Fannie Mae and Freddie Mac, government-sponsored enterprises that collectively guarantee the default risks for a large share of U.S. single-family mortgages, are adopting scoring metrics that incorporate available rent, utilities, and telecom bill payment data when assessing the default risks of mortgage applicants. This In Focus reviews the motivation for and some issues pertaining to this transition.
Background: Traditional Versus Alternative Credit Scores
A consumer credit score is a numeric metric used by lenders to determine the likelihood of default on a consumer loan. Consumer scores are computed using information obtained from one or more consumer reports. Such information, collected by nationwide credit reporting agencies (CRAs), includes consumers' credit repayment history, the amount of outstanding debts relative to credit limits, the length of credit history, and the frequency of new credit requests. From these data, consumer credit scores can be generated, representing the likelihood that prospective borrowers will repay their loans in a timely manner. Borrowers with higher credit scores are considered more creditworthy, meaning that they are less likely to default on their loan obligations. Credit scores, therefore, influence consumers' credit access and loan terms.
The Consumer Financial Protection Bureau (CFPB), however, used data from December 2010 to estimate that approximately 11% of the U.S. adult population was credit invisible, meaning that they had no credit tradelines (i.e., credit bureau records) at any of the three largest credit bureaus (i.e., Experian, Equifax, and TransUnion). The CFPB also estimated that an additional 8.3% were nonscorable, meaning they had thin (e.g., short or outdated) credit histories that cannot be used to compute accurate credit scores. Hence, updating conventional scoring metrics with additional non-credit repayment tradelines may enhance credit access.
Specifically, alternative credit scores are consumer metrics that incorporate information not generally incorporated into traditional credit scores, such as recurrent non-credit payments for rent and utilities. (See CRS In Focus IF11630, Alternative Data in Financial Services, by Cheryl R. Cooper.) The more comprehensive payment history may enhance the predictability of consumer metrics. For example, individuals without rent tradelines can still have rent debts in collection reported to credit bureaus by third-party debt collectors. (For more information, see CRS Report R46477, The Debt Collection Market and Selected Policy Issues, by Cheryl R. Cooper.) In these cases, determining repayment risk based upon isolated negative incidents may not be as accurate as using more comprehensive histories showing consistent payment patterns over time.
Rent Reporting Developments
Before alternative credit scores are computed, rent payment histories must first be collected and digitized. Before the 2000s, rent was largely paid via either cash or check rather than electronically. Renters could request and retain dated receipts from landlords for payment verification. For tenants who paid their rents via check, their (paper) bank statements could serve as official verification. Although loan applicants could turn over receipts and bank statements for review by lenders, papers can get lost, be damaged, and take time to be reproduced. Over the past two decades, however, more rents are being paid by electronic means (e.g., debit or credit cards, rent payment apps), thus enhancing the ability to collect and report these data for generating alternative credit scores.
Although rent reporting is not yet widespread, recent trends indicate that it may expand. Rental payment data have largely been reported by property management companies or landlords of multifamily properties (e.g., those with at least five or more residential units)—particularly large apartment complexes. Property managers that screen numerous tenant applicants over the course of a year are more likely to find a database of rent payment behavior patterns useful. Moreover, the per unit costs of financial reporting decreases with the volume of reporting. Consequently, property management companies (relative to small landlords) are more likely to participate in the reporting of rental payment information, thus supporting the growth of the rent reporting industry.
Some states and localities are experimenting with rent reporting for landlords, primarily focusing on residents receiving housing assistance or with low incomes. For example, these pilot programs often require landlords receiving local, state, or federal housing subsidies to offer to their tenants the option to have their rental payments reported. Under these pilots, state housing authorities are also required to offer rent payment reporting options to tenants who either reside in properties under their administration or receive state housing vouchers.
Fannie Mae and Freddie Mac, collectively known as the Enterprises, provide liquidity to U.S. single-family and multifamily mortgage markets. The Enterprises have encouraged their borrowers of multifamily loans (e.g., owners of apartment buildings) to report timely rent payment data of tenants to a vendor that subsequently forwards the information to credit bureaus. Fannie Mae offers to pay one year of fees directly to the vendor. Freddie Mac provides closing cost credits to its multifamily borrowers that participate in rent reporting.
Many landlords do not report rent data because they are not covered by any legal mandates, have not obtained financing from the Enterprises, or choose not to incur costs to transmit data. Consequently, tenants can sign up directly with vendors and pay for reporting services. Some vendors' platforms may allow tenants—after creating their online payment accounts—to send automatic reminders or rent payment transfers, further reducing the probability of being delinquent. Some tenants may be able to lower the fees for this service by selecting vendors that report to only one rather than to multiple credit bureaus. Some vendors may be able to subsidize these costs for low-income tenants by using revenues generated by higher income tenants as well as by appealing to landlords to participate on their platforms.
