Summary

Fannie Mae and Freddie Mac are government-sponsored enterprises (GSEs) that purchase mortgages from lenders and package them into securities to be sold to investors. The two GSEs guarantee the timely payment of those securities, charging fees in exchange for that guarantee. Under law and by regulation, a certain percentage of the mortgages that the GSEs purchase must be made to low-income households and certain underserved populations. The details of those purchases are described in directives known as housing goals.

Fannie Mae and Freddie Mac’s Housing Goals

Fannie Mae and Freddie Mac’s housing goals can be divided into two main categories: those covering purchases of mortgages on single-family housing and those covering purchases of mortgages on multifamily housing. The goals are designed to promote affordable and sustainable homeownership (through the single-family mortgages) and rental opportunities (through the multifamily mortgages) and to ensure that the two GSEs fulfill their mission of responsibly expanding access to mortgage credit. Although the single-family and multifamily goals are similar, the processes for achieving the goals differ. This report focuses on the GSEs’ single-family housing goals. The GSEs’ success in meeting those goals is assessed annually using the lesser of two measures: a forecast of the share of mortgages that meet the goals’ requirements, known as the benchmark level; or data showing the share of mortgages, by all lenders, that met the requirements, known as the market level.

The single-family housing goals and subgoals are generally based on two criteria: the level of borrowers’ income and whether mortgages are for purchasing a residence (purchase loans) or refinancing an existing mortgage (refinance loans). Low-income borrowers are defined as those whose income is at or below 80 percent of the area median income (AMI); very low income borrowers have income at or below 50 percent of the AMI.

The multifamily housing goals are intended to ensure that the GSEs fulfill their mission of contributing to the availability of affordable rental housing. Those goals are defined by either the income of the renting household (using the same definitions of low income and very low income that are used for the single-family housing goals) or the size of the rental property. For both single-family and multifamily housing goals, individual mortgages may count toward more than one goal.

How the Housing Goals Are Met

In this report, the Congressional Budget Office describes how the housing goals are enforced and analyzes how the two GSEs shift their purchases toward mortgages that meet the goals’ requirements (goal-eligible mortgages). Fannie Mae and Freddie Mac charge ongoing and up-front fees that are similar to insurance premiums. When averaged across all mortgage purchases, the fees cover the GSEs’ expected credit losses, administrative costs, and the cost of capital. In CBO’s assessment, to meet their housing goals, the GSEs charge lower fees than they would, absent the goals, to lenders who sell them goal-eligible mortgages. Those lower fees are an implicit subsidy. One way to recoup the cost of those lower fees is to charge higher fees—an implicit tax—than they otherwise would on single-family mortgages that do not meet the goals’ requirements. The subsidies and taxes result in an average fee, across all purchases, that covers expected credit losses, administrative costs, and the cost of capital.

Effects of Implicit Subsidies and Taxes

CBO estimates that in fiscal year 2025, nearly 750,000 households purchasing a single-family home will each receive an implicit subsidy (in the form of lower fees) averaging $2,282 because their mortgages are goal-eligible. About 1.4 million households purchasing a single-family home will pay an implicit tax (in the form of higher fees) because their mortgages are not goal-eligible. About 230,000 of the latter households whose mortgages are on investment properties and second homes will each pay an average of nearly $8,000 in higher fees; the rest will pay an average of about $160 in higher fees. Moreover, in fiscal year 2025, the implicit subsidy will result in the GSEs’ purchasing about 37,000 additional mortgages that meet the goals’ requirements, and the implicit tax will result in their purchasing roughly 29,000 fewer mortgages that do not meet the requirements.

That shift in mortgage purchases affects outcomes for borrowers and the federal budget, but an exact estimate of those effects is beyond the scope of this report. In the absence of the goals’ requirements, many of the 37,000 additional mortgages that the GSEs will purchase because of the implicit subsidy would be originated (that is, processed, approved, and funded) anyway, perhaps through federal programs such as the Federal Housing Administration’s (FHA’s) mortgage guarantee program; however, some mortgages might not be originated at all. Similarly, some of the 29,000 mortgages that the GSEs will not purchase because of the implicit tax may be originated by other lenders. CBO did not estimate the net effect of Fannie Mae and Freddie Mac’s housing goals on the lifetime cost of their mortgage guarantees, which are backed by the federal government. But in CBO’s assessment, that cost is probably close to zero because higher fees for mortgages that do not meet the goals’ requirements offset the lower fees for mortgages that do.

Overview of Fannie Mae and Freddie Mac’s Housing Goals

Fannie Mae and Freddie Mac were chartered as GSEs to ensure a stable supply of credit for mortgages nationwide, including those for low- and moderate-income households.1 The two GSEs dominate the secondary (resale) market for residential mortgages, in which they buy mortgages from lenders, pool the mortgages into mortgage-backed securities (MBSs), and sell the securities to investors with a guarantee against most losses from defaults on the underlying loans. In September 2008, the Federal Housing Finance Agency (FHFA), the GSE’s regulator, placed Fannie Mae and Freddie Mac in federal conservatorship, thus taking control of their assets and business.2

By law, Fannie Mae and Freddie Mac are subject to housing goals (a few of which are categorized as subgoals) that serve as benchmarks for measuring how the GSEs facilitate the financing of affordable housing for low-income families.3 FHFA sets the goals for purchasing mortgages on single-family housing (directed toward homeownership) and multifamily housing (directed toward rental housing) through a public rulemaking process. The goals, which are evaluated annually, aim to promote affordable and sustainable homeownership and rental opportunities for low-income families and underserved communities and are designed to ensure that the GSEs fulfill their mission of expanding access to mortgage credit responsibly.

