The voices of Urban Institute's researchers and staff
April 13, 2015

Six things that might surprise you about alternative credit scores

April 13, 2015

Consumers without credit scores are essentially ineligible to apply for lending products such as credit cards, mortgages, and auto loans. But recent advances in alternative data and credit scoring methods have made it possible to generate credit scores for millions of otherwise “unscoreable” consumers. Credit scores can give millions of additional consumers access to mainstream lending products, allowing them to forego payday lending or other products that expose them to excessive fees and interest rates, and other undesirable terms.

At a recent panel discussion hosted by the Housing Finance Policy Center, three experts discussed the benefits and challenges of alternative credit scoring. In honor of Financial Capability Month, we share six surprising facts that emerged about alternative credit scores:

  1. Alternative scoring could generate scores for 40 million previously unscored consumers. According to Sarah Davies, senior vice president of analytics and research at VantageScore Solutions, up to 47 million US consumers have “thin credit files.” This means their credit profiles don’t have enough recent activity or history of loan repayment to meet the requirements of traditional scoring models. However by employing alternative segmentation and scoring techniques, as many as 40 million of these 47 million previously unscoreable consumers can now be scored.
  2. Use of alternative data could improve affordability even for those with existing credit scores. Alternative data such as rent, utility and cell phone bill payment history can bring more consumers into the credit mainstream—and increase the scores of those already in the mainstream, according to Michael Turner, CEO of PERC. Noting that especially with respect to utility payments, negative credit behaviors are in credit files already, Turner said that full file capture of this type of data could significantly increase scoreability and, to a lesser extent, actual scores, particularly for young , minority, and low-income consumers. And by taking into account additional data points for consumers with existing credit scores, lenders could extend more credit and offer better terms to more borrowers.
  3. Lenders are beginning to use alternative data. Many lenders and credit-scoring firms have taken note of the potential influx of additional consumers into the financial mainstream and have begun to sift through alternative data to predict borrower default behavior more accurately. Some have even begun lending to borrowers who otherwise would not be considered creditworthy.
  1. Not all alternatively scored consumers are high risk. Addressing general concerns about the higher risk profile of alternatively scored consumers, Turner conceded that these consumers are “somewhat riskier than average,” but added that the group also contains a large number of low- to moderate-risk consumers. This could include thin-file consumers such as retirees who have paid off their debts, or other creditworthy borrowers who stopped borrowing during the recession. According to Davies, of the 40 million alternatively scored consumers, 21 million had scores between 500 and 580, 6 million had scores between 580 and 620, and 7 million had scores over 620.
  2. There are real impediments to widespread adoption. Because alternative data is not reported or collected as robustly as other credit data, researchers are forced to rely on sample data. Kenneth Brevoort, Chief of Credit Information and Policy Section at the Consumer Financial Protection Bureau expressed concerns about using the predictive behavior of sample data to draw conclusions about the entire population. Commenting on the need for more “observable performance,” Brevoort concluded that “…until the predictive power of these data are reliably demonstrated, we should be cautious in advocating the use of such data.” Other hurdles to widespread adoption include legal and regulatory uncertainty, as well as technological challenges associated with mining datasets stored across complex legacy systems.
  3. The GSEs and the FHA may soon play a significant role in bringing alternative credit scoring into the mainstream. The Federal Housing Finance Agency’s 2015 Scorecard for the two government sponsored enterprises (GSEs), Fannie Mae and Freddie Mac, requires them to “assess the feasibility of alternate credit score models and credit history in loan-decision models.”


    Similarly, Secretary of the Department of Housing and Urban Development, Julian Castro recently announced that the Federal Housing Administration (FHA) was exploring the possibility of using new credit scoring models. Given how much influence the GSEs and the FHA have on mortgage lending standards and mortgage credit availability, these initiatives could prove to be crucial in determining how many alternatively scored borrowers, if any, might become eligible for government-backed mortgages.

    Though these moves alone won’t remove all impediments to widespread adoption, the fact that the GSEs and the FHA are seriously evaluating alternative scoring does raise its profile and encourages risk officers to rethink their traditional positions on alternative credit scores. Indeed, these decisions could turn out to be the first serious steps towards broader adoption of alternative scoring.


As an organization, the Urban Institute does not take positions on issues. Experts are independent and empowered to share their evidence-based views and recommendations shaped by research.