December 12, 2025
Acting Director Russell Vought
Consumer Financial Protection Bureau
1700 G Street, NW
Washington, D.C. 20552
Re: Equal Credit Opportunity Act (Regulation B); Docket No. CFPB-2025-0039 or RIN 3170-AB54 (90 Fed. Reg. 50901, Nov. 13, 2025)
Dear Acting Director Russell Vought:
The National Community Reinvestment Coalition (NCRC) appreciates the opportunity to comment on the CFPB’s proposed rulemaking to amend Regulation B of the Equal Credit Opportunity Act (ECOA). We express concerns about the proposed rule’s attempt to eliminate disparate impact liability, narrow the scope of discouragement claims, and prohibit (explicitly for some groups and implicitly for others) the implementation of Special Purpose Credit Programs (SPCPs). The proposed rule will make it even more difficult for underserved markets to access credit, particularly communities of color that continue to experience redlining and lending discrimination.
NCRC is a network of more than 700 community-based organizations dedicated to creating a nation that not only promises but delivers opportunities for all Americans to build wealth and attain a high quality of life. We work with community leaders and policymakers to advance solutions and build the will to solve America’s persistent racial and socio-economic wealth, income, and opportunity divides, and to make a Just Economy a national priority and a local reality. NCRC also conducts testing, including mystery shopping and matched-pair testing, to investigate fair lending practices and access to credit, and uses those findings to advocate for stronger enforcement of fair lending laws, better banking practices, and policy changes to combat appraisal bias and disinvestment in communities of color.
I. Disparate Impact
Intent in discrimination may not be explicitly obvious and can be difficult to prove. Certain policies that are facially neutral may unnecessarily or arbitrarily exclude socially disadvantaged groups from fair and equal access to credit, and the disparate impact standard helps to identify such policies and mitigate those unwarranted disparities.
For decades, the disparate impact doctrine has driven lenders to identify and improve policies and procedures that unnecessarily exclude applicants from credit opportunities. The Supreme Court recognized the disparate impact doctrine three years before Congress enacted ECOA, holding that a particular statute prohibited not only “overt discrimination” and practices with “discriminatory intent” but also “practices that are fair in form but discriminatory in operation.”[1] In deliberations on the original bill, Congress rejected language prohibiting only “invidious” discrimination, choosing instead the broader term “discriminate.”[2] And the legislative history of that Act, and the amendments that quickly followed, demonstrate that Congress intended for disparate impact to be available under the amended ECOA.[3]
The federal agencies responsible for the enforcement of ECOA then consistently interpreted it to encompass claims of disparate impact.[4] For example, in 1994, ten federal agencies with ECOA enforcement responsibilities issued an interagency Policy Statement on Discrimination in Lending identifying “three methods of proof” to establish discrimination under ECOA, including “evidence of disparate impact.” [5] The statement explained that, for example, consideration of gross income without distinguishing between taxable and nontaxable income would have a disparate impact on elderly applicants, who are more likely to receive nontaxable income. It noted that lenders could address this issue by simply “grossing up” nontaxable income. That change—which is standard practice today—benefits lenders, who have a more accurate picture of applicants’ income, and it increases credit opportunities for elderly applicants.
This is the thrust of the disparate impact doctrine – a requirement that lenders identify the business necessity of policies that impact certain protected classes more than others (which should presumably be straightforward), and to assess whether it could meet such needs using policies or practices that have less impact. For example, a lender may require applicants to meet a minimum credit score to qualify for a mortgage. However, this policy can disproportionately exclude Black borrowers who are more likely to have low credit scores or be credit invisible because of a lack of access to mainstream financial services, and therefore face difficulties obtaining safe and responsible loans and building credit.
In this way, disparate impact has proven to be a valuable tool to eliminate practices that unfairly exclude groups without any sufficient justification and to counteract unconscious prejudices in the credit context.