The single-family housing goals and subgoals for residential mortgages purchased by Fannie Mae and Freddie Mac are categorized as follows (see Table 1):

• Purchase loans for low-income households (that is, households whose income is at or below 80 percent of the AMI);
• Purchase loans for very low income households (households whose income is at or below 50 percent of the AMI);
• Refinance loans for low-income households;
• A subgoal covering two types of purchase loans—those for households at any level of income that live in a low-income census tract that is not a minority census tract, and those for households whose income is greater than 100 percent of the AMI and that live in a low-income census tract that is also a minority census tract; and
• A subgoal covering purchase loans for households whose income is at or below 100 percent of the AMI living in minority census tracts.4

  Table 1.

  Single-Family Housing Goals

  

  Notes

  Data source: Congressional Budget Office.

  Area-based subgoals took effect in 2022. A low-income census tract is one in which median income is no greater than 80 percent of the AMI. A minority census tract is one in which at least 30 percent of the population consists of people in racial or ethnic minority groups and in which median income is less than 100 percent of the AMI.

  AMI = area median income.

Individual mortgages may count toward more than one goal or subgoal.5

The multifamily housing goals are intended to ensure that Fannie Mae and Freddie Mac contribute to the availability of affordable rental housing (see Table 2). The goals cover housing that has:

• Rental units affordable to low-income households,
• Rental units affordable to very low income households, and
• Five to 50 rental units (a size that often meets the housing needs of lower-income households) affordable to low-income households.

  Table 2.

  Multifamily Housing Goals

  

  Notes

  Data source: Congressional Budget Office.

  AMI = area median income.

Like the single-family mortgages, individual multifamily mortgages may count toward more than one goal.

Performance in Meeting the Housing Goals

Fannie Mae’s and Freddie Mac’s success in meeting their single-family housing goals is assessed by comparing the number of goal-eligible mortgages they purchase with two measures: a benchmark level and a market level. Set by federal regulation, the benchmark level mainly reflects FHFA’s forecast of the share of loans that meet the goals’ requirements. The market level is based on data showing the share of loans, by all lenders, that fell within the targets for the respective goals using the Home Mortgage Disclosure Act database.6 The percentage of goal-eligible mortgages purchased by the two GSEs must meet or exceed at least one of those levels.

Single-Family Housing Goals

Data published by FHFA for the 2018–2022 period indicate that Fannie Mae and Freddie Mac met or exceeded nearly all of their single-family housing goals and subgoals in those years (see Table 3).7 In 2022, Fannie Mae did not meet the benchmark levels for purchase loans for low-income and very low income households, partly because those levels had increased since 2021. The increase was largely attributable to the way that FHFA developed its market forecast. As a result of the factors that FHFA uses to set its benchmark level—including actual market levels in recent years—those benchmarks increased in 2022, from 24 percent to 28 percent for purchase loans for low-income households, and from 6 percent to 7 percent for such loans for very low income households.8 Fannie Mae did, however, meet market levels for both goals that year. (The implications of not achieving one or more goals are discussed in the section titled “Effects of Housing Goals on Fannie Mae’s and Freddie Mac’s Operations.”)

Multifamily Housing Goals

Multifamily housing goals are assessed using only a benchmark level. Before 2023, that level was defined as the number of affordable housing units in multifamily properties financed by mortgages that the two GSEs purchased. Beginning in 2023, however, the benchmark level has been defined as the percentage of affordable units in those properties. That change was made, in part, to reduce FHFA’s operational burden in forecasting benchmark levels and to better align the multifamily housing goals with market conditions.9

When the benchmark level was tied to the number of units, and the volume of mortgages with goal-eligible units in the market was higher than expected (based on the forecast underlying the benchmark), the GSEs did not need to acquire further mortgages with goal-eligible units after reaching the benchmark. Conversely, to meet the benchmark in a market with lower-than-expected volume, the GSEs may have needed to acquire mortgages that they would otherwise reject. But when the benchmark level is based on the percentage of affordable units, higher- or lower-than-expected volumes of mortgages with goal-eligible units in the market result in neither of those behaviors.

From 2015 to 2022, Fannie Mae and Freddie Mac met or exceeded all of their multifamily housing goals, according to data published by FHFA (see Table 4). In response to the notice of proposed rulemaking for the 2023 and 2024 benchmarks, the two GSEs, along with industry participants, expressed general support for the revised approach.10

Effects of Housing Goals on Fees for Single-Family Mortgages

The two GSEs cover their expected credit losses, administrative costs, and the cost of capital by charging ongoing and up-front fees, similar to insurance premiums, to guarantee mortgages. In CBO’s assessment, the GSEs subsidize the fees for goal-eligible mortgages by charging lower fees to lenders than they would without the goals. The GSEs pay for those implicit subsidies by taxing mortgages that do not meet the goals’ requirements with higher fees than they would otherwise charge.11 The GSEs implement that implicit subsidy and tax by:

• Offering discounts on fees to lenders who sell them goal-eligible mortgages for cash through a mechanism called the cash window;
• Increasing or decreasing the fees charged to lenders for the mortgages they provide for MBSs based, in part, on the share of goal-eligible mortgages; and
• Waiving certain up-front fees for goal-eligible mortgages.

In CBO’s assessment, the GSEs use each of those three strategies to a degree that minimizes costs, which is consistent with the aims of a profit-maximizing company.12 Cost minimization implies that on the margin, the cost of acquiring one additional goal-eligible mortgage will be the same for each strategy. Otherwise, the GSEs could reduce costs by reallocating their efforts from strategies with a higher cost to those with a lower cost.

