NCRC Comments on FHFA’s Proposed Affordable Housing Goals 2026-2028

November 3, 2025

Mr. Clinton Jones
General Counsel
Federal Housing Finance Agency
400 7^th Street SW
Washington, DC 20219

Re: FHFA Proposed Affordable Housing Goals 2026-2028, RIN 2590–AB59

Dear Mr. Jones,

The National Community Reinvestment Coalition (NCRC) opposes the proposal of the Federal Housing Fnance Agency (FHFA) to lower the affordable housing goals for the Government Sponsored Enterprises (GSEs). By requiring the GSEs (Fannie Mae and Freddie Mac) to purchase specified percentages of loans made to communities of color and low- and moderate-income (LMI) borrowers and communities, the goals also help banks and other lenders achieve their fair lending requirements of serving borrowers equitably and fairly.

A precipitous drop in the goal requirements would likely result in significantly fewer home loans and wealth building opportunities for traditionally underserved borrowers that have experienced redlining and other forms of discrimination over the decades. This would exacerbate income and wealth inequalities, making our nation less prosperous and more divided.

The proposed FHFA lowering of the affordable housing goals is inconsistent with mission of the GSEs. The statutory mission of the GSEs is “an affirmative obligation to facilitate the financing of affordable housing for low- and moderate-income families in a manner consistent with their overall public purposes, while maintaining a strong financial condition and a reasonable economic return.”[1] When implemented rigorously, the goals are a major tool for helping the GSEs achieve their mission of serving low- and moderate-income families in a safe and sound manner.

By law, the FHFA is required to consider housing needs, demographic conditions, the safety and soundness of the GSEs, and the GSE’s ability to “lead the industry in making mortgage credit available.”[2] While housing needs remain pressing for low- and moderate-income borrowers, the FHFA proposes to reduce the GSEs from leaders in providing financing for LMI borrowers to followers by setting goals below the levels of primary market lending to LMI borrowers and communities. Moreover, as demonstrated in this letter, this is not justified by adverse economic conditions nor risks to the safety and soundness of the GSEs.

One example illustrates the inequitable nature of the FHFA’s proposals. The low-income home purchase goal specified that from 2025 through 2027, the percentage of GSE purchases that must be for loans made to LMI borrowers with incomes at or below 80% of area median income was 25%. The FHFA proposes to lower this to 21% for the years 2026 through 2028. This is a diminishment of 16% and would likely result in hundreds of thousands of fewer loans for LMI borrowers financed by the GSEs. As described below, the GSEs could possibly reduce their purchases of home purchase loans for LMI borrowers by about 368,000 loans. This does not mean that LMI borrowers will receive 368,000 fewer loans over the three-year term of the goal. However, since the GSEs make up almost half of the secondary market, it is possible that the reduced purchasing activity of the GSEs for LMI loans could significantly reduce the financing and hence the number of loans received by these borrowers.

The letter is organized as follows:

• A demonstration of how the FHFA’s economic theory and analysis of economic conditions justifying FHFA goal lowering is faulty
• The FHFA’s assessment of equity considerations for their proposal would only exacerbate inequality
• A detailed examination of the flaws and impacts of FHFA’s reduced goals

The FHFA’s Crowding Out Economic Argument Does Not Justify Lowering Goals

A main pillar of the FHFA’s justification for lowering the affordable housing goals is the crowding out theory. Under this theory, overly aggressive housing goals do not expand the number of loans for LMI borrowers but mainly result in the GSEs and other secondary market entities reducing their fees and other charges assessed on sellers of loans. This in turn imperils the financial condition of the secondary market entities according to FHFA.

Even if the sellers of loans (banks, mortgage companies and other lenders) then reduce fees and interest rates on their loans for borrowers in response to the FHFA charging less to buy loans, these reductions in price do not increase the demand for loans by LMI borrowers. The main impact of the price competition among the secondary market entities would be to reduce the revenues received by them and thus increase their risk and decrease their safety and soundness according to the FHFA. Also, since the GSEs have inherent advantages due to their backing by the federal government, aggressive purchasing of LMI loans by the GSEs would mainly reduce the market share of other secondary market players and “crowd” them out.

