The Equal Credit Opportunity Act (ECOA) is a United States federal law, enacted in 1974, that prohibits creditors from discriminating against credit applicants on the basis of race, color, religion, national origin, sex, marital status, age, or because an applicant receives income from a public assistance program. The law applies to all types of credit, from mortgages and auto loans to credit cards and business lending.
ECOA also grants applicants specific procedural rights, including the right to receive a written adverse action notice explaining the reasons for a credit denial or unfavorable change in terms, giving rejected applicants a basis to understand and, if warranted, challenge an unfavorable decision.
ECOA in plain terms
ECOA says a lender's decision to approve or deny credit, and the terms it offers, must be based on legitimate financial factors — income, credit history, debt levels — not on who the applicant is in terms of protected characteristics like race, sex or age. A lender who denies credit for a prohibited reason, or even one who cannot demonstrate a legitimate financial basis for a denial, has violated the law.
The requirement to provide reasons for a denial is one of the law's most practically useful protections, since it prevents lenders from simply declining applicants without any accountability or paper trail.
Protected characteristics under ECOA
ECOA specifically prohibits discrimination based on race, color, religion, national origin, sex, marital status, age (provided the applicant has the legal capacity to contract), and the fact that an applicant's income comes in whole or part from a public assistance program, or that the applicant has exercised rights under consumer protection laws.
Regulation B and enforcement
ECOA is implemented through Regulation B, which fleshes out detailed compliance requirements for creditors, including how applications must be evaluated, what can be asked of applicants, and the specific content and timing required for adverse action notices. The Consumer Financial Protection Bureau has primary rulemaking and enforcement authority, alongside other federal banking regulators and the Department of Justice for pattern-or-practice violations.
Impact on lending practices
ECOA has shaped standard lending practices such as requiring consistent, documented underwriting criteria applied evenly across applicants, and restricting creditors from asking about certain personal characteristics during the application process that have no legitimate bearing on creditworthiness.
ECOA's treatment of spousal credit history
ECOA specifically addresses a historical practice of denying married women credit in their own name or requiring a husband's co-signature regardless of the wife's own independent income and creditworthiness, and it generally prohibits a creditor from requiring a spouse to co-sign unless the applicant's own individual qualifications genuinely do not meet the creditor's standards.
ECOA and business credit applications
Although most commonly discussed in the context of consumer lending, ECOA's anti-discrimination protections also extend to business credit applications, meaning a small business owner cannot be denied a business loan on the basis of protected characteristics any more than an individual consumer applying for a personal loan could be.
Disparate impact claims under ECOA
Beyond intentional discrimination, courts and regulators have recognized that a lending practice which is neutral on its face but produces a disproportionately negative effect on a protected group can also violate ECOA under a disparate impact theory, broadening the law's reach beyond cases where a lender's discriminatory intent can be directly proven.
ECOA's influence on modern underwriting technology
As lenders increasingly rely on algorithmic and machine-learning models to make credit decisions, regulators have emphasized that ECOA's protections apply fully to automated decision-making, requiring lenders to be able to explain and justify an adverse decision even when it was generated by a complex statistical model rather than a human underwriter.