Seven Things to Know About CBO’s Budgetary Treatment of Potential Changes to Fannie Mae and Freddie Mac

Notes

Unless this report indicates otherwise, all years referred to are federal fiscal years, which run from October 1 to September 30 and are designated by the calendar year in which they end. Numbers in the text and tables may not add up to totals because of rounding.

Fannie Mae and Freddie Mac were chartered in 1938 and 1970, respectively, as government-sponsored enterprises (GSEs) to ensure a stable supply of credit for mortgages nationwide. Government-sponsored enterprises are private companies created by federal law to fulfill a specific purpose. In the case of Fannie Mae and Freddie Mac, that purpose is to facilitate the flow of funding for home loans by purchasing mortgages from lenders, pooling them into mortgage-backed securities (MBSs), and selling the securities to investors along with a guarantee against most losses from defaults on the underlying loans.

After operating independently for decades, the two GSEs were placed in federal conservatorships in 2008. Since then, they have been controlled by the Federal Housing Finance Agency (FHFA) and effectively owned by the Department of the Treasury (for details, see Box 1). In January 2025, the FHFA announced that it will seek comments on options to end the GSEs’ conservatorships.1

This report addresses seven key issues that might arise as the Congressional Budget Office estimates the budgetary effects of potential legislation or administrative actions that could result in selling the Treasury’s ownership stake in the GSEs and releasing them from government control. In keeping with its standard practices, CBO assesses the federal budgetary and economic effects of proposed policies but does not make policy recommendations.

1. CBO projects that the GSEs’ future mortgage guarantees have a budgetary cost

An MBS created by Fannie Mae or Freddie Mac entitles its purchaser to a share of the combined mortgage payments from the pool of loans underlying the security. The GSEs guarantee that such payments will be made in full and on a timely basis—including covering missed payments if borrowers default so MBS investors are not affected. In return for providing that guarantee of payment, the GSEs receive an up-front fee when an MBS is created and a yearly fee based on the outstanding principal of the underlying mortgages. The GSEs’ net income results mainly from the difference between the fees they collect and the payments they make on defaulted loans.

In most years—including every year since 2012—the amount that Fannie Mae and Freddie Mac collect in fees exceeds their costs to cover mortgage defaults. But during periods of financial stress, such as the housing crisis of the late 2000s, the cost of defaults can exceed income from fees, resulting in net losses for the GSEs. To manage those situations, the GSEs maintain reserves that they fund by selling shares of stock or by keeping earnings that would otherwise be paid to their shareholders as dividends. Those reserves are invested in liquid assets that they can sell quickly to meet payment obligations.

CBO estimates that the new mortgages Fannie Mae and Freddie Mac are projected to guarantee during the 2026−2035 period would have a total cost to the federal government of $81.8 billion.2 That baseline estimate reflects CBO’s projections of the default payments and fees that would occur if current laws governing revenues and spending generally remained unchanged over the next 10 years. (CBO estimated that cost on a fair-value basis, which measures the market value of the GSEs’ mortgage guarantees by accounting for market risk. For details, see section 4 below.)

Fannie Mae’s and Freddie Mac’s guarantees have a budgetary cost because, in CBO’s estimation, the GSEs charge lower fees for their guarantees than private-sector investors would charge to assume the same risk. In CBO’s baseline projections, that underpricing averages about 6 to 10 basis points (0.06 to 0.10 percentage points) in each year of the 2026−2035 period.3 The two GSEs are projected to issue a total of about $15 trillion in new guarantees during that period. With such large portfolios of mortgages, a modest underpricing of risk can result in a budgetary cost of tens of billions of dollars.

2. CBO treats the GSEs as if they were part of the government

In CBO’s assessment, Fannie Mae and Freddie Mac are effectively part of the federal government for two reasons.4 First, the FHFA gained operational control of the GSEs when it put them in conservatorships. That control was evident, for example, in March 2025, when the FHFA changed the membership of the GSEs’ boards of directors. Second, the federal government gained majority ownership of the GSEs when the Treasury acquired senior preferred shares in the enterprises—and warrants to buy their common shares in the future—in exchange for agreeing to maintain their solvency (see Box 1). For those reasons, CBO’s baseline budget projections account for the GSEs’ operations as though they are being conducted by a federal agency.