However, rigorous goals would not crowd out other secondary market entities including the Federal Housing Administration (FHA) or the Private Label Securities (PLS) market. The crowding out theory assumes that the supply or number of loans for low- and moderate-income borrowers is finite. In other words, neither public sector policy nor private market developments would effectively increase the number of loans issued to low- and moderate-income borrowers.

The FHFA’s application of the crowding out theory ignores the interaction between the supply and demand for loans. Standard economic theory suggests that as the price of a good declines (interest rates and fees in the case of loans), the demand for a good (loans) will increase. Indeed, over the usual thirty years for a loan term, an interest rate reduction of a quarter or half of a percentage point would save borrowers tens of thousands of dollars. Since an interest rate reduction of this magnitude critically impacts the affordability of loans, it would make the difference for thousands of low- and moderate-income (LMI) borrowers in their abilities to afford and demand loans. Moreover, the demand for loans by LMI borrowers is likely to be elastic, meaning that they are sensitive to price changes and would significantly increase their demand for loans if interest rates decrease by what may appear to be seemingly small amounts. A paper published by the Joint Center for Housing Studies found that “a decrease in interest rates by 25 basis points (one quarter of a percentage point) results in an increase in the propensity to obtain a mortgage of about 50 percent.”[3] In other words, borrowers are considerably more likely to apply for mortgages when interest rates decline by even less than a percentage point.

The FHFA maintains that since stagnant LMI loan demand cannot be expanded through favorable pricing, rigorous GSE competition for loans against other secondary market participants only results in inefficient outcomes which mainly reduce revenues received by all secondary market entities, placing them in riskier positions with thinner capital reserves.[4] This assertion, however, rests on the erroneous assumption that LM demand for loans is immutable and cannot be increased by drops in interest rates and fees. Moreover, the FHFA maintains that GSEs have become adept at gobbling up the least risky LMI borrowers. In order for the FHA and other secondary market entities to become healthier, the GSEs should reduce their share of LMI loans so the other market entities can also purchase less risky LMI loans. The FHFA believes it can achieve lower GSE market share by lowering the affordable housing goals.[5]

The theme of ruinous competition is an oxymoron. Generally, price competition among providers of goods benefits consumers and borrowers through lower prices and interest rates. The market data also does not suggest that price competition has reduced safety and soundness or capital reserves. The Office of the Comptroller of the Currency (OCC) reported recently that 97.5% of mortgages were current and performing at the end of the second quarter of 2025, an improvement from a year earlier. Likewise, the percentage of seriously delinquent mortgages (payments 60 days overdue) also decreased from a year ago.[6] The Consumer Financial Protection Bureau (CFPB) found that rate of seriously delinquent mortgages over 90 days late has been less than 1% since 2020. These overall performance trends do not coincide with an Affordable Housing Goal regime that has been too aggressive.[7]

The FHFA indirectly agrees in its proposed rule that decreases in safety and soundness is not an issue. The FHFA states, “low- and moderate-income households represent a significant (and still profitable) proportion of the overall potential homebuying market.”[8]

Lastly, the data on market shares presented by the FHFA does not support the crowding out theory because the GSE market share has been declining noticeably over the last several years. The FHFA shows that the GSE market share decreased from 67% of all mortgages in 2020 to 47% in 2024.[9]

Economic Conditions Do Not Necessitate a Decrease in Housing Goals

In contrast to the FHFA’s portrayal, economic conditions support maintaining the existing goal levels instead of reducing them. The proposed rule reports that the National Association of Realtors (NAR) predicts that housing affordability and the ability to purchase homes for families at median income levels will improve from 2027 through 2028.[10] Moreover, after plummeting during the pandemic, housing supply has increased in recent years with active listings in spring of 2025 70% higher than they were in the spring of 2023 according to the proposed rule.[11]

Despite overall improvements in the housing affordability and housing supply, the proposed rule suggests that increased competition by millennials will continue to present difficulties for modest income households in gaining homeownership. Yet, the proposed rule does not present convincing quantitative data to back up this assertion. In fact, the NAR report the proposed rule cites has a chart showing that 26% of the homebuyers in 2022 had income levels up to $55,000, which corresponds to low- and moderate-income levels in several metropolitan areas. Twenty-six percent corresponds to the current low-income purchase goal and is considerably above the 21% proposed income goal. It seems that the proposed rule has not carefully considered all the empirical evidence in lowering the affordable housing goals.[12]