Accordingly, to estimate the amount of the implicit subsidy and tax, CBO focused on the waivers of up-front fees—the strategy with the most transparent cost—and determined that the marginal costs of the other two strategies would be the same. Directly estimating the amount of the subsidy and tax associated with MBS transactions or with transactions made using the cash window is difficult because those transactions vary by lender and the data associated with them are less available. The fee waivers, however, represent a uniform discount applied to all lenders and all mortgages with similar characteristics. As a result, the marginal cost for each goal-eligible mortgage acquired using that strategy is close to the average cost.

Using that approach, CBO estimates that in fiscal year 2025, about 1.4 million households purchasing a single-family home with a mortgage that does not meet the goals’ requirements will pay an implicit tax in the form of higher up-front fees. Nearly 750,000 households purchasing a single-family home with a goal-eligible mortgage will receive an implicit subsidy through discounts on such fees. In addition, the two GSEs will acquire about 69,000 goal-eligible mortgages (for purchasing a home) without implicitly subsidizing them, and about 115,000 mortgages that do not meet the goals’ requirements will be subsidized, in CBO’s estimation.

How Fees Are Set

The fees charged by the two GSEs on their single-family and multifamily mortgage guarantees are set at a level that is chosen to cover their expected credit losses, administrative costs, and the cost of capital.13 Although the goals of the fee-setting process for single-family and multifamily guarantees are similar, the execution differs. This report focuses on the GSEs’ single-family mortgage guarantee program.

Fannie Mae and Freddie Mac charge two types of fees for single-family mortgages: an ongoing fee and an up-front fee. An ongoing fee is incorporated into the borrower’s interest rate and assessed against the outstanding balance of the mortgage. That fee, which is quoted in basis points, is typically negotiated with the lender that originates the mortgage and is generally the same for all mortgages issued by that lender over some period, regardless of the mortgages’ individual characteristics. (A basis point is one-hundredth of a percentage point.)

When a mortgage is originated, the GSEs charge the lender an up-front fee assessed on the amount of the mortgage’s initial balance. That fee is typically the same for all lenders but is then risk-adjusted to account for a mortgage’s individual characteristics, including its purpose (for purchase or refinance), the borrower’s credit score, the original loan-to-value ratio (LTV), and the type of residency (primary residence, second home, or investment property). In general, up-front fees are higher for borrowers who are refinancing an existing mortgage (particularly if the new mortgage has a higher balance than the existing one), borrowers who have lower credit scores and higher LTVs, and borrowers who are financing a property other than their primary residence. The up-front fees are higher because each of those characteristics is associated with a greater likelihood of default.

Using data from FHFA’s report on single-family guarantee fees assessed in 2022, CBO estimates that the two GSEs charge an average ongoing fee of about 44 basis points and an average up-front fee of approximately 94 basis points on all mortgages.14 (When converted to an ongoing fee, the up-front fee equals about 17 basis points, on average.) The average fee for mortgages used to purchase a home was slightly lower than the average fee for mortgages used to refinance an existing mortgage. For purchase loans, the average up-front fee was approximately 92 basis points, CBO estimates.

How Fees Are Used to Achieve the Housing Goals

Fannie Mae and Freddie Mac have used a variety of strategies to achieve their housing goals. Before 2004, the two GSEs could largely achieve those goals without offering discounted fees related to individual mortgages.15 From 2004 through September 2008 (the month they were placed in federal conservatorship), the GSEs supplemented routine purchases with nonroutine bulk purchases of goal-eligible mortgages from specific lenders. Those bulk purchases allowed the GSEs to target mortgages with specific characteristics and thus to increase their share of goal-eligible mortgages.

Since being placed in conservatorship, Fannie Mae and Freddie Mac have largely ceased bulk purchases in favor of a combination of three other strategies:

• First, they purchase mortgages directly from lenders for cash through the cash window. The cash window was designed to allow lenders to sell their mortgages to the GSEs even if they are unable to participate in the MBS market. (Lenders who originate a small volume of loans each year are often unable to participate in the MBS market, for example). Like bulk purchases, the cash window allows the GSEs to acquire specific mortgages from specific lenders and to offer a price for them that reflects the target subsidy based on the mortgages’ goal eligibility.
• Second, the two GSEs vary the fees they charge to lenders for mortgages pooled into MBSs. The fees vary on the basis of several criteria, including the share of those mortgages that are goal-eligible. For example, when lenders consistently deliver a higher percentage of mortgages that meet the requirements of one or more housing goals, the GSEs may charge those lenders a lower ongoing fee for future deliveries than the fee paid by lenders who deliver a lower share of such mortgages.
• Third, the GSEs announced in September 2022 that beginning in February 2023, up-front fees would be waived for borrowers with mortgages that meet certain criteria related to goal eligibility.16 The waiver would apply to first-time homebuyers whose income is at or below 100 percent of the AMI; borrowers with down payments as low as 3 percent who also earn less than 80 percent of the AMI (with mortgages originated by lenders through Fannie Mae’s HomeReady program and Freddie Mac’s Home Possible program); and certain households whose income is at or below 100 percent of the AMI in underserved markets, as defined by the GSEs’ “duty to serve.”17 (For a description of the GSEs’ duty to serve, see Box 1.)

  Box 1.

  Fannie Mae and Freddie Mac’s Duty to Serve Underserved Segments of the Housing Market

  Under the Housing and Economic Recovery Act of 2008, Fannie Mae and Freddie Mac (which are government-sponsored enterprises, or GSEs) have a duty to serve three underserved segments of the housing market:

  • Manufactured housing, including manufactured homes titled as real estate or as personal property;
  • Affordable housing preservation, such as the financing of small multifamily rental properties and energy- or water-efficiency improvements; and
  • Rural housing, including housing in high-needs rural regions, rural housing financed by small financial institutions, and small multifamily rental properties in rural areas.1

  In each market, the two GSEs must undertake activities to finance properties that house families whose income is no greater than 50 percent, 80 percent, or 100 percent of the area median income, depending on the activity. That financing must occur through an array of activities specified by the GSEs.

