You need proof that a protected trait led to worse mortgage terms, plus loan papers showing what the lender did. Mortgage underwriting bias means the review used to approve, price, or deny a loan works differently by protected status. That can mean a higher rate, added fees, or a denial you would not have received otherwise. Your file must connect that outcome to race, sex, age, or another protected trait, not to credit risk alone.
Official resources:
- Register through Govinfo’s official page — Use this page to review requirements and register directly.
- Apply through Consumerfinance’s official page — Use this page to review requirements and apply directly.
Table of Contents
- Which protections apply to mortgage decisions?
- How do courts test for discrimination?
- Why do broad statistics often fail?
- What should you save and file now?
Which protections apply to mortgage decisions?
Federal mortgage bias claims arise under the Equal Credit Opportunity Act and the Fair Housing Act. The U.S.
Department of Justice explains in its Equal Credit Opportunity Act guidance that these laws bar credit discrimination based on race, color, religion, national origin, sex, marital status, age, and other protected traits. Regulators recognize three proof methods, described in the federal banking agencies' 1994 Interagency Policy Statement on Lending Discrimination. Those methods are overt discrimination, disparate treatment of applicants, and disparate impact from a neutral policy.
How do courts test for discrimination?
Disparate treatment asks whether the lender treated you worse than a similar borrower outside your group. The Justice Department used this approach in U.S. v.
GFI Mortgage Bankers, alleging Black and Hispanic borrowers paid higher rates and fees than similarly situated white borrowers. Disparate impact follows a three-step test under the HUD Final Rule on Discriminatory Effects from Feb. 15, 2013. You show the policy caused unequal harm, the lender shows a strong business need, then you can still win by showing a less harmful alternative.
Why do broad statistics often fail?
Numbers alone do not prove a lender caused the gap. On Aug.
5, 2025, a California federal court denied class certification against Wells Fargo for lack of robust causality and common policy-caused harm, as reported by the American Bankers Association Banking Journal in its class-certification ruling coverage. Mortgage data has the same limit. The Consumer Financial Protection Bureau warns that HMDA data omits credit history, debt-to-income, loan-to-value, and other underwriting factors, so reviewers must examine full loan files.
What should you save and file now?
Keep the lender's adverse-action notice. The Consumer Financial Protection Bureau explains in Regulation B Section 1002.9 that the creditor must send it within 30 days of a completed application, with specific reasons, creditor identity, ECOA rights, and the oversight agency. The same notice rule applies to AI and complex-algorithm decisions, under Consumer Financial Protection Bureau Circular 2022-03.
A lender must give the real reasons for denial or pricing, not a generic model or score explanation. Act quickly and keep a complete packet: File promptly because the U.S. Department of Housing and Urban Development notes in its intake and investigation guidance that Fair Housing Act allegations are due within one year of the last discriminatory act.
- loan estimates, rate sheets, fee quotes, and denial letters
- emails, texts, call notes, and names of staff who handled the file
- proof of income, credit, debts, and property value at application time