The FHFA’s Equity Assessment Would Result in Less Equity

As discussed above, the statutory mission of the GSEs is to facilitate financing for low- and moderate-income households. The affordable housing goals are a major means for helping the GSEs achieve this mission. Yet, a key FHFA justification for lowering the goals is its concern that middle-income borrowers have been inadvertently penalized by the goals. In pursuit of the goals, the GSEs reduce fees assessed on the sellers to the GSEs of loans that qualify for the goals. In this process, the FHFA suggests that loan fees and interest rates have been lowered for LMI borrowers but significantly increased for middle-income borrowers.

Throughout the proposed rule, the FHFA expresses its concern for the plight of middle-income borrowers. Here are a couple of passages:

FHFA, in carrying out this policy priority, is assessing the impact of the housing goals on the cost of housing, particularly to middle-class borrowers, who may be turned away or receive higher prices than they would in the absence of overly aggressive housing goals.[13]

Of particular concern is feedback from lenders that they have turned away middle-class borrowers or increased prices on middle-class borrowers in pursuit of meeting housing goals. These concerns also suggest that the benchmarks for the housing goals have been set too high.[14]

It would almost seem from these passages that mission of the GSEs is to finance housing for middle-income borrowers and that the affordable housing goals have thwarted this mission. In contrast, the law requires a GSE focus on LMI borrowers. This was likely the case not because of any desire to penalize middle-income borrowers but because the history of redlining and other barriers to homeownership experienced by people of color and LMI borrowers.

Moreover, the FHFA states that it “recognizes that it is difficult to fully isolate and quantify the extent to which setting inappropriately aggressive housing goals penalizes middle-class borrowers.”[15] Thus, it has crafted a proposed rule with hypothetical harms to the middle-class in mind, but it admits it lacks hard evidence of those harms. The inequitable result is likely to be harm imposed on LMI borrowers and a lack of alleviation of any harm currently experienced by middle-income borrowers.

NCRC’s data analysis indicates that middle-income borrowers have not experienced a significant decline in lending as a result of unfavorable pricing. From 2022 through 2024, middle-income borrowers have received 28% of all home purchase loans. LMI borrowers have received declining shares of home purchase loans from 29% in 2022 to 25% in 2024. If any group of borrowers have benefited from the market, it is upper-income borrowers whose share of loans went from 44% in 2022 to 47% in 2024.[16]

Cross subsidization is a common business practice across industries including the home loan industry. Contrary to the FHFA’s assertion, cross subsidization to help LMI borrowers does not seem to have disproportionately burdened middle-class borrowers. In contrast if the FHFA were to eliminate the cross subsidization made possible by waiving upfront fees for first time homebuyers, homebuyers with incomes below 100% area median income, and for HomeReady and Home Possible loans, LMI borrowers would experience significant cost increases and would demand fewer loans.[17]

We encourage the FHFA to engage in a historical analysis examining disadvantages in the lending marketplace and maintain the standards of the last few years focusing on LMI and communities of color. Extensive research by NCRC, academics, and federal agencies have documented that redlining originated from a collaboration among the public sector, the lending industry, and real estate professionals. Starting in the Roosevelt administration during the 1930s, maps commissioned by federal agencies and developed in part by appraisers and other private sector professionals color coded neighborhoods across hundreds of cities by perceived lending risk. The neighborhoods color coded in red as most risky for lending were usually those inhabited predominantly by African Americans and recent immigrants from Eastern Europe. Decades later these neighborhoods continue to experience less access to lending, high rates of poverty and unemployment, and poor health outcomes.