The Supreme Court has also upheld the use of the disparate impact standard as a legal tool to protect groups from civil rights discrimination in employment and housing, for exactly these reasons. In Texas Department of Housing & Community Affairs vs. Inclusive Communities Project, 576 U.S. 519 (2015), the Supreme Court affirmed that disparate impact claims are cognizable under the Fair Housing Act (FHA), explaining that:
“Recognition of disparate-impact liability under the FHA [Fair Housing Act] also plays a role in uncovering discriminatory intent: It permits plaintiffs to counteract unconscious prejudices and disguised animus that escape easy classification as disparate treatment.”[6]
Just as the Supreme Court concluded that disparate impact furthers the purposes of the Fair Housing Act, disparate impact furthers the purposes of ECOA. The Bureau’s argument to the contrary is belied by the Supreme Court’s reasoning in Inclusive Communities, which upheld the use of disparate impact under section 3605 of the Fair Housing Act, which makes it unlawful “to discriminate“ against a person “because of” an enumerated protected class.[7] Here, the statutory use of the word “discriminate” (which has historically been used to encompass the various methods of proof of discrimination, including disparate impact), the conspicuous use of the phrase “on the basis of” (which connotes impact even more strongly than the “because of” language used in section 3605), the legislative history accompanying the Act discussed above, the contemporaneously-promulgated interpretation in Regulation B, the uniform position of the regulators for fifty years, the unanimous opinions of the courts that have addressed the question, and Congress’s repeated implied ratification over decades all confirm disparate impact claims are cognizable under ECOA. The Bureau does not engage with any of this evidence, other than to acknowledge but minimize the fact that the legislative history squarely confirms Congress intended disparate impact to apply.
The proposed rule posits that the use of disparate impact may also encourage “reverse discrimination”, where reducing disadvantages for one group– Black borrowers, for example– may disfavor other protected classes, such as White borrowers. However, the example above regarding grossing up non-taxable income belies the notion that somehow disparate impact disadvantages any groups. Rather, disparate impact aims to level the playing field by addressing how the financial market has unequally distributed credit access across different groups. Black applicants are 2.9 percentage points more likely than White applicants to be denied a mortgage, controlling for risk characteristics such as credit score, loan-to-value and debt-to-income ratios.[8] Identifying additional underserved Black applicants for loans does nothing to make White applicants less likely to be approved.[9] This observation is particularly true in credit markets, which are not zero-sum. Disparate impact improves creditors’ policies and leads to the removal and adjustment of practices that unnecessarily exclude applicants, without disadvantaging anyone. Moreover, the Bureau does not engage with the fact that the Inclusive Communities majority considered and rejected the argument that disparate impact raises intentional discrimination or constitutional risks, noting that the doctrine “has always been properly limited in key respects to avoid” those issues.[10]
Discrimination is not always overt and can occur due to unconscious bias, as long recognized by the Supreme Court. This can influence the design of policies and the delivery of financial products and services in ways that impact communities unnecessarily. As discussed further below, inequity continues to exist in significant ways across financial markets and products in meaningful ways that impact communities, small businesses, and individuals – thus, the need for legal tools like the disparate impact standard to reduce inequity continues to exist, and nothing in the Bureau’s proposed rule supports removing a method of proving discrimination that Congress intended to apply, that regulators have long enforced, and that benefits both affected individuals and the institutions that serve them.
II. Discouragement
The proposed rule would significantly narrow Regulation B’s coverage of pre-application conduct. For example, discouragement would be narrowed only to oral and written statements by creditors directed at applicants that creditors know would cause applicants to believe they would be denied or granted credit on less favorable terms. Statements directed at one group, encouraging that group to apply, would not be covered.
Under current Regulation B, broader acts and statements, like public communication or targeted advertising, that discourage applicants from applying in the first place are also prohibited. The Seventh Circuit Court of Appeals recognized this implicit form of discouragement in the case, Townstone vs. CFPB.[11] An employee at a Chicago-based non-bank mortgage lender publicly called a Black neighborhood in the city “Jungle Jewel”,[12] alluding to the racist stereotype that Black neighborhoods are uncivilized and violent. The company’s Home Mortgage Disclosure Act (HMDA) data also revealed that, when compared to similar financial institutions operating in the Chicago area, it received significantly fewer mortgage applicants from Black applicants and fewer mortgage applications for properties in majority-Black (more than 50% black) and high-Black (more than 80% Black) neighborhoods, circumstantial evidence that the company was acting in various ways that discouraged Black applicants from applying.[13] Other cases brought by the federal government (including the Bureau itself) have relied on similar types of evidence indicating discouragement, including placing bank branches or loan officers in certain neighborhoods and not others, developing marketing campaigns and advertisements that are targeted to certain neighborhoods and not others, or predominantly using models that appear to be white.[14]
NCRC has looked at such evidence in identifying lending institutions that limit home mortgage lending in census tracts with high percentages of minority residents.[15] It notes that, between 2010 and 2021, only 15% of 4,130 branches that large banks opened were located in majority-minority LMI neighborhoods, in contrast to 61% of the bank branch openings in predominantly white, middle- to upper-income neighborhoods.[16] The proposed rule suggests that acts such as bank branch placement would not be seen as discouragement. But a reasonable person could conclude that, among other things, choosing to put few or no bank branches or officers in low-to-moderate income (LMI) or majority-minority census tracts does discourage prospective applicants in these census tracts from applying for loans, and the proposed rule provides no basis for excluding this long-recognized and common-sense circumstantial evidence of discouragement and discrimination.
