Summary

According to the 2020 census, approximately 66.3 million people—about 20 percent of the U.S. population—live in rural areas. Households in those areas differ in many ways—including income and education levels, racial and ethnic characteristics, and homeownership rates—from households in more urban areas. They also face different barriers to purchasing a home or finding a rental property. The federal government administers federal credit programs that specifically support homeowners and renters in rural communities, most of which are administered by the Department of Agriculture (USDA) through its Rural Housing Service (RHS). Other agencies also contribute to that assistance.

In this report, the Congressional Budget Office examines the federal credit programs designed to address the housing challenges faced by rural residents. First, CBO provides an overview of household and housing characteristics of rural communities; second, the agency describes the federal credit assistance provided to those communities by the federal government; and finally, CBO estimates the total budgetary effects of that assistance. (CBO identified rural communities using the Census Bureau’s block classification system.)

CBO estimates that in 2026, the federal government will provide $293 billion in credit assistance to rural households. RHS provides $27 billion of that assistance and non-RHS programs provide the remaining $266 billion.

CBO calculated the lifetime budgetary effects of federal credit assistance using two methods: one based on specifications in the Federal Credit Reform Act of 1990 (FCRA) and one based on a measure of fair value (that is, accounting for market risk). Using those two methods, CBO estimates that the lifetime budgetary cost of RHS’s assistance will be $0.2 billion or $0.5 billion, respectively. The lifetime budgetary effect of assistance provided through non-RHS programs will be either a savings of $3.1 billion (calculated using FCRA procedures) or a cost of $2.8 billion (calculated using a fair-value measure). The differences between the two sets of estimates—which are based on the same projected cash flows—highlight the effect of incorporating market risk into the costs of federal credit programs.1

Identifying Rural Areas in the United States

According to the 2020 census, approximately 66.3 million people, or 29.3 million households, live in rural areas. The population of rural areas accounts for about 20 percent of the U.S. population (excluding Puerto Rico and other island territories).2 By contrast, according to a report published by the Census Bureau, rural areas accounted for about 97 percent of the total land area of the United States in 2016 (see Figure 1).3

The Census Bureau determines whether a geographic area is urban or rural on the basis of either population density or housing density.4 To qualify as urban, a census block (a statistical area bounded by visible features—such as roads, railroads, and streams—and nonvisible boundaries—such as property lines, school district limits, and municipal borders) must contain at least 5,000 people or 2,000 housing units. Any census block that does not meet that threshold is considered rural.5

Counties (the basis for analysis in this report) contain more than one census block, though, so most include both rural and urban populations. Of the 3,144 counties in the United States, 1,116 counties are considered entirely rural and 52 counties are considered entirely urban.6 The remaining 1,976 counties contain both urban and rural census blocks.

Characteristics of Rural Counties

CBO examined demographic, housing, and economic characteristics of rural households. The agency grouped counties according to the share of each county’s population that is considered rural (that is, that lives in census blocks that fall below the Census Bureau’s urban threshold). The values (percentage, rate, share, or median annual household income) in this report are medians. A median value represents the midpoint of a range of values; that is, half the values are higher, and half are lower. The values are county-level medians. For each characteristic, CBO ordered counties in each group by the value of that characteristic and then determined the median of those values. Using medians (rather than averages) mitigates the outsized effects that a small number of counties with a large number of residents would have on those values. CBO used data from the American Community Survey’s (ACS’s) five-year estimates for the 2019–2023 period to determine the characteristics of rural counties.7

Demographic Characteristics

Residents in counties that include a higher share of rural population vary significantly in terms of income, racial and ethnic diversity, and education from residents in more urban areas. Compared with their urban counterparts, households in counties with a higher share of rural residents generally have lower median income, are less racially and ethnically diverse, and often have less education.

