Can the Flaws in Credit Scoring Be Fixed?

Not easily.

David Goldman / AP

That credit scoring and reporting is an opaque and flawed system isn’t a new revelation. But recent news that two of the three major credit-reporting agencies were allegedly providing customers with different scores than they advertised and signing them onto subscription-based services without proper disclosures has hit a nerve with consumers. People aren’t just upset about being misled, but because the news served as a reminder of how vulnerable Americans are to the policies of powerful, private credit-reporting agencies.

A credit score is the most ubiquitous indicator of financial responsibility. Americans rely on credit-reporting agencies to provide a fair portrait of financial health not just for their own reference, but for reporting that can be used by banks and potential employers too. The current system leaves millions with low scores that hinder their financial access despite healthy records of paying day-to-day bills responsibly. And tens of millions Americans have no score at all because of limited credit history. That means they’re essentially stuck.“If you think about the credit-invisible population in this country, their ability to enter the financial mainstream and access affordable credit instead of payday lenders and pawnshops and check-cashing services is tied to what's in their credit report,” says Michael Turner, the president of the Policy and Economic Research Council. “They're caught in the credit catch-22: In order to qualify for credit you have to have already had credit.”

Right now, the three major credit reporting agencies—Experian (which which settled a lawsuit with Federal Trade Commission in 2005 over misleading marketing), Equifax, and Transunion (the latter two both were recently fined by the CFPB for misleading consumers)—all receive data from credit-card companies, banks, and other large lenders about whether or not borrowers pay on time. For the millions of Americans without credit cards or loans at big financial institutions, the three largest credit-reporting agencies receive virtually no data about their finances, unless a household account—such as gas or water—becomes extremely late, leaving a negative mark on an otherwise blank credit report, which can cause a disproportionately large drop in score. Consumer advocates say this system is inherently unfair: excluding and punishing the millions of Americans who pay everyday bills such as rent, phone lines, and utilities responsibly and rewarding those who show similar levels of responsibility but are wealthy enough to have mortgages or plentiful access to credit lines. Darrick Hamilton, an economist and professor at the New School agrees with Turner, “The way scores are calculated definitely privileges individuals with capital in a different way than working-class individuals.”

Turner and other consumer advocates think that the answer to better, healthier credit scores—particularly for Americans at the margins of the financial industry—is the use of alternative data. They are pushing for what is known as full-file credit reporting, which would allow landlords, utility companies, phone companies, and a host of others to report payments to credit agencies. This additional data, some say, would allow people to build healthy credit histories by doing what they are already doing—paying everyday bills. “This is a simple solution that won't cost the government a penny, that builds on an existing framework, and can be implemented almost immediately,” Turner says. “I can't think of a more gift-wrapped social and economic policy that should have broad bipartisan support.”

And yet, the plan has its dissenters. Chi Chi Wu, a staff attorney at the National Consumer Law Center says that there’s reason to be cautious about the use of additional data as a way to improve credit reporting, particularly for low-income families. “We're not necessarily opposed to all alternative data, but we also don't think it's the be-all and end-all,” she told me. The NCLC’s concern is that by dumping a bunch of new data into credit profiles the risk will increase that Americans who already have passable credit scores will be exposed to a wave of negative data points. She says instead that some data points are more promising than others. Rent payments, for instance, might be a particularly useful indicator since tenants are motivated to pay in a timely manner, but the NCLC is especially wary of including reporting from gas and electrical utilities.

“We're concerned that this push toward full-file utility credit reporting, on a monthly basis, will result in a lot more negative information,” she told me. She says that it’s especially of concern for low-income families who might fall behind on a utility bill for a month or so when it is extremely hot or cold. In those instances, month-to-month reporting may not provide an opportunity to catch up before borrowers have a negative mark added to their credit report. In some cases, remaining invisible might be preferable to a bad score. That might further disparate-impact issues in the credit industry. The trouble is, the data points that the NCLC is most worried about might also prove the easiest to include, since the industries are already regulated and automated, making data collection and dissemination easier.

Another option, Wu says, is that alternative data can be culled but not added to traditional credit reports. Instead this additional information would create a different report and score, such as FICO XD—which collects data on utilities, phone accounts and other bills for those who don’t have enough traditional data to merit a credit score. But whether or not lenders will actually use it, or any other secondary scoring method remains to be seen.

Turner suggests that the fears of the NCLC about alternative data causing more harm than good are “unsubstantiated,” and that the broad consensus among consumer advocates supports the push for the use of alternative data. A slightly different interpretation of past research on the topic—which includes work from the Brookings Institution, the Philadelphia Fed, and Mission Asset Fund along with PERC, NCLC and many others—suggests that while no one quite knows what the ideal mix of alternative credit-reporting sources should be, many think that ultimately incorporating more data into the most-used credit reports will do more to help economically marginalized groups than to harm them. In other words, the potential downsides of alternative data might be worth the risk.

But there are also those who would prefer a completely different credit-scoring system. Hamilton argues that credit scores shouldn’t even be calculated and disseminated by private companies, but instead by the federal government. “The intent of credit scores is to provide information and facilitate markets. That’s something the private sector shouldn’t be doing,” he said in an interview. Relying on private companies, he says, results in a system in which those who collect and disseminate information are more beholden to the financial sector than to borrowers, since that sector is their client. The government, he suggests, would have a more clear mandate to be accountable to borrowers, and upfront about their practices. And, he adds, if that didn’t work at least the American people would have a clear form of recourse for the system not operating well: voting people out of office (though that plan may have its flaws too).

While public credit reporting might seem promising, in practice it likely isn’t tenable, according to Turner. “We don't see any cases globally where you have an optimally functioning public-credit registry,” he says. “They're not really attuned to the needs marketplace. They're not innovative. They don't make investments in relationship-building with lenders. They don't make investments in alternative data-scoring models. So, there are real questions about the social and economic efficacy of a public-credit registry.”

One thing that everyone seems to agree on is that implementing changes to help those struggling with financial invisibility probably isn’t top of mind as Congress returns and a new president is inaugurated. “This is certainly not a key issue for them,” says Turner. “This is nothing that's going to be prioritized within the first 100 days, and I just don't know what the future will bring on this issue.” That makes one of the first steps in Turner’s ideal plan, passing the Credit Access and Inclusion Act, which would introduce payment data from sources such as landlords and telecommunications companies into the credit reporting system, unlikely. But while a federal push on the issue might be the fastest or most powerful means of implementing change, it isn’t the only path. Hamilton notes that if states or municipalities were to come up with a more inclusive way of determining credit scores, and encourage lenders and businesses in their jurisdiction to use it, it might provide a model for a larger system.