The proposed rule also ignores the trust gap that exists between financial institutions and communities of color. In their national survey, the FDIC found that one of the main reasons that unbanked/underbanked households do not have accounts is because they “don’t trust banks”.[17] Therefore, banks’ broader acts, such as branch placement or targeted advertising, make a difference in whether applicants or prospective applicants feel discouraged from applying and should be considered in addition to more explicit statements.
Rather, the Bureau’s proposal is designed to immunize this type of discriminatory conduct, severely limiting the ability of the Bureau, DOJ, other agencies, or private plaintiffs to address and remediate redlining, discriminatory advertising, discriminatory targeting, and other discriminatory pre-application conduct. Indeed, under the Bureau’s proposed rule, it would not be illegal for a creditor to hang a sign in their window saying: “Whites encouraged to apply,” because that would qualify as a “statement” directed at one group of consumers, encouraging that group of consumers to apply for credit, and could not be introduced as evidence that the bank intended to discourage other applicants.
The Bureau gives no consideration to the implications of its arguments, or the harms that discouraging conduct creates. Worse, at a time when emerging technologies such as targeted digital advertising raise serious and well-known fair lending risks,[18] the Bureau should be expanding the scope of Regulation B to ensure that such practices are not used to circumvent or evade ECOA. Instead, it is limiting the applicability of Regulation B so that it would not even prohibit the most well-recognized and explicit forms of discriminatory discouragement. This is not reasoned decision-making and the Bureau should not finalize its proposal.
III. Special Credit Purpose Programs (SPCPs)
SPCPs are offered by numerous banking institutions to make credit more accessible to socially disadvantaged groups. Congress amended ECOA in 1976 to allow for the provision of SPCPs by for-profit entities because Congress did not want ECOA to be interpreted to prevent lending programs designed to expand access to credit to minority and other underserved groups, such as minority-owned small business lending programs. The statutory text provides that it is “not discrimination” if the SPCP is a (1) “credit assistance program administered by a nonprofit organization for its members or an economically disadvantaged class of persons,” or (2) “special purpose credit program offered by a profit-making organization to meet special social needs which meets standards prescribed.”[19] In other words, SPCPs do not favor one group over another, but rather recognize long-standing lending disparities faced by economically or socially disadvantaged groups, particularly people of color and women, and encourage institutions to address those inequities.
The CFPB correctly appears to agree that the statutory text intended to permit lending programs designed account for protected class status, such as programs for minority homebuyers or small business owners. But the proposed rule would prohibit such SPCPs despite this statutory text. It would do so explicitly for certain protected class groups, and implicitly for other groups by erecting a standard for implementing an SPCP that is impossible for any creditor to satisfy. The Bureau’s proposal is therefore contrary to the plain text and intent of the SPCP statutory provision (ironically, considering its insistence on the primacy of statutory text with respect to its disparate impact proposal). The Bureau even appears to acknowledge this legal problem and invokes its authority to make “adjustments or exceptions” to the statute. But it fails to explain why that provision is relevant, particularly here, where it attempts neither to “adjust” or “except”, but rather to nullify a statutory provision entirely.