Because CBO treats Fannie Mae and Freddie Mac as part of the federal government, it considers the costs of their mortgage guarantees to be federal costs, which are projected to increase the overall budget deficit. In the case of dividends that the GSEs sometimes pay to the Treasury on its shares, CBO treats those transactions as occurring between federal entities rather than between the public and the government (see Figure 1, top panel).5 Thus, CBO does not count those payments as income to the government and does not project that they affect the overall deficit.

Figure 1.

Cash Flows to and From Fannie Mae and Freddie Mac Before and After Recapitalization and Release From Federal Control

Because the GSEs are in federal conservatorships and largely owned by the Treasury, CBO considers them part of the federal government. Thus, the dividends they pay to the Treasury on its shares are considered intragovernmental payments that have no net effect on the federal budget.

If the GSEs were released from federal control, CBO would consider them nongovernmental entities, and their financial interactions with the government would have budgetary savings or costs.

Notes

Data source: Congressional Budget Office.

GSEs = government-sponsored enterprises (specifically, Fannie Mae and Freddie Mac).

a. The current solvency guarantee is the Treasury’s commitment to keep Fannie Mae and Freddie Mac afloat financially by purchasing more of their senior preferred shares whenever their net worth falls below zero. Policy changes could restructure that guarantee or make it implicit.

b. Indicates cash flow between the federal government and the public.

c. The federal credit subsidy for guaranteeing mortgage-backed securities is the difference between the present value of losses from defaults (net of recoveries) that are expected to occur over the lifetime of those securities and the present value of fees that are expected to be collected for those guarantees over their lifetime. (A present value is a single number that expresses a flow of current and future income or payments in terms of an equivalent lump sum received or paid at a specific time, such as the present.)

If the GSEs began paying dividends to their private shareholders while remaining under federal control, CBO would treat those payments as outlays from the government to the private sector, which would increase federal spending and the deficit. Under the Administration’s current policies, the GSEs do not make dividend payments to their private shareholders and are not expected to start doing so. If that changed—whether through legislation, administrative actions, or judicial decisions—CBO’s baseline would project an increase in the deficit.

Unlike CBO, the Administration’s Office of Management and Budget (OMB) treats Fannie Mae and Freddie Mac as nongovernmental entities for budgetary purposes. OMB estimates that cash transactions between the GSEs and the Treasury affect the deficit. For example, OMB treated investments that the Treasury made in the GSEs from 2008 to 2012 as federal costs and dividends that the GSEs paid to the Treasury from 2012 to 2019 as federal income.

Also unlike CBO, OMB treats the GSEs’ mortgage guarantees as private-sector transactions that do not have an immediate budgetary impact. The budgetary effects associated with those guarantees occur in future years and depend on the GSEs’ net income. If their net income is positive and the GSEs pay dividends to the Treasury, OMB will record those dividends as receipts in the budget. If the GSEs’ net income is negative, they will not make dividend payments to the Treasury, and if their losses are severe, they may require additional federal investment because of the Treasury’s commitment to ensure their solvency.

3. Releasing the GSEs from government control would result in both federal savings and federal costs

Legislation or administrative actions related to rebuilding the capital reserves of the GSEs and releasing them from their conservatorships could end federal control and ownership of the entities. That change would affect the federal budget in various ways over the typical 10-year projection period of CBO’s baseline and beyond. The budgetary effects would depend on the details of the recapitalization and release policies. As an illustration, the appendix examines two hypothetical policies: one in which the Treasury would convert its senior preferred shares in the GSEs to common shares and one in which the FHFA would place the GSEs in receivership (a condition that would involve having their assets transferred to other entities).

The first policy would begin with the Treasury converting its senior preferred stock to common stock and exercising its warrants to buy additional common stock. After that, Fannie Mae and Freddie Mac would sell more common shares to the public to raise cash to help them meet the FHFA’s capital requirements, a precondition for being released from their conservatorships (see Box 1). After the GSEs were released, the Treasury would sell its holdings of their common stock to the public over time.

