are only two of an entire cohort of fintech firms in the space as well. Others include X1, TomoCredit, and Zest AI.

So while credit scores remain very much a part of the equation in terms of judging creditworthiness now, things could change in the years ahead—and perhaps rapidly, as more startups and traditional financial institutions start utilizing the heaps of financial data that have become available to them in recent years.

“What’s spurring all of this on is the availability of all this additional financial data,” says Rosen. “We’re in the early stages of the most significant change in the way that creditworthiness is measured in 50 years.”