For example, the median annual household income of predominantly urban counties (those in which rural residents make up 20 percent or less of the population) is approximately $77,000. (The median annual household income of half of those counties is lower than $77,000, and in half it is higher.) In predominantly rural counties (those in which more than 80 percent of residents live in rural areas), the median annual household income is about $60,000. Additionally, the share of households with a White head of household is 71 percent in predominantly urban counties. By contrast, in predominantly rural counties, that share is 92 percent. In predominantly urban counties, the share of occupied properties headed by a person with a college degree is 38 percent. In predominantly rural areas, that share is 20 percent.

Rates of Homeownership

Counties with a higher share of rural residents tend to have higher homeownership rates (see Figure 2). In predominantly urban counties, the homeownership rate is approximately 65 percent; in predominantly rural counties, that rate increases to about 78 percent.

Types of Residence

The types of housing units also vary markedly depending on how rural the county is. In predominantly urban counties, the share of households residing in single-family detached homes is about 66 percent—that is, in half of those counties, fewer than 66 percent of households live in single-family detached homes, and in half of those counties, more than 66 percent do. In predominantly urban counties, the share of households living in multifamily buildings with ten or more units is 11 percent. By contrast, in predominantly rural counties, the share of households residing in single-family detached homes is 78 percent. The share living in large multifamily structures is only 1 percent for households in those counties (see Figure 3).

In both predominantly urban and predominantly rural counties, the share of homeowners living in single-family detached homes is 86 percent. Therefore, most of the difference in the type of housing stems from the fact that in rural areas, renters are more likely to live in single-family homes and less likely to live in large multifamily buildings than renters in urban areas are. In predominantly urban counties, the share of renters living in detached single-family homes is 27 percent; that share rises to 53 percent for renters in predominantly rural counties. The difference in the prevalence of renters living in multifamily buildings with 10 or more units is similarly large: In predominantly urban areas, the share is 31 percent, whereas that share is 6 percent for renters in predominantly rural areas. Additionally, the share of households (both homeowners and renters) in predominantly rural counties occupying other types of housing—which consist of boats, mobile homes, and trailers—is 13 percent. In predominantly urban counties, that share drops to only 3 percent.

Housing units in rural counties also tend to be larger. In predominantly urban counties, the share of households living in one-bedroom units is 9 percent. By contrast, in predominantly rural counties, that share is 6 percent. That trend may be driven by household composition, housing affordability, and the availability of land for the development of larger properties in rural areas.

Housing Costs

Households in counties with higher shares of rural population tend to have lower median income. For homeowners, the median annual household income in predominantly urban areas is about 46 percent higher than it is in predominantly rural counties. For renters, the difference is about 36 percent. Those lower incomes are accompanied by correspondingly lower housing costs for both homeowners and renters (see Figure 4). Monthly housing costs are generally consistent between owners and renters regardless of the share of rural population in the county; the median housing cost for owners is about 7 percent more per month than renters in predominantly urban counties and about 10 percent less in predominantly rural counties.

Cost Burden of Housing

The share of income—or cost burden—spent on housing differs depending on whether the household owns or rents (see Figure 5). Despite lower housing costs in rural areas, the proportion of homeowner households spending more than 30 percent or more than 50 percent of their income on mortgage payments is relatively consistent across counties with different rural population shares. That suggests that although housing prices are lower in rural areas, income is proportionately lower as well.

By contrast, the cost burden for renters varies by a county’s share of population that is rural; in general, the more rural the county, the smaller the share of their income renters spend on rent. In predominantly urban counties, the share of renters spending more than 30 percent of their income on housing costs is 24 percent; and 12 percent of renters spend more than 50 percent. In predominantly rural counties, the share of renter households spending more than 30 percent of their income on rent is only 8 percent and just 4 percent for renters spending more than 50 percent. The lower cost burden that rural renters bear may be attributable to some combination of lower market rents, greater availability of subsidized rental options, or differences in household composition.

Mortgages

In predominantly urban counties, the share of homeowners with a mortgage is nearly 65 percent. By contrast, in predominantly rural counties, that share drops to around 42 percent (see Figure 6). That pattern may reflect several factors, including lower home values that reduce the need for financing and a higher incidence of types of homes that tend to be mortgage-free: those occupied by older residents, inherited properties, or long-held residences.

