FICO admitted this week that its credit scores don’t do such a good job of predicting which consumers might not pay their bills someday soon. The credit-scoring firm rolled out a new product called the FICO Resilience Index, which theoretically will tell banks which consumers will be able to weather a recession or economic shock — like a pandemic — and which can’t.
That means credit scores *don’t* do that.
For years, I’ve complained that credit scores provide cover to lenders who make bad decisions, and to investors who buy up pools of such bad decisions, but their real-life accuracy is suspect. One piece of evidence I’d glibly point to: The Great Recession. If scores worked so well, why banks lend money to so many people who stopped paying their mortgages, I’d ask? I’d often be told in response that the scores were fine, but lenders misused them. Launch of the Credit Resilience Index defies that logic, at least in part. The design of the product sounds to me like confession that traditional credit scores are inadequate when there are outside forces, like a recession, at play. As the past few months have made abundantly clear, a wide swath of America lives as if every day is an economic shock.
I am thinking of this new scoring tool as a “precarity” index. If you haven’t read the word precarity before, here’s one story I wrote about it recently. The word explains itself readily, however. Any way you look at it, a stunning number of Americans are one paycheck away from a debt crisis, or from losing their health insurance. Most Americans didn’t have anywhere close to the recommended six months’ living expenses saved before Covid hit; roughly three in 10 Americans have no emergency savings. So it’s hard to understand what good a credit score is that *doesn’t* take precarity into account.
All three credit bureaus are offering the Credit Resilience Index to their clients as an add-on to credit scoring products. FICO says the new index is calculated using existing credit data, such as recently opened accounts. That’s also confusing; why not just improve credit scores?
Many consumers don’t realize that credit scores take only an odd slice of their personal financial lives into account. A person with one low-limit credit card but $1 million in savings would likely have a low credit score and have trouble borrowing based on that score. Best as I can tell, the new precarity index will work the same way. FICO will identify consumers who have ramped up credit applications, perhaps predictive of coming credit trouble — but it won’t see who has a year or two of savings squirreled away.
“One of the odd things about credit scoring is how blunt it is,” said Chi Chi Wu, attorney at the National Consumer Law Center. “After all, most people with a low credit score actually do not default on their credit obligations – there’s research that 70 to 80% could actually be good performing borrowers. I appreciate that FICO has developed a score to be more precise. But I wonder who this will benefit – is it your hypothetical person with a $1 million in savings who is a disorganized mess or is it the restaurant worker who lost their job and suffered credit score damage, but will be an excellent borrower once they are employed again? I would love a score that could help the latter person.”
As is often the case, this week’s announcement has been couched in the language of consumer benefit. During the Great Recession, as lenders panicked, many consumers had their credit limits cut — seemingly at random. The Resilience Index will allow banks to be more precise with such actions during economic crises like the current pandemic. The net effect will be that more consumers will be able to keep generous lending terms when they need money most.
“We now can predict which consumers are best positioned to withstand a downturn and which ones are not so well positioned,” Jim Wehmann, executive vice president, at FICO, told the Washington Post. “The FICO Resilience Index can allow lenders to keep credit flowing to even low FICO scorers and below-average FICO scorers who we can identify for the very first time that they are resilient.”
That may well be true. But will it end up kicking consumers when they are down, as many credit decision products do? And will it kick certain groups of consumers more than others?
“Call it a credit score or pretend it’s a shiny new toy and call it a resilience index, it’s still a score that will likely perpetuate the inequities of the racial wealth gap,” said Ed Mierzwinski, director of the federal consumer program at PIRG, a consumer advocacy group. “All other factors being equal, black and brown people have much lower credit scores and, I’d suspect without seeing data, generally less resilience. As for transparency, will FICO or the bureaus now provide consumers with a dual score, eg, 690/42 or 720/21 or 780/65? This product raises more questions than it answers but so far all the info I’ve seen is from the marketing departments.”
There are years of evidence that credit scores are not the colorblind calculations they might seem to be. One example: A study published this year by Duke University’s William Darity Jr. and others examined the never-ending cycle of incarceration, low credit scores, and stunted wealth accumulation that Black Americans face. The research found that Black consumers tend to have lower credit scores than white consumers. The paper includes this stark finding; Blacks who’d never been to jail have average credit scores equal to whites who’d been incarcerated. In other words, the credit score punishment for being Black is the equivalent of having a criminal record.
Like all new scoring products, it may take years before lenders truly incorporate it into their lending decisions; even then, it might only be used as a sort of tie-breaker. Time will tell. In the meantime, consumers should do what they have always done: Pay the bills on time, avoid running a credit card balance as best you can, keep your credit utilization as low as you can, and check your credit report and score frequently for any signs of trouble.
n. Even within a narrow FICO® Score band,
for example near the common FICO Score cut-off of 680, a range of so-called
“stress sensitivity” can be observed. FICO Resilience Index can provide insights
into, which “680s” are more likely to go seriously delinquent when economic stress
is exerted on a consumer population – giving lenders a new tool to use and
potentially avoid taking broad measures that impact insensitive or more “resilient”