The Bureau also appears to argue that, “[r]egardless of whether instances of credit discrimination continue to occur in the marketplace,” discrimination is less of a problem today than when the Board first promulgated the SPCP provisions in Regulation B, and so SPCPs are not justified. The Bureau does not even attempt to explain why such an unworkable, undefined standard would be appropriate, even if it were true. But it clearly is not, and the Bureau makes no attempt to actually demonstrate its point. Data, reports, and analyses regarding persistent and widespread discrimination in credit are readily available—in fact, the Bureau has produced and compiled many such studies itself, including a very recent report on matched-pair testing showing discrimination adverse to Black-owned small businesses[20] and via its yearly statutorily-mandated reports to Congress about fair lending issues. Not to mention the countless studies showing ongoing disparities across lending markets and product. NCRC’s own analysis demonstrates that Black borrowers remain underrepresented at 8.9% of home purchase loans, well below their 11.7% share of the adult population.[21] Banks continue to place fewer branches in Black-majority neighborhoods, and the racial wealth divide has only grown.[22] Pricing discrimination is occurring, leading to over $8 billion in overpayment in interest.[23] In 2022, the median White household held $284,130 in wealth, more than six times that of the Black median household at $44,210.[24] There is no justification or citation to support the Bureau’s conclusion that credit markets have been “substantially reshaped” compared to 1976, even if that was the relevant standard. Given that the Bureau’s logic rests on a purported scarcity of discrimination today, it is the Bureau’s burden under the Administrative Procedures Act to justify that rationale by identifying and weighing relevant evidence, but it makes no attempt to do so.
The Bureau vaguely gestures at arguments that SPCPs may violate the Equal Protection Clause of the Constitution because they consider race and other protected class characteristics when determining eligibility. However, the Equal Protection Clause only applies to actions taken by the government, and because SPCPs are implemented by private institutions, such as banks, they cannot be deemed unlawful on those grounds. For this reason, SPCPs are governed by ECOA and Regulation B. ECOA and Regulation B, moreover, are protected-class neutral and therefore do not implicate constitutional questions.[25] And of course, these arguments are particularly unfounded with respect to protected class groups that are not afforded special constitutional protections.
NCRC also notes that the eradication of SPCPs is particularly flagrant in the small business lending context. As noted above, the Bureau itself identified patterns of potentially discriminatory treatment in the small business lending market,[26] echoing the same patterns identified by NCRC.[27] And studies show that minority small business owners face “inordinate barriers to accessing capital, are less likely to secure loans, and often face predatory payment terms at higher rates.”[28] And the proposed rule notes that the availability of HMDA data has significantly changed the market environment such that SPCPs are no longer warranted – but it does so after repeatedly delaying the publication of a rule implementing Section 1071 of the Dodd-Frank Wall Street Reform and Consumer Protection Act, which amended ECOA to collect certain demographic data on applications for credit from women-owned and minority-owned small businesses. The Bureau cannot simultaneously claim that there is no discrimination in the small business lending market while continuing to delay and drastically limit the collection and publication of data that could confirm patterns of discrimination that have been identified by its own testing, as well as the experiences of countless small business owners.
SPCPs are economically viable models for generating more revenue for for-profit institutions, as they open the credit market to a wider range of customers. Indeed, the American Bankers Association did three case studies of SPCPs specifically serving minority and small business owners and all three institutions reported that the programs were successful for both the institutions and the recipients of the funding.[29] SPCPs allow institutions to acknowledge long-standing inequities in the market and work to increase access to credit for all. Eliminating this tool serves neither customers nor institutions, and in fact costs the U.S. economy significantly each year.[30] The Bureau should not enact its proposed changes.
Conclusion
For the reasons above, we urge the CFPB to rescind the proposed rule. The proposed rule will create more barriers for underserved communities, particularly people of color, to access credit. Redlining continues to have a present-day impact and the core elements of Regulation B and ECOA ensure that all communities have a fairer shot at accessing credit, building wealth, and contributing to the economic growth of the nation.
Should you have any questions, you may reach out to Manan Shah, Policy Advisor, mshah@ncrc.org, or Eden Forsythe, Chief Policy Counsel, eforsythe@ncrc.org.
Sincerely,
Jesse Van Tol
President and CEO
National Community Reinvestment Coalition
[1] Griggs v. Duke Power Co., 401 U.S. 424, 431-32 (1971) (holding that statutory language deeming it unlawful for an employer to “limit, segregate, or classify his employees in any way which would deprive or tend to deprive any individual of employment opportunities or otherwise adversely affect his status as an employee, because of such individual’s race, color, religion, sex, or national origin” suggested that Congress sought the removal of artificial, arbitrary, and unnecessary barriers . . . when the barriers operate invidiously to discriminate on the basis of racial or other impermissible classification.”)