Home Values

Home values also differ significantly between urban and rural counties. In predominantly urban counties, the share of homeowner households with a mortgage living in homes valued at more than $300,000 is 45 percent (see Figure 7). In addition, the share of those households living in homes valued at more than twice their annual household income is 67 percent (see Figure 8). Conversely, in predominantly rural counties, most homeowners live in lower-valued properties. The share of homeowners with a mortgage in those counties living in homes valued at less than $300,000 is approximately 76 percent. Moreover, the share of those homeowners living in homes worth less than twice their annual household income is 48 percent.

Those figures suggest that housing costs more and appreciates faster in urban areas, both on an absolute scale and when measured against household earnings. They also indicate that homes in rural counties are generally more affordable and that households in those areas are less likely to face high debt-to-income ratios associated with mortgage borrowing.

Federal Credit Programs That Support Homeowners in Rural Areas

Rural households may face barriers to obtaining conventional credit, such as lower incomes and limited access to mortgage lenders, that can make buying a home challenging.8 The Housing Act of 1949 (Public Law 81-171) established a federal framework to support housing development and homeownership in rural areas to address those challenges. Title V of the law authorized the USDA to administer programs through RHS to eligible rural households. Those programs provide direct loans, loan guarantees, and other forms of financial assistance.

Section 502 Guaranteed Loan Program

The Section 502 Guaranteed Loan Program is one of RHS’s principal tools for expanding access to mortgage credit in rural areas. The program guarantees loans made by private lenders to borrowers with low or moderate income to purchase, build, or improve a primary residence in an eligible rural area. The program permits 100 percent financing, which eliminates the need for a down payment.

RHS provides lenders with a guarantee against credit losses on up to 90 percent of the outstanding loan balance. In return, RHS collects an upfront guarantee fee at loan origination and an annual fee for the life of the loan. The guarantee reduces lender exposure to default risk, which in turn lowers borrowing costs and expands access to mortgage credit for qualified rural households.

Other RHS Programs

RHS also manages a collection of smaller programs designed to address specific needs in rural housing markets. Those programs include the following:

• Section 504 housing repair loans support homeowners with very low income as well as those age 62 or older by providing financial assistance to repair their homes and remove health and safety hazards.
• Section 514 farm labor housing loans support the development of housing for domestic farm laborers.
• Single-family housing credit sales facilitate the sale of government-owned properties to eligible rural families with low income.

Together, those programs are designed to fill perceived gaps in credit availability and improve affordability in rural communities, particularly among populations that may be underserved by private lenders.

Other Federal and Conventional Loan Programs

Rural households may also access other federal and conventional mortgage credit channels. Those channels include loan guarantees provided by the Federal Housing Administration (FHA) or the Department of Veterans Affairs (VA), which operate in a similar manner to section 502 loans, and conventional mortgages guaranteed by Fannie Mae and Freddie Mac (the government-sponsored enterprises, or GSEs).

Typically, those programs have conditions, such as FHA’s and the GSEs’ down payment requirements and VA’s military service requirement. In addition, they often impose credit score, income, or loan limit requirements. Those requirements are generally more restrictive than requirements imposed by RHS programs.

Federal Credit Programs That Support Renters in Rural Areas

Rural renters face a range of housing market conditions, including a limited supply of rental units, older housing stock, underdeveloped infrastructure, fewer transportation options, and lower average wages. The Housing Act of 1949 also established programs to address renters’ challenges. Some of those programs focus on increasing the supply of rental units by making it less expensive to build; others are designed to increase demand by reducing the cost of renting to eligible households.

Section 538 Guaranteed Rural Rental Housing Program

Under the Section 538 Guaranteed Rural Rental Housing Program, RHS provides partial loan guarantees to qualified lenders against losses on multifamily mortgages. The program is designed to increase the supply of rental housing for households with low to moderate income in rural areas; it does so by extending credit to developers and property owners for construction and rehabilitation of multifamily rental housing in eligible rural areas. Borrowers must agree to income limits for tenants and to rent restrictions to ensure the properties serve the targeted rural households. The program facilitates long-term financing and can be combined with other private financing tools or government programs.