Redlining combined with multiple other forms of discrimination across industries and sectors to compound economic distress. Exclusionary zoning laws mandating single family homes, large lot sizes, and other unaffordable features restricted the ability of modest income residents of redlined neighborhoods to find decent housing in other neighborhoods with improved economic opportunities. As a result, people of color and recent immigrants were confined to neighborhoods with deteriorating and unhealthy housing. Labor market discrimination depressed wages of residents of redlined neighborhoods and segregation in schools denied educational opportunities. Decades later, the subprime crisis led to another wave of predatory lending and wealth extraction in communities of color practiced by unregulated mortgage companies largely financed by Wall Street firms.[18]

Federal agency redlining maps digitized by the University of Richmond illustrate the focus of redlining on working class immigrant neighborhoods and African American communities in the 1930s. In Milwaukee, the Haymarket neighborhood in the eastern part of the city near Lake Michigan was a neighborhood inhabited by African Americans and Eastern European immigrants that worked in local industry. The map of Milwaukee had accompanying narrative that described the Haymarket neighborhood in these terms: “This is the Negro and slum area of Milwaukee. It is old and very ragged. Besides the colored people, a large number of lower type Jews are moving into the section.”[19]

Today, almost 75% of Haymarket residents are people of color, the poverty rate is 40%, 39% of the residents have high blood pressure, and the life expectancy is 72.9 years while the highest is 88.5 years in the neighborhoods with better socioeconomic conditions.[20] Redlining and a myriad of other discriminatory practices ranging from exclusionary zoning, racial covenants, and predatory lending targeting areas with a dearth of mainstream banks have produced disadvantages that still afflict too many neighborhoods across the country.

In addition, working-class white Americans as well as people of color suffered from exclusionary practices. In his comprehensive book, Excluded: How Snob Zoning, Nimbyism, and Class Bias Build the Walls We Don’t See, Richard Kahlenberg describes in detail exclusionary zoning practices that not only block people of color from thriving communities but also present stiff barriers for LMI families and households. Income segregation (or the separation of modest income from affluent Americans) significantly increased over the decades from 1970 through 2010. According to a study Kahlenberg cites, the percentage of Americans living in rich or poor neighborhoods increased from 15% in 1970 to 34% in 2012.[21]

Towards the end of the civil rights era, Congress focused on improving economic prospects and enacted laws outlawing discrimination in housing and lending aimed at communities of color, LMI neighborhoods, and other disadvantaged populations. These laws included the Fair Housing Act of 1968, the Equal Credit Opportunity Act of 1974, the Home Mortgage Disclosure Act of 1975, and the Community Reinvestment Act of 1977. The laws not only outlawed discrimination but monitored the fairness of lending via public data disclosure and imposed an affirmative obligation on financial institutions to serve traditionally underserved communities.

Congress was responding to the vicious discrimination suffered by communities of color and working-class neighborhoods as documented over the decades starting with the federal agency redlining maps in the 1930s. While Congress sought to also extend consumer protections to middle-income Americans, this group was not the focus of fair lending laws because they had largely benefited from the American Dream of homeownership financed by the Federal Housing Administration and the GI Bill. It was African Americans, other people of color, poor immigrants, and working-class Americans that were largely left out of the American Dream whose plight was addressed by the fair lending laws.

When Congress was considering the Community Reinvestment Act in 1977, the late Senator William Proxmire, the legislative father of CRA, defined redlining as:

banks and savings and loans will take their deposits from a community and instead of reinvesting them in that community, they will actually or figuratively draw a red line on a map around the areas of their city, sometimes in the inner city, sometimes in the older neighborhoods, sometimes ethnic and sometimes black, but often encompassing a great area of their neighborhood.[22]

This quote reinforces the focus on communities of color and white ethnic working-class neighborhoods. Years later the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 maintained this emphasis when it established the affordable housing goals. Testifying before Congress, Thomas J. McCool Director, Financial Institutions and Markets Issues of the Government Accountability Office maintained:

As part of the 1992 Act, Congress concluded that the financial benefits that the enterprises derive from their government sponsorship involve a corresponding obligation to meet the mortgage credit needs of all potential home buyers, including those with low- and moderate-incomes.[23]

In passing the 1992 Act, Congress concluded that HUD’s regulatory framework had not been effective in ensuring that the enterprises’ activities benefit low- and moderate-income Americans and those who live in underserved areas, such as central cities and rural communities (targeted groups). The 1992 Act required HUD to develop, implement, and enforce a comprehensive housing mission regulatory framework.  Among other provisions, the 1992 Act directed HUD to set housing goals, which require the enterprises to meet specified criteria each year for the purchase of mortgages serving targeted groups.[24]