[2] Credit Discrimination: Hearing on HR 14856 and HR 14908 Before the H. Subcomm. On Consumer Affairs of the H. Comm. on Banking & Currency, 93 Cong. 5 (1974).
[3] See, e.g., S. Rep. No. 94-589, at 1 (1976) (noting that “judicial constructions of anti-discrimination legislation in the employment field, in cases such as [Griggs] . . . are intended to serve as guides in the application of this Act . . .”).
[4] 40 Fed. Reg. 49298, 49301 (Oct. 22, 1975) (observing that the Board believed a policy discounting income solely because it was derived from part-time employment would have “a disproportionately heavy impact on one sex” because “the majority of the part-time workforce are women”); 41 Fed. Reg. 29870, 29874 (July 20, 1976) (“The amended Act prohibits intentional discrimination and also may be interpreted as prohibiting actions that have the effect of discriminating against applicants on any prohibited basis.”); 42 Fed. Reg. 1242, 1246 (Jan. 6, 1977) (“Congress intended certain judicial decisions enunciating the effects test in the employment area to be applied in the credit area.”).
[5] Policy Statement on Discrimination in Lending, 59 Fed. Reg. 18266, 18268 (April 15, 1994).
[6] Texas Department of Housing and Community Affairs v. Inclusive Communities Project, Inc., 576 U.S. 519, 540 (2015), available at https://supreme.justia.com/cases/federal/us/576/519/
[7] 576 U.S. at 533-540.
[8] Ky, K.-E., & Lim, K. (2022). The role of race in mortgage application denials. Federal Reserve Bank of Minneapolis.https://www.minneapolisfed.org/research/community-development-working-papers/the-role-of-race-in-mortgage-application-denials
[9] See Latimore v. Citibank Fed. Sav. Bank, 151 F.3d 712, 714 (7th Cir. 1998) (“A bank does not announce, ‘We are making a $51,000 real estate loan today; please submit your applications, and we’ll choose the application that we like best and give that applicant the loan.’”).
[10] 576 U.S. at 521.
[11] Cons. Fin. Prot. Bureau v. Townstone Financial, Inc., 104 F.4th 768 (7th Cir. 2024).
[12] Id. at 772.
[13] Id.
[14] See, e.g., Cons. Fin. Prot. Bureau, et al. v. Fairway Indep. Mortg. Corp., No. 2:24-cv-01405-AMM, Compl. (Dec. 3, 2024 N.D. Ala.) (alleging, among other things, that Fairway operated its three retail loan offices and three loan production desks in majority-white areas and predominantly directed its marketing to majority-white areas); Cons. Fin. Prot. Bureau, et al. v. Trident Mortg. Co., No. 2:22-cv-02936 (July 27, 2022 E.D. Penn.) (alleging, among other things, that Trident avoided sending its loan officers to market to majority-minority neighborhoods, developed marketing campaigns and advertisements that featured primarily white models and Trident employees, and targeted its marketing materials to majority-white neighborhoods); United States of America, et al. v. Trustmark Nat’l Bank, No. 2:21-cv-02664 Compl. (Oct. 22, 2021 W.D. Tenn.) (alleging, among other things, that Trustmark avoided locating branches in and assigning loan officers to majority-Black and Hispanic communities, and that similarly situated lenders generated 2.5 times more mortgage loan applications from majority-Black and Hispanic neighborhoods in the Memphis MSA than Trustmark);
[15] See, e.g., NCRC, Redlined: KeyBank Failed Black America Despite Its Commitments to Improve (November 2022).
[16] Rocha, P. (2022, September 29). Report: Large banks opened far fewer branches in minority areas since 2010. American Banker. https://www.americanbanker.com/news/report-large-banks-odoes not considercreatespened-far-fewer-branches-in-minority-areas-since-2010
[17] Federal Deposit Insurance Corporation, 2023 FDIC National Survey of Unbanked and Underbanked Households — Executive Summary, accessed December?2,?2025, https://www.fdic.gov/household-survey/2023-fdic-national-survey-unbanked-and-underbanked-households-executive-summary
[18] Carol A. Evans, From Catalogs to Clicks: The Fair Lending Implications of Targeted, Internet Marketing, Federal Reserve Board Consumer Compliance Outlook (2019), https://www.consumercomplianceoutlook.org/2019/third-issue/from-catalogs-to-clicks-the-fair-lending-implications-of-targeted-internet-marketing/; see also NCRC, et al., Statement on Request for Guidance on Implementation of Disparate Impact Rules Under ECOA (June 29, 2021).