Section 515 Rural Rental Housing Direct Loan Program

Another RHS program, the Section 515 Rural Rental Housing Direct Loan Program, also finances the development of affordable multifamily rental housing in rural communities to increase the supply of housing units. Unlike the section 538 program, which relies on guarantees to private lenders, through the section 515 program, RHS provides direct loans to eligible entities. The loans finance the construction, rehabilitation, or acquisition of multifamily rental properties for low-income, elderly, or disabled individuals and families. Properties financed under the section 515 program must meet rent and income restrictions. The amount of loans made by that program has declined, from more than $500 million in the early 1990s to around $50 million in recent years.

Other Federal Programs for Renters

FHA and the GSEs administer multifamily guarantee programs that support financing for the construction, acquisition, and rehabilitation of rental properties as a means of increasing the supply of rental units.

Secondary Market Support for Rural Housing Finance

The federal government operates programs that facilitate the sale of mortgages in the secondary (resale) market as a means of expanding mortgage credit access in rural communities. Those programs allow lenders to sell loans and obtain new capital, which can increase the supply of mortgage credit available to rural homebuyers and rural rental housing developers. Several government entities guarantee the timely payment of principal and interest on mortgage-backed securities (MBSs).9

Ginnie Mae

The Government National Mortgage Association (Ginnie Mae), a government-owned corporation within the Department of Housing and Urban Development (HUD), guarantees the timely payment of principal and interest on MBSs backed by loans insured or guaranteed by RHS, FHA, and VA. For rural borrowers, particularly those served by RHS programs, Ginnie Mae’s guarantee enhances the liquidity of mortgages and increases the ability of lenders to originate loans in rural areas.

Farmer Mac

The Federal Agricultural Mortgage Corporation (Farmer Mac) is a GSE that serves as a secondary market for agricultural and rural housing loans, including those guaranteed or originated by RHS. Farmer Mac guarantees the timely payment of principal and interest on MBSs backed by USDA loans. Farmer Mac’s activities are designed to increase liquidity for lenders operating in rural areas and to broaden the availability of capital for housing and infrastructure needs in those regions.

Fannie Mae and Freddie Mac

Fannie Mae and Freddie Mac also play a role in supporting mortgage financing in rural areas. Both entities guarantee the timely payment of principal and interest on MBSs that include mortgages secured by properties in rural communities. Although their programs are not specific to rural areas, Fannie Mae and Freddie Mac include rural loans within their broader portfolios and may participate in targeted initiatives to promote the financing of rural housing.10

Other Federal Programs That Support Rural Communities

Support for rural communities is not limited to federal credit programs. Other federal housing programs—administered primarily by FHA and the Department of the Treasury—are available to participants in rural areas, including the following:

• HUD’s Housing Choice Voucher Program (section 8) provides direct subsidies to low-income renters, thus increasing housing demand by offsetting the cost of renting in the private market.
• Public housing—owned and operated by local housing authorities and subsidized by HUD—provides a limited increase in the supply of rental units in some rural communities.
• The Low-Income Housing Tax Credit (LIHTC) Program offers federal income tax credits to eligible private developers that build new affordable rental housing or substantially rehabilitate existing housing, thus increasing the supply of housing units.11

The federal government also operates various programs to promote infrastructure investment that support rural homeowners and renters in rural communities. Those investments include the development of utility infrastructure, health care facilities, community centers, fire departments, police stations, and libraries. The programs that support rural communities include the following:

• Rural Utilities Service’s water, waste disposal, electrification, and telecommunication direct loans and loan guarantees;
• RHS’s rural development community facilities guarantees, direct loans, and grants;
• FHA’s hospitals loan guarantees; and
• FHA’s residential care facilities loan guarantees.

