THE NUMBER DOESN'T KNOW YOUR NAME. IT STILL KNOWS YOUR HISTORY.

VAULT · DEFEATING THE EPSTEIN CLASS

Published by Langhorne P. Breedlove

MONDAY, NOVEMBER 9, 2026 · 2:02 AM ET

Reading time: 9 minutes

Tags: Credit Scores · FICO · Redlining · Racial Inequality · Consumer Reporting · Lending

Dear fellow American proletariat: America has developed a number that can follow you from your first credit card to a mortgage application, a car dealership, and sometimes a rental office. The number claims to summarize risk. It does not tell a lender whether you are honest, generous, competent, or the kind of person who returns a neighbor's borrowed ladder. Yet a complicated human life can be reduced to three digits and handed to an automated gatekeeper wearing an extremely respectable mathematical expression.

The question behind this Vault essay is not whether every credit score was secretly programmed to dislike Black Americans. That would require evidence we do not have. The historical question is more consequential: how did a society that explicitly denied generations of Americans fair access to mortgages, neighborhoods, and financial assets later develop ostensibly race-neutral systems that rely on the records those unequal opportunities helped create?

ROOTS · FIRST THE RED LINES, THEN THE DATA

In the 1930s, federal housing policy helped institutionalize racially discriminatory mortgage practices. The Federal Housing Administration's underwriting guidance favored segregation, including racially restrictive covenants, and the agency systematically underserved many predominantly Black urban neighborhoods (Source 01). The Fair Housing Act of 1968 later prohibited racially discriminatory housing practices. That legislative remedy did not retroactively give denied families decades of home equity.

Popular accounts frequently place all responsibility on the Home Owners' Loan Corporation's color-coded neighborhood maps. The historical evidence is more complicated. Federal Reserve Bank of Chicago research found that FHA exclusionary lending practices developed independently of and before many of those maps, while HOLC itself served some Black and urban borrowers (Source 02). Other economic research finds that HOLC map boundaries had persistent effects on later neighborhood development and credit outcomes (Source 03). Different studies ask different causal questions; flattening their findings into a single villain is not serious history.

Fair, Isaac and Company was founded in 1956. Its general-purpose FICO credit bureau score was introduced in 1989, decades after discriminatory federal housing policies had helped shape unequal household wealth and credit access (Source 04). Those dates are important. FICO did not invent 1930s redlining, and no source reviewed establishes that the 1989 score was intentionally designed to advance white supremacy. A scoring system can nevertheless reproduce historical inequalities without explicitly asking applicants their race.

WHAT EXACTLY IS A CREDIT SCORE?

Credit reporting companies assemble information from lenders and other furnishing institutions. Scoring models evaluate selected elements of a person's reported credit history to estimate a defined lending outcome, such as the likelihood of serious delinquency. The familiar FICO score is not identical to the complete credit report, and there are multiple score versions. A lender may also use income, debt obligations, collateral, employment, and other lawful underwriting factors.

Under the Equal Credit Opportunity Act, creditors generally may not discriminate based on race, color, religion, national origin, sex, marital status, age under specified conditions, receipt of public assistance, or the exercise of protected consumer rights (Source 05). Simply removing a racial identifier from a formula does not prove that all outcomes are fair. Neither does unequal approval prove that every lender violated the law. The inquiry requires specific data, comparisons, legal standards, and causal analysis.

TODAY · WHO GETS DENIED?

The Federal Reserve's Survey of Household Economics and Decisionmaking reported that 33% of adults applied for some form of credit in 2025. Among applicants, roughly one-third were denied or approved for less than requested. Black applicants reported a 54% denial-or-reduced-approval rate—more than twice the rate reported by White applicants (Source 06). This was a survey measure of application outcomes, not a count of discriminatory lending judgments or a direct comparison of FICO scores.

A refusal can reflect verifiable repayment risk, insufficient income, debts, inaccurate records, lender policies, unlawful discrimination, or several factors together. To isolate discrimination, analysts must compare similarly situated borrowers, review underwriting variables, and examine outcomes within a relevant market and period. The existence of racial gaps is a reason to investigate, not a substitute for investigation.