[19] Consumer Financial Protection Bureau, “§?1002.8?Special purpose credit programs,” Regulation?B, accessed December?2,?2025, https://www.consumerfinance.gov/rules-policy/regulations/1002/8/
[20] CFPB, Matched-Pair Testing in Small Business Lending Maproductsrkets (Nov. 13, 2024), https://www.consumerfinance.gov/data-research/research-reports/matched-pair-testing-in-small-business-lending-markets/
[21] National Community Reinvestment Coalition, Mortgage Market Report Series — Part?2: Lending Trends by Borrower and Neighborhood Characteristics, accessed December?2,?2025, https://ncrc.org/mortgage-market-report-series-part-2-lending-trends-by-borrower-and-neighborhood-characteristics
[22] Broady, K., McComas, M., & Ouazad, A. (2021, November 2). An analysis of financial institutions in Black?majority communities: Black borrowers and depositors face considerable challenges in accessing banking services. Brookings Institution. https://www.brookings.edu/articles/an-analysis-of-financial-institutions-in-black-majority-communities-black-borrowers-and-depositors-face-considerable-challenges-in-accessing-banking-services/
[23] Bradford, Wl,, et al., (Mis)Pricing in Loans to Businesses Owned by People of Color, IZA Discussion Paper No. 17730 (Feb. 2025).
[24]Dean, J. (2024, October). The racial wealth gap,?1992 to 2022. National Community Reinvestment Coalition. https://ncrc.org/the-racial-wealth-gap-1992-to-2022/
[25] See Rothe Dev., Inc. v. U.S. Dep’t of Def., 836 F.3d 57, 62 (D.C. Cir. 2016) (“Section 8(a) uses facially race-neutral terms of eligibility to identify individual victims of discrimination, prejudice, or bias, without presuming that members of certain racial, ethnic, or cultural groups qualify as such. That makes it different from other statutes that either expressly limit participation in contracting programs to racial or ethnic minorities or specifically direct third parties to presume that members of certain racial or ethnic groups, or minorities generally, are eligible.”).
[26] CFPB, Matched-Pair Testing in Small Business Lending Markets (Nov. 13, 2024), https://www.consumerfinance.gov/data-research/research-reports/matched-pair-testing-in-small-business-lending-markets/
[27] See, e.g., NCRC, Does Perceived Socio-Economic Status Impact Access To Credit In The Smal Business Arena? (Dec. 2021); NCRC, Racial and Gender Mystery Shopping for Entrepreneurial Loans; Preliminary Overview; NCRC, Lending Discrimination within the Paycheck Protection Program (July 2020)
[28] Goldman Sachs, New Survey Data Shows Black Small Business Owners Less Likely to Secure Loans (Feb. 6, 2024), available at https://www.goldmansachs.com/pressroom/press-releases/2024/new-survey-data-shows-black-small-business-owners-less-likely-to-secure-loans; see also, e.g., Federal Reserve Bank of Cleveland, “It shouldn’t be this hard”: Small Business Credit Survey highlights funding challenges for startups owned by people of color (July 10, 2023), available at https://www.fedsmallbusiness.org/analysis/2023/it-shouldnt-be-this-hard-small-business-credit-survey-highlights-funding-challenges-for-startup; Federal Reserve Bank of Atlanta, “Minority Firms Have Harder Time Obtaining Bank Financing, Fed Analysis Finds (Jan. 28, 2020), available at https://www.atlantafed.org/economy-matters/community-and-economic-development/2020/01/28/minority-firms-have-harder-time-obtaining-bank-financing-fed-analysis-finds
[29] American Bankers Association, Case Studies, available at https://www.aba.com/banking-topics/commercial-banking/small-business/special-purpose-credit-programs/case-study
[30] McKinsey Institute for Economic Mobility, The economic impact of closing the racial wealth gap (August 13, 2019), available at https://www.mckinsey.com/industries/public-sector/our-insights/the-economic-impact-of-closing-the-racial-wealth-gap