  Within the manufactured housing and rural housing markets, each GSE must undertake at least four activities (such as financing manufactured housing titled as personal property, manufactured housing owned by nonprofits, housing in high-needs rural regions, and small multifamily rental properties in rural areas). In the area of affordable housing preservation, each GSE must consider at least seven separate activities (such as financing small multifamily properties, financing energy- or water-saving projects, and rehabilitating distressed properties). The Federal Housing Finance Agency monitors the GSEs’ progress and compliance in serving underserved segments of the housing market.

  ---

  1. 1. High-needs rural regions are those located in the following areas: Middle Appalachia, the Lower Mississippi Delta, a colonia, or counties elsewhere with persistent poverty.

In practice, the GSEs use a combination of all three strategies to help achieve their housing goals. Using cash purchases and the practice of discounting MBS fees will remain the most common strategies for acquiring goal-eligible refinance loans, CBO estimates.

Before introducing fee waivers in February 2023, the GSEs mainly relied on lowering ongoing fees for lenders with disproportionately high shares of goal-eligible mortgages. The amount of the discount on those ongoing fees was negotiated with each lender. Since February 2023, the two GSEs have used fee waivers as an explicit and uniform discount for goal-eligible mortgages, which, all else being equal, would lessen the need to offer implicit discounts by reducing ongoing fees.

Negotiating ongoing fees with lenders still has advantages. It gives the GSEs flexibility to offer greater discounts on their fees during periods when achieving housing goals is expected to be difficult, and lesser discounts when goal-eligible mortgages can be acquired more easily. However, negotiating ongoing fees is less transparent than waiving up-front fees and may be less likely to result in an implicit subsidy passing through to borrowers. (Borrowers have access to the GSEs’ published up-front fees but generally do not know the amount of ongoing fees included in their mortgage rate.) Moreover, that lack of transparency is inconsistent with Fannie Mae and Freddie Mac’s mandate to promote access to affordable mortgage credit. CBO anticipates that in 2025, the GSEs will mainly rely on fee waivers to subsidize goal-eligible purchase loans, which also reflects the value of transparency to the GSEs, mortgage market participants, and regulators.

Because of that expectation, CBO also anticipates that the GSEs will incur an average cost of shifting toward goal-eligible purchase loans that is close to the discount offered on fee waivers. The amount of those fee waivers determines the goals’ effect on the price of mortgages. The GSEs could negotiate a smaller discount for each goal-eligible mortgage in the cash window from some lenders. They could also offer additional discounts on ongoing fees for those goal-eligible mortgages packaged into MBSs, primarily in pursuit of goal-eligible refinance loans. In CBO’s estimates, the effect of the housing goals on the price of mortgages—particularly the price of purchase loans—equals the amount of the fee waiver because the waiver is the most transparent discount and is available to every lender in a competitive market.

The amount of the implicit subsidy offered through the three strategies depends on the degree to which the GSEs decide they are at risk of not achieving a particular housing goal.18 For goals that are achievable without changing their acquisitions, mortgages will generally be priced without a subsidy or tax. However, when a particular goal is at risk of not being met, the GSEs may need to offer concessions that depend on the composition of mortgages they have acquired that are subject to their standard fees.

Determining which goals might not be met is complicated by the nature of the targets set for the goals. During a given year, the GSEs base their progress on the benchmark level established by FHFA in advance. However, their performance will ultimately be judged against the lower of the benchmark level or the market level, and the latter is unknown until after the year has ended. As the calendar year progresses, the GSEs probably gain a sense of how much the benchmark and market levels differ and can adjust their concession strategies accordingly.

Waivers Offered on Up-Front Fees for Goal-Eligible Mortgages

Although waivers on up-front fees are not explicitly linked to the housing goals, they largely align with the goals, except in the case of refinance loans. Goal-eligible refinance loans originated through the HomeReady or Home Possible program are eligible for the fee waiver on the basis of household income; but some of those mortgages do not receive the waiver because they fail to meet those programs’ other criteria.19 As a result, any subsidy necessary to achieve the goal for refinance loans for low-income households will still be provided through the cash window and through discounted fees on MBSs, CBO estimates. For that reason, the analysis in this report focuses on the use of fee waivers for acquiring goal-eligible purchase loans.

Fee waivers align more closely with the goals associated with purchase loans, but not completely. For example, mortgages for first-time homebuyers whose income is 100 percent of the AMI or less and who receive a fee waiver also count toward the low-income purchase goal if the borrower’s income is 80 percent of the AMI or less. Such mortgages count toward the very low income purchase goal if the borrower’s income is 50 percent of the AMI or less, and toward one of the area-based subgoals if it is 100 percent of the AMI or less and the property is located in an eligible census tract.

In addition, mortgages originated by lenders through the HomeReady or Home Possible programs, and for which up-front fees are waived, also meet the goals for low-income purchase loans or low-income refinance loans, depending on the purpose of the mortgage. (Both programs both require that households make no more than 80 percent of the AMI.) Those mortgages would also meet the goal for purchase loans for very low income households if the borrower’s income was 50 percent of the AMI or less, and they would meet one of the area-based subgoals if the property was located in an eligible census tract. Mortgages for households meeting criteria under the two GSEs’ “duty to serve” obligation also meet the housing goals if the borrower’s income is below certain thresholds or if they are purchasing properties located in an eligible census tract.