THE INVISIBLE FILE AND THE THIN FILE

People without an established credit record, or whose records lack sufficient recent information to generate a score, can face additional barriers. In June 2025, the Consumer Financial Protection Bureau corrected a widely cited earlier estimate: its original credit-invisible population estimate should have been approximately halved, with more consumers reclassified as having records that were unscorable (Source 07). That correction demonstrates why old headline statistics should not be repeated as if the methodology had never changed.

A person can have a history of paying rent, utilities, or family expenses without a score reflecting every such payment. Incorporating additional payment data may help some applicants, but introduces questions about consent, reporting mistakes, surveillance, and whether missed payments on necessities should become another long-lived penalty. More data are not automatically more justice.

THE ERROR THAT BECOMES A PRICE

The Fair Credit Reporting Act establishes rights to obtain consumer reports and dispute inaccurate or incomplete information. The CFPB explains that nationwide reporting companies must provide free reports under the applicable rules, and consumers may request weekly reports through AnnualCreditReport.com (Source 08). Disputes require a reasonable investigation, though the real-world quality of that process deserves scrutiny.

Medical debt offers a recent example of the importance of legal precision. In January 2025, the CFPB finalized a rule intended to restrict medical-debt reporting and use in lending. On July 11, 2025, a federal district court vacated the rule, finding that the Bureau exceeded its statutory authority under the Fair Credit Reporting Act (Source 09). A proposal to change medical-debt treatment now needs to account for that ruling, statutory authority and the separate question of whether a reported debt should affect credit risk.

THE CASE FOR SCORING, AND THE CASE AGAINST COMPLACENCY

Proponents of standardized credit scoring argue that consistent, validated models can be more predictable than purely discretionary judgments, reduce some subjective bias, and improve pricing of default risk. Lenders must manage real losses and the cost of providing credit; abandoning risk assessment altogether is not an implementation plan.

Critics question how errors, historical disparities, thin files, opaque variables and commercial incentives influence outcomes. A model that predicts past repayment patterns can still deliver unequal results when the underlying society distributed assets and credit access unequally. Predictive accuracy, individual fairness, group disparities, market competition and legal compliance are separate tests. A credit algorithm should have to survive more than one of them.

WHY IT MATTERS

Credit determines not only whether money is available but sometimes how expensive it becomes. More expensive borrowing can increase debt service, make emergency savings harder, and constrain the next financial choice. When inherited disadvantage affects the ability to obtain affordable credit, the consequences can persist even after explicit discriminatory rules have been repealed.

The historical claim supported here is that public and private lending discrimination helped shape unequal access to financial assets, and current credit outcomes remain unequal across surveyed groups (Sources 01, 02, 06). The further claim that a particular scoring formula is intentionally white supremacist has not been established by those sources. The title poses a provocative question; the evidence must decide the answer.

WHAT A REFORM WOULD NEED TO PROVE

American Proletariat's related Justice mandate, Audit & Modernize Credit Scores, could examine data accuracy, consumer appeal rights, explainable adverse-action notices, model validation, competing scoring products, inclusion of verified alternative payment histories, and independent testing for unlawful disparities. None is current legislation merely because the policy has a title.

Federal regulators already require creditors to give specific reasons for adverse credit actions, and the CFPB has explained that vague reasons do not become adequate merely because a lender uses a complex algorithm (Source 10). Reform could improve enforcement of those rights, require understandable explanations and provide effective correction channels, while guarding sensitive personal information.

Replacing one score with a government-issued score would still require choices about data, model governance, lender adoption, appeals, security, bias testing and the right to contest decisions. A public model can be inaccurate too. We should not take an opaque black box, paint it red, white and blue, and declare the Republic healed.

THE TEN-YEAR QUESTION

Any national scoring reform would need a separate first-year forecast and ten-year estimates for data audits, investigations, consumer dispute systems, model research, privacy protection, regulator staffing, lender transition, court challenges and public reporting. Current agency budgets belong in the baseline. Economic benefits from lower errors or wider access to credit must be tested independently rather than recorded as automatic federal revenue.

