Published on September 17, 2025
Introduction: A Pivotal Shift in Fair Lending Oversight
On August 29, 2025, the FDIC announced a major shift in its supervisory approach that has mortgage organizations scrambling to update their compliance efforts. Going forward, the FDIC will evaluate potential discrimination under the Equal Credit Opportunity Act (ECOA) and the Fair Housing Act (FHA) only through evidence of disparate treatment, eliminating disparate impact analysis from its Consumer Compliance Examination Manual. This decision impacts key sections of the manual – specifically, Fair Lending and Unfair, Deceptive, or Abusive Practices (UDAAP) – removing all references to disparate impact and how examiners are to evaluate it.
This follows a broader regulatory trend: the OCC made similar changes earlier in 2025, and these actions stem from Executive Order 14281 (“Restoring Equality of Opportunity and Meritocracy”) issued by President Trump, directing agencies to remove disparate-impact liability from their enforcement toolkits.
What Was Specifically Changed – and What Wasn’t
Removed Enforcement Components
Any FDIC guidance referencing policies or practices that are facially neutral but have a disproportionately negative effect on protected classes – even when not intentionally discriminatory – has been stripped. Examiners will no longer assess statistical disparities or outcome-based analysis to determine fair lending risk. The updated manual no longer references these so-called “effects tests.”
Continuing Enforcement Components
Disparate treatment remains prohibited. That means any policy or practice that intentionally treats someone differently based on a protected characteristic (ex: race, sex, national origin) is still actionable under ECOA and FHA. The underlying laws and private rights of action remain intact. The Supreme Court’s decision in ‘Texas Department of Housing & Community Affairs v. Inclusive Communities Project (2015)’, which confirmed the FHA’s jurisdiction over disparate impact issues, has not been overturned.
Why It Matters to Mortgage Lenders
With disparate impact effectively removed from FDIC supervisory scrutiny going forward, mortgage lenders must recalibrate compliance frameworks. Policies that previously were vulnerable due to outcome disparities, such as minimum loan amounts or geographic overlays, are now less likely to trigger FDIC fair lending exam findings, unless there's clear evidence of intent to discriminate.
Lenders can and should still audit loan programs, pricing, underwriting rules, and credit-scoring models for unintended consequences. However, the emphasis will be less on statistical disparities and more on whether decision-makers intended or knowingly facilitated discriminatory outcomes.
Previously, a strong fair lending compliance program included statistical models and testing to detect disparate impact, controls or mitigation plans when such disparities were discovered, and documentation of business necessity and less discriminatory alternatives.
Now, while statistical tools can still inform internal risk oversight, their role will be diminished from defensive to more diagnostic. Documentation of intent, rationale for decision-making, underwriting exceptions, and executive oversight will likely become even more critical.
Simply put, FDIC examiners will no longer demand disparate impact risk assessments, evidence of whether neutral policies disproportionately burden protected groups, or business necessity defenses for policies generating disparate outcomes.
However, they will expect clear reasoning behind policies and decisions, transparent escalation where exceptions or discretionary underwriting has occurred, and robust record-keeping around decisions affecting protected classes.
Though the FDIC has dropped disparate impact from its manual, some state regulators or state laws (including UDAP statutes) may still consider outcome-based discrimination. For example, in Massachusetts, the attorney general recently used disparate-impact theory under state law to penalize a lender whose AI pricing model disproportionately harmed protected groups.
Mortgage companies operating in multiple jurisdictions must continue to assess the risk across federal and state realms. Even if FDIC risk is lower, exposure under state laws or through private lawsuits still remains.
With AI, machine learning, and automated underwriting more prevalent than ever, lenders must revalidate their models for bias by design or effect, ensuring transparency in algorithms. How these models govern intent and fair treatment should be closely scrutinized, even if the FDIC won’t examine for impact disparities.
Continuing to monitor possible outcomes is still important so that any disparities are discoverable and addressed before they create potential liability under non-FDIC frameworks.
Key Actions for Leaders to Take
Conclusion
The FDIC’s elimination of disparate-impact evaluation from its exam manual is a seismic shift in fair lending oversight. Mortgage leaders must adapt, but should not relax. Disparate treatment remains a primary enforcement vector, and broader legal and reputational risks still loom where outcome disparities remain unexamined or unaddressed.
The industry’s response should be a renewed focus on intent-based compliance, clear and well-documented underwriting and pricing logic, forward-looking model governance, and careful vigilance regarding state and third-party enforcement. Staying ahead of this shift means ensuring your compliance machinery remains rigorous, transparent, and aligned with both current regulatory guidance and broader fair lending imperatives.
One of the most insidious forms of discrimination in lending over the past several decades has been...
Having a compliance management system (CMS) is a well-known regulatory imperative, but did you know...
Manage your compliance confidently with our easy-to-use, affordable suite of regulatory compliance products.
Try ActiveComply Today!