The fair lending laws including the Safety and Soundness Act of 1992 directed federal agencies to correct market failure and eliminate the deleterious impacts of discrimination victimizing communities of color and working-class neighborhoods. These discriminatory barriers to the market did not victimize Americans and the middle class broadly but were focused on subgroups in the population. The pressure of public sector oversight and enforcement of affirmative obligations to serve communities was focused on victims of discrimination and did not need to be broadly extended since middle- and upper-income Americans benefited from access to housing and lending markets. The proposed rule’s distortion of the Affordable Housing Goals and diminishment of their focus on LMI borrowers and communities is contrary to their statutory purpose and is not an efficient use of resources for addressing priority shortcomings in the housing and lending markets.

Unjustified Reduction in Single Family Purchase Goals Could Dramatically Reduce Home Mortgage Lending for LMI Borrowers and Neighborhoods and Borrowers in Communities of Color

The inequitable nature of the proposed affordable housing goals is dramatically illustrated by a goal setting that ignores GSE abilities to lead the market and their recent performance. Moreover, the loss of lending is likely to be significant and substantially reduce economic development and wealth building in formerly redlined neighborhoods.

Table 1 below in the appendix uses data presented by the FHFA and displays the performance of the primary market (banks, mortgage companies, and credit unions) compared to the that of the GSEs during the years of 2019 through 2024.[25] Under the single-family low-income purchase goal, the five-year average for the primary market in terms of the percentage of loans made to low-income borrowers with incomes up to 80% of area median income was 26.8% (the FHFA did not have primary market data for 2024). The six-year average for Fannie Mae and Freddie Mac over those years was almost identical at about 28% for each of them. In other words, over a multiple year period, the GSEs lead the market by about one percentage point in financing loans for LMI borrowers. Despite their clear ability to lead the market, the FHFA now wants to reduce the goal they need to reach to 21%. Since this is a substantial decrease from the current 25% goal, it is likely that the GSEs will relax their efforts and the percentage of mortgages they finance for low-income borrowers pursuing the American dream of homeownership.

Table 2 below in the appendix shows the possible reduction in the numbers of mortgages financed by the GSEs under the low-income purchase goal.[26] The top series of rows for the low-income purchase goal shows a four-year average of the number of purchases for each GSE from 2021 through 2024 and the percentage of mortgages for low-income borrowers. Fannie Mae’s four-year average of purchases of mortgages for low-income borrowers was 258,264 or 27.5% of their total purchases. If Fannie Mae reduced its percentage of mortgages to this income group to 21% or the goal proposed by the FHFA, this would result an annual average purchase level of just 197,350 mortgages or 60,914 fewer mortgages yearly. Over the three-year term of the proposed goal, the total reduction could be as high as 182,741. For Freddie Mac, the reduction could be as high as 185,509 loans. Combining the loss for both GSEs could possibly equal 368,250 loans.

Admittedly, this is an upper bound estimate because the GSEs may not reduce their percentages of loans purchased to the proposed goal level. However, it seems that the FHFA desires a significant reduction based on their concern for middle-income borrowers. The upper bound estimate is not inconsistent with the FHFA’s objectives and starkly illustrates that access to mortgages for low-income borrowers could decrease substantially. Other secondary market players could pick up some of the slack, but it is probable that a substantial number of low-income mortgages would not be financed by the secondary market. Lenders, for their part, are unlikely or unable to keep all the loans not purchased by the secondary market in their portfolios. Hence the number of mortgages for low-income borrowers could decline markedly over the next three years.[27]

The middle rows of Table 1 reveals that for the very low-income purchase goal, the FHFA goal reduction is also not necessary. In this case, the multiple year average for the GSEs equaled that of the primary market at 6.7% of mortgages for households with up to 50% of area median income. The FHFA proposed goal would be 3.5%, about half the current goal and a disservice for very low-income households. Moreover, Table 2’s calculation reveals that the difference between the GSE’s multi-year average of loan purchases for very low-income borrowers and those under the three-year term of the proposed goal could be upwards of 174,000 fewer loans.[28] Again, this does not mean that very low-income borrowers would experience a decline of that number of loans over the next three years. However, it is likely that they could see thousands if not tens of thousands fewer loans, frustrating their attempts of upward economic mobility.