Comparisons should evaluate accuracy, default performance, costs to borrowers and lenders, protected-class disparities under lawful testing, correction times, data security and unintended exclusions. Without a specified legal mechanism and implementation scope, no defensible national net fiscal cost exists.

A SCORE IS NOT A MORAL VERDICT.

There is nothing radical about asking a lender to explain a decision, a reporting company to correct an error, or a regulator to distinguish a genuinely predictive variable from a proxy for structural disadvantage. There is also nothing responsible about pretending that debt risk vanishes when a score does. The task is to make a useful financial tool accountable to the people whose lives it affects.

The score may not know your race. History certainly knew it. The job now is to discover which remnants still shape financial opportunity, document them, and correct what the evidence and law justify. The mathematics should be examined. So should the people and institutions hiding behind it.

RELATED JUSTICE MANDATES

→ Explore Audit & Modernize Credit Scores

→ Explore Public Banking & Postal Financial Services

→ Explore Break Up Concentrated Corporate Power

SOURCES & FURTHER READING

01 · Federal Reserve History — Redlining, June 2, 2023. https://www.federalreservehistory.org/essays/redlining

02 · Federal Reserve Bank of Chicago, Jonathan Rose — Revisiting How Two Federal Housing Agencies Propagated Redlining in the 1930s, April 2022. https://www.chicagofed.org/research/content-areas/mobility/policy-brief-federal-housing-programs-redlining

03 · Aaronson, Hartley & Mazumder — The Effects of the 1930s HOLC Redlining Maps, American Economic Journal: Economic Policy, November 2021. https://www.aeaweb.org/articles?id=10.1257/pol.20190414

04 · FICO / myFICO — The History of the FICO Score, company founding and 1989 score introduction. https://www.myfico.com/credit-education/blog/history-of-the-fico-score

05 · Consumer Financial Protection Bureau — Credit Discrimination Is Illegal, Equal Credit Opportunity Act. https://www.consumerfinance.gov/consumer-tools/credit-reports-and-scores/credit-discrimination-is-illegal/

06 · Board of Governors of the Federal Reserve System — Economic Well-Being of U.S. Households in 2025: Credit, 2026. https://www.federalreserve.gov/publications/2026-economic-well-being-of-us-households-in-2025-credit.htm

07 · Consumer Financial Protection Bureau — Technical Correction and Update to the CFPB's Credit Invisibles Estimate, June 23, 2025. https://www.consumerfinance.gov/data-research/research-reports/technical-correction-and-update-to-the-cfpbs-credit-invisibles-estimate/

08 · Consumer Financial Protection Bureau — Consumer Reporting Companies; report access and disputes. https://www.consumerfinance.gov/consumer-tools/credit-reports-and-scores/consumer-reporting-companies/

09 · Consumer Financial Protection Bureau — Medical Information Rule and July 11, 2025 Vacatur. https://www.consumerfinance.gov/rules-policy/final-rules/prohibition-on-creditors-and-consumer-reporting-agencies-concerning-medical-information-regulation-v/

10 · Consumer Financial Protection Bureau — Circular 2023-03: Adverse Action Notification Requirements, September 19, 2023. https://files.consumerfinance.gov/f/documents/cfpb_adverse_action_notice_circular_2023-09.pdf

10 RECOMMENDED ARTICLES, ESSAYS & POLICIES

→ Explore Audit & Modernize Credit Scores

→ Explore Federal Reserve, Currency & Financial Reform

→ Explore Why Can't Everyone Bank at the Post Office Again?

→ Explore National Climate Adaptation

→ Explore Public Ferries, Waterways, Locks & Water Access

→ Explore Greenbacks

→ Explore Libertarians

→ Explore Reformers, Abolitionists & Civil Rights

→ Explore Elite & Industrial Power

→ Explore Public Transportation for Everyone

WE ARE ONE.  ·  E PLURIBUS UNUM.