Those programs, although part of the broader landscape of federal rental assistance and development that extends into rural housing markets, are outside the scope of this report and were not used to calculate the budgetary cost of federal credit programs’ support for rural housing and homeownership.

Budgetary Effects of Federal Credit Programs for Rural Housing

The federal government administers 31 credit programs that provide assistance to homeowners and renters, 10 of which are administered by RHS and focus exclusively on rural households. The remaining 21 programs are administered by other agencies and partly support rural households.12 How much of those programs’ activities provide credit assistance specifically to rural communities is uncertain, but estimates can be made on the basis of Home Mortgage Disclosure Act (HMDA) data and the community characteristics described in this report. On the basis of those estimations, CBO included some or all of each federal credit program’s activities in its analysis of the budgetary effect of federal credit programs for rural housing (see Figure 9). In CBO’s projections, the federal government provides $293.4 billion in credit assistance to rural households in 2026; $27.4 billion is administered by RHS and $266.0 billion by HUD, VA, and the GSEs.

HMDA data, which are collected by lenders on single-family mortgages they originate, classify loans as either conventional (typically guaranteed by Fannie Mae or Freddie Mac or held in originator portfolios) or as guaranteed by the federal government through FHA, VA, or RHS.13 Using county-level HMDA data for 2022 and 2023 combined with Census data, CBO estimated that approximately 20 percent of the single-family guarantees made by FHA and VA and 15 percent of the single-family guarantees of Fannie Mae and Freddie Mac are made to rural households. (For more details about federal programs that support single-family home mortgage guarantees, see Box 1.)

To estimate the composition of other programs’ portfolios, CBO relied on the demographic and financial characteristics of rural communities. Rural households make up 20 percent of the U.S. population, but given the characteristics of rural households, they may make up a smaller or larger share of a particular program’s portfolio. For example, rural communities have a higher homeownership rate than non-rural communities do, so the share of rural borrowers in single-family loan and loan guarantee programs may be larger than their share of the total population. Their participation in certain multifamily housing programs may be lower. Likewise, rural communities have lower median incomes than non-rural communities do, so single-family loan and loan guarantee programs geared toward low-income households (such as FHA’s programs) may have a larger share of rural borrowers. Their share of certain loan and loan guarantee programs for moderate-income households (such as those of the GSEs) may be smaller.

FCRA and Fair Value Estimates

The budgetary effects of the federal credit assistance can be calculated in two ways. The first uses procedures specified in FCRA and is the standard way that costs of credit programs are measured in the federal budget. The second uses a fair-value approach, which measures the market value of the government’s obligations by accounting for market risk. Market risk is the component of financial risk that is associated with the overall performance of the economy rather than with the performance of a specific investment; it results from shifts in macroeconomic conditions, such as productivity and employment, and from changes in expectations about future macroeconomic conditions.14

Incorporating market risk into budget estimates has advantages and disadvantages depending on the circumstances and purpose of the estimates. FCRA estimates, which discount cash flows at Treasury rates, reflect the likely effect of a policy on projected federal debt. Fair-value estimates, which include a premium for market risk and thus provide a more comprehensive measure of costs, are more useful for comparing different forms of government assistance.

Lifetime Budgetary Effects of Assistance Administered by RHS

The amount of federal credit assistance projected to be administered by RHS for 2026 is $27.4 billion (see Table 1). Using FCRA procedures, CBO estimates that new loans and loan guarantees issued by RHS in 2026 would generate savings to the federal government of $0.2 billion over their lifetime. Using the fair-value approach, CBO estimates that those loans and loan guarantees would have a lifetime cost of $0.5 billion.

The single-family guarantee program represents more than 90 percent of the total credit assistance projected to be provided by RHS in 2026. On a FCRA basis, that program is projected to generate savings of nearly $0.2 billion. The remaining programs—representing less than 10 percent of the assistance—produce FCRA savings and costs that nearly offset each other.

