The use of more-modern credit scores that incorporate nontraditional data, such as rent reporting, is long overdue in mortgage underwriting. While other industries have adopted newer credit scores, the mortgage industry is still relying on more traditional FICO scores (Classic FICO 2, 4, or 5). Those scores influence decisionmaking for numerous housing finance actors, including lenders, the government-sponsored enterprises (GSEs), servicers, mortgage insurers, credit and rate investors, and regulators.
But in recent years, the Federal Housing Finance Agency (FHFA) has approved the use of new credit scores—VantageScore 4.0 and FICO 10T—that incorporate rent payments in their scoring models. In 2022, FHFA director Sandra Thompson announced that both new credit scores were validated and approved for use by the GSEs. Once implemented, lenders would be required to submit both scores for the loans they sell to the GSEs. Also approved was the move from tri-merge to bi-merge, or requiring two credit reports for borrowers instead of three. And last month, FHFA head Bill Pulte announced on social media that Fannie Mae and Freddie Mac will allow lenders to use VantageScore 4.0 scores effective immediately. The announcement didn’t mention FICO 10T scores.
Though not yet implemented, these changes could have a significant impact on the mortgage industry. At a recent Urban Institute event, The Past, Present, and Future of Credit Scores in Housing Finance, participants discussed the implications and implementation of this announcement. They mentioned that newer scores have greater predictive power in estimating borrower risk and that having a choice of two scores may also lower costs for borrowers in the long run. But it’s also possible that implementing these changes could harm consumers through increases in mortgage rates, at least in the short run.
Clear guidelines on how to implement these new credit scores are necessary to reduce uncertainty and facilitate a successful transition to new scoring models in the mortgage process. Here, we outline key implementation steps and highlight unintended consequences that may affect mortgage pricing and risk evaluation.
What steps are involved in implementing new credit scores in the mortgage process?
The GSEs don’t currently use credit scores in their automated underwriting systems. Instead, they use credit bureau attributes, including trended data, obtained directly from the credit bureaus. But credit scores are used to determine minimum eligibility for borrowers, which is currently a 620 Classic FICO score. (If there are two scores, the lower score is used.) Credit scores are also used in risk-based pricing models that determine fees and interest rates.
Given these varied uses, there are three key areas where the GSEs and FHFA will need to provide additional clarity as to how the new credit scores will be implemented in mortgage lending.
Eligibility thresholds and pricing grids. With the introduction of VantageScore, the GSEs must establish an additional new minimum eligibility threshold and a new grid for loan-level pricing adjustments based on VantageScores.
VantageScores are, on average, 14 points higher than Classic FICO scores, with considerably larger gaps at the lower end of the score distribution. As a result, the VantageScore minimum eligibility threshold and new loan-level pricing adjustments grid should reflect these differences to ensure pricing and eligibility are appropriately aligned. Even so, VantageScores will not rank borrowers exactly the same as FICO scores, and some borrowers may qualify for loans under one score but not the other. The loan-level pricing adjustment they receive could also vary depending on which score is used.
- Adoption of FICO 10T. It’s unclear whether the VantageScore and FICO 10T will be implemented together or independently. Waiting to implement both scores together would allow lenders and investors to update their systems only once, reducing transition costs and operational complexity. But this approach may conflict with the recent announcement that appeared to prioritize the adoption of VantageScore.
- Data disclosure. The GSEs collect and regularly release loan-level origination and performance data to support market transparency and investor confidence, but new data fields will need to be added to these disclosures to differentiate score types. If lenders submit both scores, will both be disclosed or will one be suppressed? If FICO 10T is introduced separately, when will yet more data fields be needed? Without clear answers, data comparability and usability may be compromised.
Score selection could come with unintended risks
The FHFA’s FAQs state that lenders must use only one score per loan, but they can choose which credit score model to use for each loan. This policy provides flexibility but introduces the potential for adverse score selection.
Originators may review both VantageScores and Classic FICO scores for a given borrower and select the score that results in more favorable terms for the borrower. This might slightly reduce loan-level pricing adjustment revenues to the GSEs, but it would have a more significant effect on how other market participants adjust to this shift. For example, mortgage insurers, who use credit scores to price risk, have built pricing models around Classic FICO scores. Although they can recalibrate to VantageScores, the uncertainty of the method of score selection—and the possibility that originators routinely choose the more advantageous score—may lead mortgage insurers to price in an added risk premium. This could ultimately raise insurance costs for borrowers, even those with strong credit.
Similarly, buyers of GSE credit risk transfer deals and non-agency mortgage-backed securities, who currently depend on Classic FICO scores to evaluate credit quality, may impose a similar “uncertainty premium.” Although these investors could also rely on VantageScore or FICO 10T, they may discount credit quality assumptions if there’s no transparency into score selection practices. This pricing conservatism could translate to higher financing costs or reduced investor demand, at least until greater certainty is established.
For borrowers to benefit, the FHFA must ensure a successful transition to new credit scoring models
The decision to allow a choice of credit scoring models in mortgage underwriting brings potential benefits—including greater innovation and broader access to credit—but it also introduces operational, pricing, and transparency challenges, at least in the near term.
To ensure the transition to new credit scores achieves the intended benefits, the FHFA should release clear implementation timelines, standardization in pricing frameworks, and robust data disclosure practices. Although the costs of adverse selection can’t be eliminated, they can be minimized through careful planning, problem resolution, sufficient time to prepare and test systems and mortgage infrastructure, and clear communication. These efforts will be essential to realizing the benefits of the new credit scores while preserving the mortgage finance ecosystem’s integrity and efficiency.
Let’s build a future where everyone, everywhere has the opportunity and power to thrive
Urban is more determined than ever to partner with changemakers to unlock opportunities that give people across the country a fair shot at reaching their fullest potential. Invest in Urban to power this type of work.