CBO estimates that in fiscal year 2025, 35 percent of purchase loans will meet one or more housing goals (see Table 5).20 That estimate is based on CBO’s assessment that 21 percent of all purchase loans will meet only the low-income or very low income purchase goal and that an additional 7 percent will meet only an area-based subgoal. The remaining 7 percent will meet the low-income or very low income purchase goal (or both), as well as an area-based subgoal.

The criteria for a purchase loan to count toward one or more housing goals and the criteria for waiving up-front fees are not completely aligned. Nevertheless, CBO’s analysis indicates that in fiscal year 2025, approximately 90 percent of the purchase loans that meet a housing goal—or 30 percent of all purchase loans that the two GSEs will guarantee that year—will involve a fee waiver. CBO estimates that the remaining goal-eligible purchase loans that do not involve fee waivers are mortgages that meet the subgoal for families in low-income census tracts whose income is above the threshold (100 percent of the AMI) to receive the first-time homebuyer waiver.

Some purchase loans involve a waiver but do not qualify for a housing goal. Those loans are for first-time homebuyers whose income is above 80 percent of the AMI but at or below the income threshold (100 percent of the AMI) necessary for the waiver. For those purchase loans, the weighted average up-front fee will be approximately 92 basis points in fiscal year 2025, CBO estimates. (That fee is the weighted average of up-front fees for all purchase loans, some of which involve a waiver.) In CBO’s estimation, 60 percent of purchase loans do not meet any goal’s criteria and involve larger up-front fees. In those cases, the amounts vary depending on the characteristics of the loans: 37 percent of loans (including 32 percent of goal-eligible mortgages and 5 percent of non-goal-eligible ones) involve a waiver of 92 basis points and involve no up-front fee; and 3 percent (representing the goal-eligible mortgages that do not qualify for a waiver) involve the average up-front fee of 92 basis points, resulting in the average of 92 basis points for all mortgages.21

How Fee Waivers for Purchase Loans Are Determined

For the purpose of setting fees, Fannie Mae and Freddie Mac classify purchase loans in three distinct groups, in CBO’s estimation:

• The first group contains nearly all of the goal-eligible loans, most of which receive a waiver of up-front fees.
• The second group, which consists of what are sometimes called mission-remote loans, includes a very low percentage of goal-eligible loans and instead includes loans on investment properties and second homes and larger loans to support the purchase of more expensive homes.
• The third group, which consists of what are known as core loans, contains all other purchase loans—those that are generally not goal-eligible but are still important for supporting the GSEs’ role of providing liquidity to the mortgage markets. Core loans comprise the majority of the GSEs’ mortgage guarantees and include 30-year fixed-rate mortgages to assist moderate-income households in purchasing their primary residence.22

To meet their housing goals, the two GSEs increase up-front fees slightly on core purchase loans and implement larger increases to fees on mission-remote purchase loans. They then use the additional revenues from those higher fees to pay for the cost of fee waivers for loans in the first group. The waived fees amount to an implicit subsidy, and the increased fees amount to an implicit tax.

The amount of the implicit subsidy and tax varies within each group on the basis of the characteristics of their mortgages as defined by the GSEs’ up-front fee schedules. The amount of the subsidy is greater for purchase loans with higher up-front fees (as assigned by the fee schedules), such as loans for households with low down payments and credit scores. The amount of the implicit tax on mission-remote purchase loans and core purchase loans may differ, but that difference is more difficult to determine because the GSEs do not publish a fee schedule without the adjustments required to cover the subsidy associated with the waivers. Therefore, in the course of its analysis, CBO estimated the average implicit subsidy or tax for each of the three groups; the subsidy or tax for any given mortgage within each group might vary.

A review of data showed that households with goal-eligible mortgages would probably purchase a smaller property and thus require a smaller mortgage. Therefore, CBO used different average loan amounts for goal-eligible and non-goal-eligible mortgages in its analysis. The agency used an average loan amount of $248,000 for the 35 percent of purchase loans that will be goal-eligible in fiscal year 2025, and an average loan amount of $328,000 for the 65 percent of purchase loans that will not be goal-eligible that year.

Amounts of Implicit Subsidies and Taxes

CBO estimates that in fiscal year 2025, the two GSEs will waive up-front fees for 737,000 households purchasing a single-family home. Given an average up-front fee equal to 92 basis points and an average loan balance of $248,000, the waivers amount to an implicit subsidy of $2,282 per household (see Table 6). An additional 115,000 households purchasing a single-family home with a non-goal-eligible mortgage will each receive a subsidy of $3,018 (reflecting a fee waiver equaling 92 basis points) based on an average loan amount of $328,000. Those households represent first-time homebuyers whose income is above 80 percent of the AMI and at or below 100 percent of the AMI.

CBO also estimates that 60 percent of purchase loans will not be goal-eligible and will involve an implicit tax; 50 percent (1.2 million loans) are core loans, and 10 percent (230,000 loans) are mission-remote loans. Households with core loans will be charged an implicit tax averaging 5 basis points (for a total average up-front fee of 97 basis points)—which is consistent with other analysts’ estimates of the increase in fees charged on loans in that group.23

CBO estimates that in fiscal year 2025, mission-remote loans will involve an implicit tax averaging approximately 244 basis points (for a total average up-front fee of approximately 336 basis points). That tax is equal to the amount necessary for the higher fees on 60 percent of loans to cover the cost of the subsidies (averaging 92 basis points) for the 37 percent of loans receiving a waiver. Given an average balance of $328,000 for mission-remote loans and core loans (for which fees are not waived), the implicit tax on each loan in those groups will equal $7,990 and $164, respectively. (The remaining 3 percent of goal-eligible mortgages, totaling less than 70,000 loans, that do not involve a fee waiver will pay the average up-front fee of 92 basis points.)

