Redlining was federal policy, not fringe conduct. Beginning in the 1930s, two New Deal agencies built a national system for denying mortgage credit to Black neighborhoods. The Fair Housing Act of 1968 outlawed explicit discrimination, but did not reverse the structural conditions it created. Those conditions persist today through property tax-based school funding, algorithmic lending disparities, and a student debt system that extracts wealth from communities redlining never allowed to accumulate it.
How Redlining Was Built
The standard account of redlining begins with the Home Owners’ Loan Corporation — the New Deal agency whose color-coded neighborhood maps became the era’s most recognizable symbol of institutional racism. The account is not wrong, but it is incomplete in a way that matters. Because if the HOLC gets all the credit for redlining, then the practice looks like the product of one misguided bureaucracy, something that was corrected when the maps stopped circulating. That is not what the record shows.
Federal Reserve economists have spent years digitizing actual loan records from both the HOLC and the Federal Housing Administration to understand which agency’s practices actually shaped the mortgage market. Their conclusion, published by the Chicago Fed, is that the red lines drawn by the FHA were likely far more impactful than the HOLC’s. The FHA largely excluded core urban areas and Black mortgage borrowers from its insurance operations, while the HOLC did not. The FHA was not reacting to the HOLC maps. The FHA developed its own methodology to redline core urban neighborhoods, which it did from day one of its operations.
This distinction is not academic. It means that the federal government’s most consequential tool for shaping the American housing market — mortgage insurance — was designed from its inception to exclude Black homebuyers. The HOLC maps became the iconic image of redlining partly because they are visually legible. The FHA’s block-level grading system, used to determine where its insurance would and would not flow, was less visible and more operationally significant.
The term “redlining” stems from the neighborhood maps drafted by the HOLC. HOLC tasked local officials to classify neighborhoods using four color-coded letter grades: D = “hazardous” (red), C = “definitely declining” (yellow), B = “still desirable” (blue), and A = “best” (green). “A” ratings were assigned to affluent white neighborhoods, while D ratings were assigned to neighborhoods with a greater share of Black, lower-class, or immigrant residents. Virtually all Black families lived in D-rated areas. The assessors’ field notes made no attempt to conceal the basis for their judgments. In Miami, a neighborhood was downgraded because of its proximity to “dump and Negro area.” In Portland, Oregon, a declining area was attributed to “infiltration of subversive racial elements.”
Those notes are not the exception to the system. They are the system’s documentation of its own logic.
What the Policy Produced
The Chicago Fed’s research into the long-run effects of the HOLC maps found that the maps led to reduced homeownership rates, house values, and rents and increased racial segregation in later decades, concluding that the HOLC maps had meaningful and lasting effects on the development of urban neighborhoods through reduced credit access and subsequent disinvestment.
Reduced credit access is doing a lot of work in that sentence. What it means in practice is this: when a neighborhood is denied mortgage lending, residents cannot build home equity. When they cannot build home equity, they have no wealth to pass to their children. When they have no wealth to pass to their children, those children have no down payment, no cushion for a financial emergency, no resource to finance a college education without debt. The deprivation compounds with each generation. It does not require any subsequent discriminatory act to continue producing racially disparate outcomes. The initial structure is sufficient.
The numbers that describe what this produced are not historical. According to the U.S. Census, in 2023, white homeownership rates were 73.8 percent, compared to only 49.8 percent and 45.9 percent for Hispanic and Black homeownership, respectively. The George Washington University’s research on this gap describes it as roughly as large as it was in 1968 — the year explicit discrimination became illegal. The Fair Housing Act removed the legal sanction for the practice. It did not reverse the structural conditions the practice had created over the preceding three decades.
The wealth implications are direct. Black household wealth increased from 12.7% to only 15.5% of median white household wealth between 2019 and 2022. Home equity accounts for the largest share of that wealth. Racial disparities in homeownership continue to throttle this growth, with only 44% of Black individuals owning a home, compared to nearly 73% of white individuals — a vestige of discriminatory housing practices such as redlining and blockbusting.
The homeownership gap between Black and white households is not a residual effect that will close on its own trajectory. It has not closed in the 57 years since the Fair Housing Act. The same property tax mechanism that tied school funding to home values is still in place. The same credit scoring practices that disadvantage renters and people without inherited assets are still in place. The structure that produces the gap is still producing it.