The lower series of rows on Table 1 reveal the performance of the market and the GSEs on the low-income areas goal compared to the FHFA’s proposed goal. This goal combines lending in low-income census tracts and in predominantly minority tracts. Over a multiple year period, lenders issued 24.1% of their loans in these tracts while Fannie Mae’s and Freddie Mac’s purchases in these tracts were 25.9% and 24.8%, respectively, of their total purchases. Despite both GSEs being modestly ahead of the market, the FHFA proposes to drastically reduce the goal to 16% for the years 2026-2028. The bottom rows of Table 2 show a reduction in total loans purchased by the GSEs of upwards of 286,361 loans.[29] While this is an upper bound estimate, it is likely that the amount of financing for homeownership in formerly redlined communities struggling to revitalize could decrease significantly, threatening to increase wealth inequalities in our country.

FHFA set its proposed goal for low-income areas considerably lower than market and GSEs level of financing out of a concern that most loans financed by the GSEs in these tracts were for upper-income borrowers and that a higher goal level may promote gentrification and displacement. Interestingly, FHFA suggests that gentrification is not an overwhelming concern and cites NCRC’s study that just 15% of census tracts in urban areas experienced gentrification over a fifty-year period.[30] Nevertheless, the agency decided to lower its requirements for GSE service in these neighborhoods. Instead of dramatic declines in the proposed goal, the FHFA could have proposed techniques to ameliorate any displacement associated with gentrification since such techniques would be feasible given the occurrence of modest gentrification over the last half century. For example, median rent or home price data could have been used to identify LMI census tracts with rapid price increases, high probabilities of gentrification, and speculative and displacement-prone lending. After identifying this subset of tracts, the FHFA could prohibit any loans made to upper-income borrowers in these tracts from counting towards the goals.

The FHFA emphasizes concerns about procedures such as these being burdensome throughout the proposed rule, but experience under the Community Reinvestment Act (CRA) suggests that this type of census tract categorization is feasible and useful. Each year, the Federal Financial Institutions Examination Council (FFIEC) releases a set of underserved and distressed tracts in rural counties in which banks can finance community development activities and receive favorable consideration on their CRA exams.[31]

FHFA Must not Combine the Low-Income and Minority Tract Goals

The proposed low-income areas home purchase goal combines two previous goals, the minority census tract goal and the low-income census tract goal. The FHFA states that it “is proposing to restore the single low-income areas subgoal to simplify the regulatory framework, improve operational clarity for the regulated Enterprises, and better align subgoals with existing borrower-based metrics.”[32] However, the lending challenges in low-income and minority tracts are different, and combining the two subcategories into one would reduce the GSEs’ incentives to address needs and conditions unique to each category of tracts. For example, some low-income tracts may face gentrification pressures, which should prompt the GSEs to work with lenders and public agencies to develop products and programs that finance affordable rental and homeownership housing in LMI tracts undergoing market changes. In contrast, several minority neighborhoods continue to confront the legacy of discrimination and experience low levels of lending. A NCRC report had identified underserved tracts (the quintile of tracts across the country with the lowest lending levels) based on low levels of lending on a per capita basis; 57 percent of the residents of these underserved tracts, on average, were people of color.[33] Products and programs designed to serve these tracs would likely differ from those geared towards preventing displacement in low-income tracts.

Another difficulty with combining the two current area categories is that the GSEs may emphasize purchasing loans issued in middle-income minority tracts to the detriment of LMI tracts. As the table immediately below shows, at least 31% of the tracts that are majority minority are middle-income. Since housing and economic conditions are likely to be more favorable in the middle-income tracts as compared to the LMI tracts, the GSEs may embark upon a disproportionate amount of purchasing activity in the middle-income tracts, leaving the LMI tracts that are predominantly minority as well as non-minority with less financing than if the two goals remained separate.