In the second policy, the FHFA would end the conservatorships by putting Fannie Mae and Freddie Mac in receivership, creating a new corporation to replace each GSE, and transferring the GSEs’ assets and liabilities to those entities. The new corporations would sell common stock to the public and keep some of the proceeds from those sales to meet the FHFA’s capital requirements. The remaining proceeds would be allocated to the former GSEs’ shareholders on the basis of their seniority.6

Sources of Budgetary Savings From Recapitalization and Release

Ending federal control of Fannie Mae and Freddie Mac would produce budgetary savings, primarily because it would eliminate federal subsidies for future mortgage guarantees that are priced below market rates, in CBO’s estimation. Other sources of budgetary savings would be receipts from selling the Treasury’s shares to private investors, which would be offset by a corresponding cost from reductions in future profits, and receipts from fees paid by the GSEs to federal affordable housing funds.

  • Elimination of subsidies on future mortgage guarantees. Under either of CBO’s illustrative scenarios, after the GSEs’ release, their future mortgage guarantees would be private transactions rather than government commitments. The federal government would no longer be responsible for guaranteeing MBSs at prices below their market value. That change would reduce the deficit by eliminating the budgetary cost that CBO assigns to future guarantees in its baseline projections.
  • Payments to the Treasury. Under either of CBO’s illustrative scenarios, recapitalization and release of Fannie Mae and Freddie Mac would involve the Treasury selling its shares in the GSEs to private investors. Under either scenario, the payments from private investors in the newly independent GSEs to the Treasury would reduce the deficit by bringing in receipts to the federal government. The size of the payments would depend on the details of the policy. CBO would recognize the significant cash inflows that the government would receive from the sale of shares in the newly independent GSEs. In such a market transaction, the amount the Treasury would receive for those shares would match the market valuation of the GSEs’ reduced future profits, resulting in no net budgetary effect, in CBO’s assessment.
  • Reclassification of affordable housing fees. The fees that the GSEs pay to the Housing Trust Fund and the Capital Magnet Fund to increase and preserve the supply of affordable housing—currently classified as intragovernmental transfers—would be reclassified as receipts to the government (see Figure 1, bottom panel).7 As government receipts, they would reduce the deficit.

Sources of Budgetary Costs From Recapitalization and Release

Ending federal control of the GSEs would also create costs. For example, the government would no longer receive the income that Fannie Mae and Freddie Mac earn from their existing mortgage guarantees and from their portfolios of assets. (Existing mortgage guarantees, those in place before the GSEs’ release, are projected to generate net income for the government in future years because the risk of default by borrowers tends to decline over time, while guarantee fees continue to be paid at a fixed rate. In contrast, future mortgage guarantees, those made after the GSEs’ release, would no longer be backed by the government or be included in the federal budget.) Another possible source of budgetary costs is that any commitments by the Treasury to maintain the solvency of the GSEs would no longer be considered intragovernmental transfers. The effects of those costs on the deficit would be largely offset by the payments to the Treasury for the value of its shares in the GSEs.