Modern Redlining: The Practice Continues Without the Maps
Calling contemporary lending discrimination “modern redlining” is not rhetorical shorthand. It is the terminology used by the Department of Justice and the Consumer Financial Protection Bureau in active enforcement proceedings. The maps are gone. The outcomes are documented in current HMDA data, tested against peer lender benchmarks, and adjudicated in federal court.
In October 2024, the CFPB and DOJ announced action against Fairway Independent Mortgage Corporation, the nation’s third-largest mortgage lender in 2023. From 2015 through 2022, Fairway operated three retail loan offices and three loan production desks located in real estate offices in the Birmingham metropolitan area, all of which were in majority-white areas. Fairway predominantly directed its marketing to majority-white areas. Only 3.7% of Fairway’s applications from 2018 through 2022 were for properties in majority-Black areas, compared to 12.2% for Fairway’s peer lenders. No maps required. The same exclusion, accomplished through branch placement, referral network construction, and marketing strategy.
This is not an isolated case. In January 2023, a California bank settled allegations of redlining for $31 million — the largest-ever bank redlining settlement at the time. The DOJ alleged the bank avoided making mortgage loans in majority-Black and Hispanic neighborhoods in Los Angeles, with other banks in their market receiving six times more mortgage applications from these neighborhoods. Of the 11 branches it opened or acquired in Los Angeles over 20 years, just one was in a majority-minority community.
Since the Combating Redlining Initiative was launched, DOJ has filed 11 redlining complaints and simultaneously entered into 12 redlining settlements resulting in $122 million in relief. The initiative has also expanded its targets to include independent mortgage companies, which accounted for 72.1 percent of all first-lien residential mortgage originations in 2022.
The enforcement record demonstrates two things simultaneously. First, that the practice has not stopped. Second, that the infrastructure of enforcement has been built and is functional. What remains uncertain is whether it will be used. Beginning in 2025, the DOJ sought and obtained court approval for the early terminations of multiple redlining consent orders, including the September 2022 consent order with Evolve Bank and Trust, the October 2023 consent order with Ameris Bank, and the January 2024 consent order with Patriot Bank. The CFPB also moved to rescind at least one consent order during this period, a motion the court denied. The enforcement posture has shifted.
The algorithmic dimension of modern lending discrimination is not hypothetical. Research published in peer-reviewed economics journals documents that disparities in interest rates and loan approval rates persist for minority applicants even when controlling for credit scores — meaning the discriminatory outcome survives the nominally race-neutral screening mechanism. Algorithmic systems trained on historical data reproduce historical patterns. This is not a flaw in the systems. It is how they were built.
The Classroom Is Downstream of the Redline
Property tax-based school funding is the mechanism through which redlining’s suppression of Black home values continues to determine educational outcomes for children who were not born when the practice was formally occurring. The logic is inescapable: if redlining reduced home values in Black neighborhoods and increased them in white ones, and if schools are funded through taxes on those home values, then the educational resources available to children in formerly redlined areas are a direct function of the original deprivation.
Research bears this out explicitly. Within metropolitan areas, the HOLC redlining maps from 80 years ago consistently correspond with district racial and ethnic composition, school neighborhood poverty, and K-12 funding adequacy today. The Annenberg Institute at Brown University’s analysis of school district data found that districts and schools currently located in formerly redlined neighborhoods have significantly less per-pupil revenues, larger shares of Black and non-white student bodies, less diverse student populations, and lower average test scores compared with those located in neighborhoods that were not redlined.
Schools in redlined areas face stark funding disparities compared to schools in areas rated more favorably. On average, they spend nearly $2,500 less per pupil than schools in top-rated zones, and over $3,000 less than schools in neighborhoods rated second-tier. Differences in home values largely explain that gap. Local education revenues derive from property taxes. Expensive houses produce more cash for neighborhood schools. The pattern compounds: better-funded schools increase property values, which increase school funding, which increases property values. The cycle at the top reinforces the cycle at the bottom.
This is not a side effect of school funding policy. It is school funding policy operating as designed, in a housing market that was deliberately distorted along racial lines and has not been structurally corrected.
Student Debt as a Downstream Product of Housing Policy
The connection between redlining and the student debt crisis is not a rhetorical leap. It follows a documented causal chain. Redlining suppressed Black homeownership. Suppressed homeownership means suppressed home equity. Suppressed home equity means no wealth transfer to the next generation. No wealth transfer means no parental savings, no help with a down payment, no resource to finance a college education without borrowing. Black families, denied for generations the primary mechanism through which white families accumulated and transferred wealth, are structurally dependent on debt to access the same educational credentials.