In CBO’s analysis, the total amounts of the implicit subsidy and tax are approximately equal. That is because profit-maximizing companies would rationally react to a binding requirement such as the housing goals by raising the prices of non-goal-eligible mortgages and lowering the prices of goal-eligible mortgages in equal measure.

If the GSEs’ mortgage purchases are such that they are meeting the housing goals, then the GSEs must acquire approximately one goal-eligible mortgage for every two non-goal-eligible mortgages they acquire. Thus, acquiring two non-goal-eligible mortgages imposes the cost of extending the discount to one additional goal-eligible mortgage, and each non-goal-eligible mortgage results in roughly one-half of that cost. (In CBO’s analysis, 60 percent of purchase loans involve an implicit tax averaging $1,468 per loan, and 37 percent of purchase loans involve a subsidy averaging $2,381 per loan; thus, the ratio of tax to subsidy per loan is not exactly one-half.) A profit-maximizing company would raise the price of non-goal-eligible mortgages above the total of all other costs it faces (expected losses, market risk, and administrative costs) to cover the cost of meeting the goals; that higher price is an implicit tax. Thus, the implicit tax and the implicit subsidy fully offset one another because one is the result of the other.

Effects of Implicit Subsidies and Taxes on the Volume of Purchase Loan Guarantees

To estimate the change in the volume of purchase loans and the total dollar value of the implicit subsidy and tax stemming from the housing goals, CBO reviewed studies that estimate the elasticity of mortgage volume to price changes.24 (Elasticity is calculated as the change in the quantity of mortgages originated, divided by the change in the interest rate charged.) Those studies found that a change of 50 basis points in mortgage rates—equivalent to a 3 percent change in up-front fees—is associated with a 14 percent change in the volume of mortgages insured by FHA.

Although that elasticity was based on the behavior of borrowers with FHA loans, who tend to have lower income than borrowers with loans guaranteed by Fannie Mae or Freddie Mac, CBO expects that the response of demand to changes in price would be similar for both types of borrowers. The two GSEs’ higher-income borrowers might be less likely than FHA’s borrowers to be priced out of a mortgage because of the implicit tax. But the GSEs’ borrowers may have more options for nongovernment mortgages and therefore may be as sensitive to fee increases as FHA’s borrowers are, making the elasticities in the studies generally applicable.

Because of that similarity, CBO used the elasticity measure from the studies to estimate the change in the volume of purchase loans that would result from a subsidy amounting to 92 basis points on goal-eligible mortgages. The agency did the same for a tax amounting to either 5 basis points or 244 basis points on purchase loans not meeting the goals’ requirements.

That elasticity implies that for goal-eligible mortgages, the average reduction in price of 92 basis points would increase annual volume by roughly 4.3 percent, which would equal approximately 37,000 additional goal-eligible purchase loans’ being originated. An implicit tax amounting to 5 basis points and 244 basis points would decrease volume by about 0.2 percent and by about 11.4 percent, respectively, leading to roughly 29,000 fewer non-goal-eligible purchase loans than if the housing goals did not exist. Some of the additional goal-eligible mortgages may come from loans that would have been guaranteed by other federal programs (such as FHA) or that would not have been originated at all, absent the goals’ requirements. Alternatively, the non-goal-eligible loans no longer guaranteed by the GSEs may be guaranteed by those same federal programs, may be originated without a government guarantee, or may not be originated at all.

Effects of Housing Goals on Fannie Mae’s and Freddie Mac’s Operations

Although the federal government, through FHFA as conservator, retains control and effective ownership of Fannie Mae and Freddie Mac, there are repercussions if the GSEs fail to meet their housing goals. If FHFA determines that a GSE has failed to meet a housing goal, FHFA has discretionary authority to require the GSE to submit a housing plan describing the specific actions it will take to meet the goal. The GSE must submit that plan within 45 days, unless FHFA extends that deadline. FHFA then reviews the plan and either approves or disapproves the proposed actions. If FHFA disapproves the actions, the GSE must submit an amended plan within 15 days. If the amended plan is not approved, FHFA may allow the GSE an additional 15 days to submit a new plan. At that point, no further action is mandated by the applicable regulations.

The compensation of executives at Fannie Mae and Freddie Mac is partially determined by the GSEs’ performance in meeting the goals. For example, a portion of executive compensation is determined by FHFA, which bases its assessment on an evaluation of performance in what is known as a conservatorship scorecard.25 That scorecard includes criteria related to the GSEs’ support for affordable housing, including their housing goals.

Effects of Housing Goals on CBO’s Baseline Budget Projections

In CBO’s judgment, Fannie Mae and Freddie Mac are effectively part of the federal government while operating in their conservatorships. Thus, in its baseline budget projections for the coming 10 years, CBO accounts for the two GSEs’ operations as though they are being conducted by a federal agency. CBO measures the cost of the GSEs’ mortgage guarantees on a fair-value basis by using the equivalent of market prices for those guarantees. (The fair value of a liability, such as a loan guarantee, is the price that would have to be paid to induce a private financial institution to assume the liability.)26

The effects of the GSEs’ housing goals on CBO’s budget baseline are difficult to determine. In general, the structure of the up-front and ongoing fees the GSEs use to achieve their housing goals—in which fee waivers for goal-eligible mortgages are recouped through higher fees on non-goal-eligible mortgages—leaves average fees roughly the same across the entire portfolio of loan guarantees. As such, the effect on CBO’s estimate of the subsidy rate is probably close to zero. (The subsidy rate is defined as the present value of the lifetime cost of credit losses net of recoveries and fees, expressed as a percentage of the original loan amounts.)27

