The FHFA is implementing sweeping changes to mortgage credit assessments beginning in 2026, removing minimum credit score requirements and introducing new scoring models that incorporate trended and alternative data. Fannie Mae has already eliminated its minimum score threshold, while lenders prepare to adopt VantageScore 4.0 and FICO 10T. The shift promises broader access to credit, but underwriting standards remain strict, meaning borrowers will need to navigate mixed adoption across lenders.

The rules governing how Americans qualify for mortgages are entering their most significant transformation in decades. Beginning in 2026, the Federal Housing Finance Agency will require lenders selling conforming loans to Fannie Mae and Freddie Mac to adopt new credit scoring models—changes that could reshape access to homeownership, especially for first-time buyers and historically underserved borrowers. The shift comes as part of a broader mandate to modernize mortgage risk analysis and reduce reliance on traditional, static credit scores.

Fannie Mae Drops Minimum Credit Score Requirements

The most immediate change is Fannie Mae’s elimination of its minimum credit score requirement, effective November 15, 2025. The move represents a structural shift in how eligibility is determined for more than half of U.S. mortgages. While credit scores were once a hard threshold, Fannie Mae now relies on a more holistic risk framework that incorporates borrower reserves, debt levels, property characteristics, and loan purpose.

For buyers financing homes under $832,750 without using government-backed mortgages, these changes could broaden access to conforming loans. However, analysts note that removing minimum credit score requirements does not equate to easier approval. Borrowers without traditional credit histories may still face stringent underwriting, particularly as lenders adjust to the new models at varying paces.

New Credit Models Introduce Trended Data and Alternative Credit

The introduction of VantageScore 4.0 and FICO 10T marks a modernization of credit assessment tools used in mortgage markets. Unlike classic FICO, these models incorporate “trended data,” which evaluates borrower behavior over time rather than a single snapshot. That includes month-to-month credit utilization, spending patterns, and payment habits—data points considered more predictive of future default risk.

Equally significant is the incorporation of alternative credit inputs such as rent, utility payments, and phone bills. Industry leaders argue that this approach better captures the financial reliability of millions of consumers who lack deep credit histories. FICO executives emphasize that the expanded data sets provide “a more nuanced understanding of borrower behavior,” potentially opening the door for young adults, renters, and minority borrowers who have historically struggled to qualify under legacy scoring.

Lenders Face Transitional Friction Despite Expanded Access

Despite the structural overhaul, the mortgage underwriting process itself remains unchanged. FHFA Director William J. Pulte underscored this point, noting that lenders are still required to meet established risk standards. As a result, the rollout could create an uneven landscape in which some lenders adopt the new models early while others continue to rely on classic FICO scores out of caution or operational inertia.

For borrowers with thin files or no traditional credit, securing approval may still depend on finding lenders willing to use alternative data. Firms such as Guild Mortgage and AmeriHome are already aligning with the new standards, but broader lender adoption may take time.

Looking Ahead

The FHFA’s overhaul signals a long-term shift toward more inclusive and dynamic mortgage underwriting. Still, 2026 is likely to be a transitional year marked by uneven adoption, lender discretion, and borrower confusion. Over time, expanded credit visibility may reshape demographic access to homeownership, but the practical impact will depend heavily on how quickly lenders integrate — and trust — the new models.