Proceed with Caution on Credit Scoring with Alternative Data
FICO discards incomplete and old credit history information for a good reason. The solution to today's credit-access problems is not to use this unreliable data but to turn to alternative data such as the payment of everyday bills.June 2
The CFPB found that more than 26 million consumers are effectively "credit invisible" because they have no credit record and another 19 million are "unscored" because they have an insufficient or stale credit history. But it's unclear how the CFPB plans to tackle the issue.May 5
The issue of "credit invisibility" has attracted a great deal of attention in recent months. The Consumer Financial Protection Bureau released in May a study finding that 26 million Americans do not have a credit history, and another 18 million are unscorable because their histories are too limited. The CFPB study also found that African American, Hispanic, and low-income consumers are more likely to be credit invisible.
Policymakers, advocates, and industry have all proposed ways to help people without credit scores get on the grid, including promoting the use of alternative data. But a cautious and thoughtful approach is needed in developing solutions to the problem, since alternative data may do more harm than good.
The manner in which alternative data is used is important. For example, using alternative data to create special scores for otherwise unscorable consumers might be preferable to the wholesale addition of the alternative data to all consumers' traditional credit reports, where it might damage consumers who already have a credit score.
While credit invisibility poses real and significant problems for many consumers, in some areas, no credit history is better than a bad one. When it comes to employment and insurance, a negative credit report or low score could harm job prospects or increase rates. It's often better to be a blank slate. A low score could also put a consumer on the radar of predatory lenders that target vulnerable consumers.
We must ensure that the cure for credit invisibility is better than the disease. Not all alternative data is created equal. Recent efforts to add positive rental data appear to be promising, especially those approaches that do not report late payments prior to the debt being sent to collections agencies.
By contrast, "full file" monthly reporting of gas and electric bill payments could give millions of low-income consumers bad credit scores or make their current scores even worse by adding payments that are only 30 or 60 days late. Most gas and electric companies currently only report late payments on traditional credit reports when they are very seriously delinquent.
The impact of "full file" monthly reporting on gas and electric payments could be especially harsh for families that need a little more time to pay off winter or summer bill spikes. Reporting late payments could also undermine state consumer protections, such as prohibitions against wintertime shut-offs for vulnerable consumers.
Some approaches have attempted to incorporate data from cell phone and cable providers into credit scores. Unlike electric and gas service, these industries have fewer consumer protections that could be undermined by monthly reporting. But the jury's still out on the accuracy of the data and the treatment of disputes over issues like cramming and questionable surcharges.
The scores of vulnerable borrowers may also be hurt if alternative data includes payday loans and other subprime credit. High-cost credit is often designed to trap borrowers in a cycle of debt. And even merely using subprime credit can negatively affect a credit score.
There are many unanswered questions about the predictiveness and accuracy of big data in general. Big data used to determine credit, employment, or insurance is covered under the Fair Credit Reporting Act, and so big-data providers must be in compliance with the law. If big data is used in credit decisions, the Equal Credit Opportunity Act also applies, and lenders must ensure that there is no disparate impact on protected groups.
Addressing credit invisibility should require the industry to examine the basic fairness of credit scoring. Minorities are more likely to be credit invisible, and numerous studies have found that when minorities do have credit scores, as a group, their scores are significantly lower than those of white people.
This is because credit scoring is a reflection of the racial economic divide in this country. Communities of color have less income and far fewer assets to cushion the blow of financial catastrophes such as job loss or sickness. These income and wealth disparities are caused by centuries of discrimination, which still have a huge impact to this day.
Credit scoring perpetuates this unequal playing field. Poor credit scores deny minority and low-income consumers the chance to access affordable credit, insurance, and other services. If they are forced to pay more for these services, the drain on their income necessarily affects their ability to pay their bills on time, let alone build assets to move ahead. Even if alternative data is used responsibly, it will not address the fundamental injustices reflected in credit scoring. At best, it will just lead to more of the same inequality.
Chi Chi Wu is a staff attorney at the National Consumer Law Center with a focus on credit reporting and scoring.