Lifetime Budgetary Effects of Assistance Administered by Other Entities

In 2026, federal credit programs administered by HUD, VA, and the GSEs are projected to provide $1.6 trillion in credit assistance in the form of housing and real estate loans and loan guarantees.15 Of that total, about $266 billion supports households in rural areas, CBO estimates (see Table 2). Calculated on a FCRA basis, the lifetime budgetary savings generated by those programs’ support for rural housing are projected to be $3.1 billion. Calculated on a fair-value basis, the lifetime cost is projected to be $2.8 billion.

Combined Lifetime Budgetary Effects

Including RHS and non-RHS programs, the total support for households in rural areas is $293.4 billion in 2026, CBO estimates. Calculated on a FCRA basis, the lifetime budgetary savings generated by that support for rural housing are projected to be $3.4 billion. Calculated on a fair-value basis, the lifetime cost is projected to be $3.2 billion.

Sensitivity Analysis of the Budgetary Effects of Federal Credit Programs for Rural Housing

Rural borrowers in non-RHS programs may have higher or lower subsidy costs than the average borrowers in those programs for various reasons. For example, if lower median incomes caused rural borrowers to default at higher rates than non-rural borrowers in those programs, those rural borrowers might have a higher subsidy rate than non-rural borrowers. (The subsidy rate is the cost divided by the amount disbursed; a positive rate indicates a government subsidy, and therefore a cost to the government, and a negative rate indicates budgetary savings.)16 Alternatively, if less expensive housing relative to income caused rural borrowers to default at lower rates than non-rural borrowers in those programs, those rural borrowers might have a lower subsidy rate than non- rural borrowers.

The degree to which rural borrowers produce higher or lower subsidy costs than non-rural borrowers in programs outside of RHS—in which all borrowers are rural—is uncertain. CBO analyzed how the lifetime budgetary effects would be different if subsidy rates for rural borrowers in non-RHS programs were 10 percent higher or 10 percent lower than the average subsidy rate in those programs.

If subsidy rates were 10 percent higher, rural borrowers in the 31 federal credit programs supporting rural housing would save the federal government $3.0 billion on a FCRA basis and cost the federal government $3.5 billion on a fair-value basis (see Table 3, upper panel).17 If subsidy rates for rural borrowers in non-RHS programs were 10 percent lower than the average subsidy rate in those programs, those rural borrowers would save the federal government $3.7 billion on a FCRA basis and cost the federal government $3.0 billion on a fair-value basis (see Table 3, lower panel). Adjustments larger or smaller than 10 percent would generate linearly larger or smaller changes in budgetary costs and savings.

---

1. 1. For further discussion of the two methods, see Congressional Budget Office, How CBO Produces Fair-Value Estimates of the Cost of Federal Credit Programs: A Primer (July 2018), www.cbo.gov/publication/53886.

2. 2. Census Bureau, “County-Level Urban and Rural Information for the 2020 Census” (December 16, 2024), https://tinyurl.com/4e2repbe, and “2019–2023 ACS 5-Year Estimates” (December 12, 2024).

3. 3. Census Bureau, Understanding and Using American Community Survey Data: What Users of Data for Rural Areas Need to Know (June 2019), p. 2, https://tinyurl.com/mrx84mkd.

4. 4. Not all agencies and programs use the same threshold to identify rural areas. As a result, the classification of areas as rural often depends on the agency, the purpose of the program, or the specific statistical standard being applied. However, the Census Bureau’s geographic classification system serves as the foundation for their determinations. For more about that classification system, see Census Bureau, “County-Level Urban and Rural Information for the 2020 Census” (December 16, 2024), https://tinyurl.com/4e2repbe.

5. 5. Census Bureau, “Urban and Rural” (December 16, 2024), https://tinyurl.com/4c8hk8e6.

6. 6. Counties are considered entirely rural if less than 0.5 percent of the population is defined as urban and are considered entirely urban if more than 99.5 percent of the population is defined as urban.

7. 7. The Census Bureau conducts a full census every 10 years, most recently in 2020, but it also contacts people at a sample of addresses annually to collect information about people and their households through the ACS. From the survey responses, the Census Bureau produces five-year estimates of demographic, social, housing, and economic characteristics for census blocks.