The research on this is clear. In order to finance higher education, Black families — already disadvantaged by generational wealth disparities — rely more heavily on student debt, and on riskier forms of student debt, than white families do. Due to lower family wealth and racial discrimination in the job market, Black students are far more likely than white students to experience negative financial events after graduating, including loan default, higher interest rate payments, and higher graduate school debt balances.
The education funding gap produced by redlining’s property tax effects further concentrates Black students in the highest-cost and least-accountable institutions. Underfunded K-12 schools produce weaker preparation, which narrows the range of institutions students can access, which funnels them toward for-profit colleges and under-resourced open-enrollment institutions where the debt-to-outcome ratio is worst. The pipeline from redlined neighborhood to high-cost, low-quality credential to unpayable debt is not accidental. It is the structural consequence of policy choices made at every stage of the chain.
This is where it becomes important to scrutinize not just the original practice but the current management of its downstream. Clutch Justice has reported on Ascendium Education Group, a nonprofit that controls a substantial share of student loan guaranty operations and wields significant influence over how Pell Grant funding flows to institutions. The organizations managing the infrastructure of student debt — determining which institutions receive guaranty backing, which students receive Pell eligibility, which default rates trigger institutional consequences — are operating at the exact intersection where redlining’s educational legacy becomes actionable debt policy. That infrastructure warrants the same structural scrutiny as the lending practices that preceded it. [See: Ascendium’s Control Over Student Loan Infrastructure and Pell Grant Funding]
What Institutional Design Requires to Persist
Redlining does not need to be actively practiced today in order to continue producing its original effects. It needs only for the conditions it created to remain uncorrected, and for the systems that interact with those conditions — school funding, credit scoring, appraisals, loan guaranty operations — to continue operating on rules that treat the present distribution of wealth as a legitimate baseline.
The mechanisms that carry it forward are documented:
Home values depressed by 80 years of deliberate exclusion still determine the educational resources available to children in formerly redlined neighborhoods. The funding gap has widened over time, not narrowed.
Standard credit scoring models weight mortgage and installment loan history heavily while historically discounting rent and utility payments. Families excluded from homeownership for generations are structurally disadvantaged in the credit scoring systems used to determine access to the homeownership their parents and grandparents were denied.
Research has documented persistent undervaluation of properties in majority-Black neighborhoods and overvaluation when ownership appears to change to white hands. The appraisal profession remains nearly 90 percent white. The valuations that determine home equity — the primary wealth vehicle available to Black families — are subject to documented racial bias at the point of execution.
As enforcement actions against Fairway, the California bank, and others document, modern redlining does not require a map. It requires consistent decisions about where to place branches, whose referral networks to cultivate, and which neighborhoods to target with marketing. Each decision is individually defensible as business strategy. The aggregate effect is geographic exclusion based on race.
Machine learning models used in lending, insurance, and credit decisions are trained on historical data that reflects decades of discriminatory outcomes. A model optimized on historical default rates inherits the distortions that exclusion from credit access produced. Algorithmic neutrality is not structural neutrality.
What the Counterargument Gets Right — and Wrong
The counterargument to characterizing present disparities as effects of redlining takes several forms. One version argues that current disparities reflect income differences, not discrimination — that if you control for income, the racial gap in homeownership largely disappears. A second version argues that the Fair Housing Act’s enforcement mechanisms, supplemented by the CRA and HMDA, provide adequate ongoing correction. A third version accepts the historical analysis but questions whether the legal and policy interventions available today can or should be directed at reversing conditions set 80 years ago.
The first argument is partially true and analytically insufficient. Income differences do explain a portion of the homeownership gap. But the income differences themselves are partially a product of the same exclusion: neighborhoods denied investment develop fewer businesses, generate fewer jobs, produce weaker tax bases for public services, and funnel residents toward lower-wage employment. You cannot control for income as an independent variable when income is itself a downstream product of the policy you are analyzing.
The second argument has become harder to sustain in 2025. The enforcement infrastructure built under the Combating Redlining Initiative is being dismantled through consent order terminations and reduced CFPB activity. The argument that existing mechanisms provide adequate correction depended on those mechanisms being actively used.
The third argument deserves honest engagement. The timeline between the policy and its present effects spans generations, multiple administrations, and institutional actors who were not parties to the original decisions. Designing interventions that are both legally sustainable and structurally adequate is genuinely difficult. Acknowledging that difficulty is not the same as concluding that no intervention is warranted. The structural mechanism is still running. The question is whether any institution with the power to modify it is willing to do so.
Sources