Despite recent developments, the volume of rent reporting is still considered low. In a November 2022 report, the CFPB estimates that 1.7% to 2.3% of adults living in rental housing are likely to have their rent payments reported to credit bureaus. Thus, many updated credit scores are still unlikely to include this information due to lack of availability.
Enterprises' Updates to Credit Scores
Section 310 of P.L. 115-174, enacted in 2018, establishes requirements for the Enterprises' use of third-party credit scores. On October 24, 2022, the Federal Housing Finance Agency (FHFA), their primary regulator and current conservator, approved the inclusion of two credit scores—FICO 10T and VantageScore 4.0—for all Enterprises' single-family mortgage purchases. This transition is expected to be completed by the fourth quarter of 2025.
Both FICO 10T and VantageScore 4.0 incorporate traditional credit repayment data as well as alternative data such as rent, utilities, and telecom repayment data. The updated credit scores are considered more accurate because they incorporate trends in consumer debt accumulation and repayment behaviors—in addition to changes and averages—over a 24-month period (as opposed to a snapshot at a single point in time, typically on the day a loan applicant would apply for a mortgage). The scores, however, have some differences. For example, the numerical ranges used for interpreting credit scores differ for the two metrics. The firms also apply different weights to the different categories (e.g., payment history, credit utilization) that comprise their credit scores. The FHFA director, however, testified that various performance tests revealed minimal differences between the two credit scores, and they both exhibit acceptable levels of accuracy.
Tri-Merge Versus Bi-Merge
The Enterprises will switch from a tri-merge system, in which mortgage originators were sending three scores with every mortgage delivery, to a bi-merge system, sending only FICO 10T and VantageScore 4 scores. For investors in the mortgage-backed securities (MBSs) markets issued by the Enterprises, the bi-merge system may improve transparency. Specifically, credit scores are informative of both credit risk, the risk of borrowers repaying late or not at all, and prepayment risk, the risk borrowers will repay their mortgages ahead of schedule. (Higher credit scores typically indicate both lower credit risk and higher prepayment risk.) Because investors in MBSs issued by the Enterprises face exposure only to prepayment risk, the disclosure of one fewer credit score may simplify the ability to estimate when an MBS might generate less cash flow due to early mortgage repayments. Less complexity for investors aligns with FHFA's initiative to minimize pricing differentials between the Enterprises' MBS issuances, discussed in CRS Report R46746, Fannie Mae and Freddie Mac: Recent Administrative Developments, by Darryl E. Getter.
The transition from tri-merge to bi-merge still poses a risk of information loss. Individual CRAs have their own policies for storing and reporting information. For example, one CRA may report delinquent debt obligations separately from the penalties and fees; another CRA may combine both items into one entry. Consequently, when creating credit scores, firms that purchase consumer information from different CRAs may acquire similar but different information. Furthermore, individual scoring firms have their own proprietary statistical models, decide what consumer information should be included and excluded from calculations, and choose whether to equally weight or place heavier weights on information more recent or deemed more pertinent—perhaps as instructed by the requestor of the scores. To the extent using more data sources can be more informative, then transitioning from tri-merge to bi-merge may result in some loss of information that could arguably benefit invisible and nonscorable consumers.
Continuing Underwriting Challenges
Credit scores are only part of the overall underwriting process. Underwriting—the process of assessing the creditworthiness of prospective borrowers—is typically more expensive and time-consuming for those who lack sufficient credit histories, face greater income variability, or lack collateral sufficient for backing loans. Incorporation of alternative credit scores may still not increase credit access for these groups. For example, some people may still lack digital non-credit payment histories if they are paid in cash, engage primarily in cash transactions, or choose not to have bank or credit union checking accounts. Furthermore, the inclusion of rent and utility tradelines may enhance accessibility to and lower the costs of consumer credit (e.g., credit cards, automobile loans) for some invisible and nonscorable individuals, but it may not significantly increase the share of creditworthy mortgage applicants due to other requirements. Some individuals may still lack the Enterprises' minimum tradeline requirements, currently either three credit or four non-credit tradelines of recurring or fully repaid obligations. (Various federal agencies that provide mortgage guarantees generally require at least two tradelines that can be credit, non-credit, or a combination.) For this reason, the mortgage industry (particularly small loan originators)—which is incurring fixed costs to ensure that mortgage originations comply with the Enterprises' updated credit score requirements—is somewhat skeptical of the benefit in terms of the total return on investment.