Because Fannie Mae and Freddie Mac use an implicit subsidy and tax to achieve their housing goals, the portfolio of loans that they guarantee differs from what it would be if the goals did not exist. That difference changes expectations about the number of households that will default on their mortgages. In general, mortgage defaults increase as borrowers’ income (in relation to the AMI) decreases.28 Although loans that meet the housing goals’ requirements are likely to default at higher rates than loans that do not, the GSEs and FHFA mitigate the potential impact of those higher defaults when setting criteria for underwriting the mortgages that the GSEs purchase. In CBO’s assessment, the existence of those underwriting criteria limits the potential increase in default rates attributable to the greater number of lower-income borrowers that results from implementing the implicit subsidy and tax.

---

1. 1. Those charters specify that in activities related to low- and moderate-income households, Fannie Mae and Freddie Mac may earn a “reasonable economic return that may be less than the return earned on other activities.” That provision provides a basis for subsidizing the goal-eligible mortgages described in this report. See Federal Housing Finance Agency, About Fannie Mae and Freddie Mac (accessed February 2, 2024), www.fhfa.gov/about-fannie-mae-freddie-mac.

2. 2. For details, see Congressional Budget Office, Accounting for Fannie Mae and Freddie Mac in the Federal Budget (September 2018), www.cbo.gov/publication/54475.

3. 3. Federal Housing Finance Agency, “Fannie Mae and Freddie Mac Affordable Housing Goals” (August 22, 2024), https://tinyurl.com/32ahuwkz. According to the usual definition of affordability, homeowners or renters would spend no more than 30 percent of their income on housing costs. The single-family goals are based on household income without regard to the affordability of the housing payment. For the multifamily goals, however, the 30 percent rule is required by statute (12 U.S.C. §4563(c)).

4. 4. A low-income census tract is one in which median income is no greater than 80 percent of the AMI. A minority census tract is one in which at least 30 percent of the population consists of people in racial or ethnic minority groups and in which median income is less than 100 percent of the AMI.

5. 5. Individual mortgages that count toward more than one goal are more valuable to the GSEs than mortgages meeting only one goal. In this report, CBO does not estimate the relative value of mortgages that meet more than one goal.

6. 6. FHFA uses the market level to determine the extent to which the GSEs’ purchases of goal-eligible mortgages meet the share of mortgages originated in the entire market (including loans not sold to the GSEs). For more information about FHFA’s method of estimating the market size for single-family housing goals, see Federal Housing Finance Agency, The Size of the Affordable Mortgage Market: 2022–2024 Enterprise Single-Family Housing Goals (December 2021), https://tinyurl.com/294yzxsb.

7. 7. Federal Housing Finance Agency, “Housing Goals Performance” (October 27, 2023), https://tinyurl.com/y4mxu33t.

8. 8. Federal Housing Finance Agency, The Size of the Affordable Mortgage Market: 2022–2024 Enterprise Single-Family Housing Goals (December 2021), https://tinyurl.com/294yzxsb.

9. 9. 2023–2024 Multifamily Enterprise Housing Goals, 87 Fed. Reg. 50794–50803 (proposed August 18, 2022), https://tinyurl.com/a2wnc492.

10. 10. 2023–2024 Multifamily Enterprise Housing Goals, 87 Fed. Reg. 78837, 78837–78846 (December 23, 2022), https://tinyurl.com/2r4dj7vy.

11. 11. In addition to adjusting their guarantee fees to achieve their housing goals, the GSEs charge two other guarantee fees not directly related to covering their costs. First, they assess a 0.042 percent up-front fee on all new guarantees to cover required contributions to the Housing Trust Fund and the Capital Magnet Fund. For more information, see Congressional Budget Office, How the Housing Trust Fund and Capital Magnet Fund Support Affordable Housing (November 2022), www.cbo.gov/publication/58427. Second, as required by the Infrastructure Investment and Jobs Act, they assess a 0.10 percent ongoing fee on all new loan guarantees, to be paid to the Treasury Department.

12. 12. Although the GSEs are structured as for-profit corporations with a fiduciary responsibility to their shareholders, they probably have other objectives besides maximizing profits. First, under their conservatorships, the GSEs did not pay dividends or otherwise make payments to those shareholders—with any profits instead directed to the Treasury—potentially resulting in reduced concern for their profitability. In 2019, they began to retain earnings in preparation for a potential plan to recapitalize the GSEs and release them from conservatorship, which may have increased their focus on profit maximization. Second, their charters mandate that they promote mortgage credit nationwide and maintain stability and liquidity in the secondary market in addition to earning a reasonable return on their activities. Third, compensation for some of Fannie Mae’s and Freddie Mac’s executives depends, in part, on whether the GSEs meet the housing goals. As a result, in CBO’s assessment, the GSEs are incentivized to meet the housing goals and do so in a way that minimizes costs.

13. 13. For a description of the GSEs’ capital requirements, see Congressional Budget Office, Effects of Recapitalizing Fannie Mae and Freddie Mac Through Administrative Actions (August 2020), www.cbo.gov/publication/56496.

