Reparations, Race, and Reputation in Credit: Rethinking the Relationship Between Credit Scores and Reports with Black Communities
Credit reports and credit scores reflect stunning racial disparities. Over a dozen studies have found that Black consumers have lower credit scores as a group than whites (listed in this policy brief). As a result, Black consumers have less access to mortgages, credit cards, and even government lending intended to help those most in need — a July 2020 study by ClimateWire found credit scores kept many Black consumers from obtaining Small Business Administration disaster relief loans. Credit reports or credit scores are also used by insurance companies, employers, and landlords, creating racial disparities in those settings as well.
The racial disparities in credit scores are due in part to the racial wealth gap, which makes it more difficult for Black consumers to weather periods of financial distress and to recover from them. It’s a lot harder to keep paying the bills during a rough financial patch, such as the current COVID economic crisis, when Black families have one-tenth of the assets of white families ($17,150 in assets for Black families versus $171,000 for white families). As a result, the credit reports and credit scores of Black families suffer.
The racial wealth gap, in turn, is due to decades of systemic discrimination and centuries of enslavement. Housing discrimination, in particular, is responsible for much of the racial wealth gap. Redlining by the Federal Housing Administration, which explicitly mandated segregation by race, deprived Black communities of the ability to accumulate wealth through homeownership.
Racial disparities in credit scoring are also due in part to current discrimination against Black communities, such as racist policing practices, employment discrimination, continuing housing discrimination and segregation, the school-to-prison pipeline, and poorer health outcomes (as starkly seen during this COVID-19 pandemic). All of these discriminatory practices have an impact on family financial situations. For example, a March 2020 study found that a history of incarceration heavily impacts the credit scores of both the incarcerated individuals and their families, and it’s been well-established that Black communities are racially targeted by the criminal justice system. Black communities also bear the bulk of financial burdens when municipalities make heavy-handed use of criminal fees and fines to obtain revenue.
“Objectivity” Can Be Racist
It’s not surprising for credit scores to reflect the racial disparities in the economic conditions of Black and white communities. As a measurement tool, they work well, in that they reveal the trenchant inequality that current and historical discrimination has engendered. The problem is when credit scores are used as a decisionmaking tool without consideration of these disparities. Using tools that “bake in” racial disparities results in perpetuating and reinforcing these same disparities, and is one of the key elements of structural racism. Disparities in credit scoring are part of the reason for the black-white homeownership gap , which in turns contributes to the racial wealth gap, creating a vicious cycle.
Challenges to the racial disparities are often met by arguments that “credit scores are objective and do not consider race” and “credit scores objectively measure risk, not using them is unsafe and unsound lending.” But so-called objective measures are not free from the legacy and impacts of racism. We have seen numerous other examples of purportedly objective algorithms producing shockingly biased results, from healthcare to criminal justice to employment.
Building a Better System
When faced with challenges to the use of credit history information, defenders of the current system often argue this: if lenders, employers, or decisionmakers cannot use credit history information or the use is limited in any way, they will simply say “no” more often to Black applicants. For example, a June 2020 paper arguing against legislative efforts to suppress negative credit information resulting from the COVID-19 pandemic asserts that such efforts “will greatly reduce access to affordable sources of credit, harming consumers, but particularly lower to moderate income persons, as well as the young and members of minority communities…” A study by Federal Reserve and Harvard researchers claimed that state restrictions on the use of credit reports in employment actually hurt Black applicants because employers then required other qualifications, such as additional education or job experience, that Black applicants are less likely to have.
Defenders of credit scoring note that without scoring, lenders may fall back on more subjective forms of evaluation, which introduces elements of bias. And there is an element of truth in that. Lenders and employers may implicitly (or explicitly) assume that Black applicants are unqualified unless they have a so-called objective measure to counter that assumption. They may use higher credit scores as a requirement to say: “this is a good Black borrower.”
The solution, however, is not to continue using credit scores so that a limited number of Black consumers can get credit to the exclusion of many others. It is to create a better measure to gauge creditworthiness. This could mean tapping different sources of less biased data — admittedly, a difficult endeavor given that any data that is financially related will similarly reflect racial disparities. But it is an endeavor that racial economic justice demands. And while some forms of “alternative data” can be harmful to consumers, including Black consumers, other forms have looked promising.
A better measure of creditworthiness might also involve modifying current credit scoring models to reduce racial disparities while maintaining — and hopefully, improving — predictiveness. Such modifications might need to actively take race into account, but doing so would be an explicit acknowledgment that we need to counter not only 400 years of slavery and legal discrimination, but current existing and virulent anti-Blackness — all of which is baked into credit scores.
Reparations Lending
To address structural racism in the credit markets, we also need reparations in credit underwriting. Reparations are usually thought of as monetary payments. But we also need to undo structural racism in the many systems that hold such importance in our lives — from education to housing to criminal justice. For example, one organization has called for banks to adopt four measures to atone for their role as “underwriters of American racism”: (1) cancel consumer debts for Black customers, (2) eliminate banking fees, including much-abused overdraft fees, for Black customers; (3) provide interest-free mortgages to Black homebuyers; and (4) provide interest-free loans to Black-owned small businesses. The third and fourth measures would be a form of “reparations lending,” and would require foregoing the use of credit reports and scores for those programs.
What would credit reparations for Black communities look like? At a minimum, it should include:
● Stopping or severely restricting the use of credit scores and credit reports in employment, insurance, rental housing, and other non-credit purposes.
● Eliminating the practice of risk-based pricing, in which lenders charge higher rates to consumers with lower credit scores, thus making that credit less affordable and more likely to actually cause the default. Risk-based pricing in mortgages are part of the reason the foreclosure crisis of 2008–2009 caused so much greater harm to Black communities.
● Establishing a national lending reparations program for low- and moderate-income Black communities, which offers low or no interest loans for small business and home purchases and is not based on lending by credit scores, similar to the proposal by PolicyLink. Such a program could constitute a “special purpose credit program” allowed under the Equal Credit Opportunity Act.
● As part of the public option credit registry proposed by Demos, developing a credit scoring model that actively takes past and present discrimination into account and is intentionally designed to reduce racial disparities.
In the wake of George Floyd’s death, banks are now touting their support of the Black community, with even JPMorgan Chase’s CEO Jamie Dimon taking the performative knee. But public statements of support are entirely insufficient if banks continue to blithely use a biased tool they know judges Black consumers worse than whites — worse yet, if deprived of this tool, simply deny credit to Black consumers instead of using less discriminatory alternatives and offering reparations-based credit. In a country that will soon be majority-minority, the continued use of biased-baked algorithms is not only unjust, but a recipe for commercial failure.
For more information, see Past Imperfect: How Credit Scores and Other Analytics “Bake In” and Perpetuate Past Discrimination.