8. 8. Alaina Barca and Harry Hou, U.S. Bank Branch Closures and Banking Deserts (Federal Reserve Bank of Philadelphia, February 2024), https://tinyurl.com/2cc4hpxr.

9. 9. MBSs are financial securities whose payments of principal and interest are backed by the payments from a pool of mortgages.

10. 10. For a discussion of the GSEs’ purchases of mortgages made to low-income families and certain underserved populations, including those in rural areas through the GSEs’ “duty to serve” requirements, see Congressional Budget Office, Fannie Mae and Freddie Mac’s Housing Goals (November 2024), www.cbo.gov/publication/60190.

11. 11. Andrew M. Dumont, Rural Affordable Rental Housing: Quantifying Need, Reviewing Recent Federal Support, and Assessing the Use of Low Income Housing Tax Credits in Rural Areas, Finance and Economic Discussion Series 2018-077 (Board of Governors of the Federal Reserve System, July 2018), https://doi.org/10.17016/FEDS.2018.077.

12. 12. Congressional Budget Office, Estimates of the Cost of Federal Credit Programs in 2026 (January 2026), www.cbo.gov/publication/61645; and Office of Management and Budget, Technical Supplement to the 2026 Budget: Credit Supplement (June 2025), https://tinyurl.com/48dfrn7p.

13. 13. The Consumer Financial Protection Bureau has developed a tool that allows users to filter, aggregate, and download datasets. See Consumer Financial Protection Bureau, “HMDA Data Browser” (accessed August 26, 2025), https://ffiec.cfpb.gov/data-browser/.

14. 14. For more information about the fair-value approach, see Congressional Budget Office, Estimates of the Cost of Federal Credit Programs in 2026 (January 2026), www.cbo.gov/publication/61645.

15. 15. In CBO’s projections, much of that assistance—$1.1 trillion in mortgage guarantees—is provided by Fannie Mae and Freddie Mac. Fannie Mae and Freddie Mac primarily buy mortgages from lenders and then pool the mortgages to create mortgage-backed securities, which they guarantee against default and sell to investors. Because the GSEs are in federal conservatorship (that is, the federal government retains operational control and effective ownership of them), CBO regards their loan guarantees as government commitments. By contrast, the Administration does not consider them to be government entities and therefore accounts for their loan guarantees differently.

16. 16. To calculate the budgetary costs or savings, CBO and other federal agencies multiply the size of the commitment or obligation by the subsidy rate, so programs with high subsidy rates do not necessarily have the largest total budgetary impact.

17. 17. The subsidy rates for RHS programs were not adjusted as a part of this sensitivity analysis because all borrowers in those programs are rural.

About This Document

This report was prepared to enhance the transparency of the work of the Congressional Budget Office. In keeping with the Congressional Budget Office’s mandate to provide objective, impartial analysis, the report makes no recommendations.

Mitchell Remy wrote the report with guidance from Sebastien Gay. Julia Aman, Michael Falkenheim, Ron Gecan, Paul B. A. Holland, Justin Humphrey, Zunara Naeem, Erik O’Donoghue, Lara Robillard, Emily Stern, David Torregrosa, and James Williamson offered comments. David Torregrosa also fact-checked the report.

Andrew M. Dumont of the Board of Governors of the Federal Reserve System, Tobias Peter of the American Enterprise Institute, and Lance George, Jonathan Harwitz, and David Lipsetz of the Housing Assistance Council commented on an earlier draft. The assistance of external reviewers implies no responsibility for the final product; that responsibility rests solely with CBO.

Jeffrey Kling reviewed the report. Caitlin Verboon edited it, and Jorge Salazar created the graphics and prepared the text for publication. The report is available at www.cbo.gov/publication/61946.

CBO continually seeks feedback to make its work as useful as possible. Please send any comments to communications@cbo.gov.

Phillip L. Swagel

Director

April 2026