14. 14. Federal Housing Finance Agency, Fannie Mae and Freddie Mac Single-Family Guarantee Fees in 2022 (May 2024), pp. 5, 13–14, https://tinyurl.com/mpeykhv4. According to FHFA’s report, “For the purposes of reporting to FHFA, the Enterprises annualize up-front fees by dividing the up-front fee for a given loan by that loan’s specific present value multiplier (PVM). For example, a loan with an up-front fee of 72 basis points and a PVM of 6 would have an annualized up-front fee of 72/6 = 12 basis points. PVMs are modeled by the Enterprises, based on interest rates and loan characteristics.” The PVM is a measure of the expected average time a mortgage is outstanding. Higher PVMs reflect the expectation of a longer period between the time the mortgage is originated and the time it is fully repaid. For this report, all annualized up-front fees are converted back to the up-front equivalent by multiplying the reported annualized up-front fee by a PVM of approximately 5.75.

15. 15. Financial Crisis Inquiry Commission, The Financial Crisis Inquiry Report: Final Report of the National Commission on the Causes of the Financial and Economic Crisis in the United States (February 2011), Chapter 9, pp. 185–187, www.govinfo.gov/app/details/GPO-FCIC.

16. 16. The fee waivers were among several changes made to the up-front fees as a part of that announcement. In general, the changes were made to better align up-front fees with the GSEs’ new capital framework. See Fannie Mae, “Lender Letter (LL-2023-01): New Loan-Level Price Adjustment Framework” (March 22, 2023), https://tinyurl.com/59xnttxr; and Freddie Mac, “Update to Credit Fees and Exhibit 19 Redesign” (January 19, 2023), https://tinyurl.com/muxrxks4.

17. 17. HomeReady and Home Possible are programs designed to provide affordable mortgages for low-income borrowers with a history of timely repayment of their debt obligations. For more information, see Fannie Mae, “HomeReady Mortgage” (accessed August 19, 2024), https://tinyurl.com/w32v4e65; and Freddie Mac, “Home Possible” (accessed August 19, 2024), https://tinyurl.com/yc3zhws7.

18. 18. In addition to offering discounts on fees for mortgages that meet the goals’ criteria, the GSEs could limit the purchase of mortgages that do not meet the requirements of one or more goals. That strategy is often referred to as “denominator management” because it reduces the number of mortgages in the calculation of the percentage of housing-goals loans in relation to the number of total loans. Denominator management has the benefit of reducing the explicit concessions that the GSEs make for goal-eligible mortgages. However, it reduces the amount of the GSEs’ loan guarantees (which could reduce revenues and net income) and may leave lenders with fewer options for selling their mortgages in the secondary market.

19. 19. Fannie Mae and Freddie Mac have other programs designed to provide refinance options to households with affordability concerns—RefiNow and Refi Possible, respectively. However, those programs do not offer the same up-front fee waivers that the HomeReady and Home Possible programs offer.

20. 20. CBO estimates that 70 percent of all mortgages purchased by the GSEs in fiscal year 2025 will be purchase loans and that 30 percent will be refinance loans. Of those refinance loans, 26 percent will meet the low-income refinance goal.

21. 21. CBO estimates that goal-eligible purchase mortgages that do not receive a full waiver will pay the average up-front fee (92 basis points) rather than the up-front fee charged to mortgages that do not meet a goal criterion. That estimate is based on the expectation that the GSEs have calibrated other up-front fees to mitigate the implicit tax on goal-eligible purchase mortgages that do not receive a waiver.

22. 22. For a more detailed description of the groups of purchase loans, see Edward Golding and others, How to Think About Fannie Mae and Freddie Mac Pricing (Urban Institute, August 2023), https://tinyurl.com/dncmy4a4.

23. 23. Ibid.

24. 24. See Neil Bhutta, “GSE Activity and Mortgage Supply in Lower-Income and Minority Neighborhoods: The Effect of the Affordable Housing Goals,” Journal of Real Estate Finance and Economics, vol. 45 (2012), pp. 238–261, https://doi.org/10.1007/s11146-010-9258-z; and Neil Bhutta and Daniel Ringo, “The Effect of Interest Rates on Home Buying: Evidence From a Shock to Mortgage Insurance Premiums,” Journal of Monetary Economics, vol. 118 (March 2021), pp. 195–211, https://doi.org/10.1016/j.jmoneco.2020.10.001.

25. 25. Federal Housing Finance Agency, “Conservatorship” (February 6, 2024), www.fhfa.gov/conservatorship. For more information about executive compensation, see Federal Housing Finance Agency, “Executive Compensation” (August 28, 2023), www.fhfa.gov/policy/executive-compensation.

26. 26. For details, see Congressional Budget Office, Accounting for Fannie Mae and Freddie Mac in the Federal Budget (September 2018), www.cbo.gov/publication/54475.

27. 27. For a description of how CBO calculates the subsidy rate for federal mortgage guarantee programs, see Congressional Budget Office, “Modeling the Subsidy Rate for Federal Single-Family Mortgage Insurance Programs” (January 2018), www.cbo.gov/publication/53402.

28. 28. Hamilton Fout and others, “Credit Risk of Low Income Mortgages,” Regional Science and Urban Economics, vol. 80 (January 2020), https://doi.org/10.1016/j.regsciurbeco.2018.07.013.

About This Document

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

Michael Falkenheim and Mitchell Remy wrote the report with guidance from Sebastien Gay. Justin Humphrey, Wendy Kiska, Kyoung Mook Lim, Shannon Mok, Zunara Naeem, and Natalie Tawil (formerly of CBO) contributed to the analysis and offered comments. Byoung Hark Yoo fact-checked the report.

Neil Bhutta of the Federal Reserve Bank of Philadelphia, Edward DeMarco of the Housing Policy Council, and Jim Parrott of the Urban Institute 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, Scott Craver edited it, and R. L. Rebach created the tables and prepared the text for publication. The report is available on CBO’s website at www.cbo.gov/publication/60190.

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

Phillip L. Swagel

Director

November 2024