  • Loss of income from existing mortgage guarantees. In CBO’s assessment, the GSEs’ mortgage guarantees create costs for the government in the early years of repayment on the loans (when default rates tend to be highest) and create savings in later years (when default rates tend to decline). Having incurred costs in the early years of existing mortgage guarantees, on average, the government is projected to benefit from the savings on those existing guarantees in later years, under current law. CBO incorporated all of the projected lifetime costs and savings associated with outstanding mortgage guarantees in its budget estimates when the guarantees were made, consistent with its treatment of other federal credit programs (see the next section for details). Any changes to those cash flows would be reflected in a cost estimate for proposed legislation. If the GSEs were recapitalized and released, the income from existing guarantees would go to the GSEs’ private shareholders rather than to the government, a change that would increase the deficit.
  • Loss of income from other assets. Ending federal control would cause the government to lose net income from the GSEs’ other assets as well as from its mortgage guarantees. In 2019 and 2020, the Treasury changed its agreements with the GSEs to allow them to keep their earnings rather than pay dividends to the Treasury. Those earnings have been invested in a mix of assets that earn interest and other forms of income. Under CBO’s budgetary treatment, that income is received by the government. If the GSEs were reclassified as private entities, the government would no longer earn that income, a change that would increase the deficit.8
  • The Treasury’s commitments to keep the GSEs solvent. The Treasury’s commitments to maintain the solvency of the GSEs are currently viewed as intragovernmental and thus as not having budgetary costs. If federal control of the GSEs ended, such commitments by the Treasury would be reclassified as transactions with private entities and would result in costs. Under CBO’s illustrative scenarios, the Treasury would continue to guarantee the GSEs’ solvency but would charge a commitment fee (see the appendix). Depending on its size, such a fee could partly or wholly offset the cost of the commitment. If the Treasury charged a fee that was less than the cost of its backstop commitment, it would be providing a federal subsidy. If the Treasury charged more for its backstop than private investors would charge to provide a similar amount of capital support, the federal backstop would reduce the deficit.

Net Budgetary Effects

The budgetary savings from eliminating federal subsidies on the GSEs’ future mortgage guarantees would be largely offset by costs from recapitalization and release under a wide range of policy approaches. The reason is that the amount investors would pay for the Treasury’s shares in the GSEs would depend on the same policies that would determine the size of all other savings and costs. For example, investors would pay for the GSEs’ existing mortgage guarantees and other assets, and that payment would be shown in the budget as a savings. But after the sale of the Treasury’s ownership stake, the government would no longer receive income from those assets. Those two budgetary effects from recapitalization and release—receipts from selling the government’s shares and forgone revenues from the GSEs’ assets—would generally be estimated as offsetting each other.

Similarly, the Treasury’s commitments involve costs, but they increase the projected profits of the GSEs. Those higher projected profits lead to a higher sale price for the Treasury’s shares in the GSEs. Eliminating the Treasury’s commitments would reduce costs and the profitability of the GSEs, reducing that sale price. Requiring the GSEs to pay fees to the Treasury for its solvency commitment would generate budgetary savings but also reduce the GSEs’ profitability and the sale price for the Treasury’s shares.

4. CBO projects the budgetary effects of the GSEs’ mortgage guarantees on a fair-value basis

Since 2008, CBO, in consultation with the House and Senate Committees on the Budget, has used fair-value accounting to measure the cost of the GSEs’ new mortgage guarantees. That approach estimates the market prices that private entities without a backstop from the Treasury would charge to make the same mortgage guarantees. (The fair value of a liability, such as a loan guarantee, is the price that would have to be paid in an orderly market to induce a private financial institution to assume the liability.)

Fair-value estimates are calculated on an accrual basis, meaning that the full costs projected to result from a cohort of mortgage guarantees over their lifetime are recorded in the year in which the guarantees are made. That approach differs from the cash basis used for most federal programs in the budget. Cash-based accounting records a program’s spending and revenues in the years when they occur.

The fair-value accounting that CBO uses for the GSEs in conservatorship also differs from the approach used for other federal programs that guarantee mortgages. Those programs are accounted for using the procedures specified in the Federal Credit Reform Act of 1990 (FCRA). Both FCRA and fair-value estimates are accrual measures, but fair-value estimates incorporate market risk—the risk that adverse movements in overall market conditions will reduce the value of a financial asset.9 FCRA estimates do not incorporate market risk.

CBO projects that Fannie Mae and Freddie Mac will issue a total of $15.2 trillion in new loan guarantees from 2026 to 2035. In CBO’s January 2025 baseline, those guarantees are projected to have a total cost to the government of $81.8 billion on a fair-value basis.10 That cost occurs because the GSEs charge slightly lower fees than fully private insurers would charge for the same guarantees, in CBO’s view. As a result, on a fair-value basis, the net present value of the guarantee fees that the GSEs charge is less than the net present value of the guarantee obligations for which they are liable.11 Ending federal control of the GSEs would eliminate that cost from the budget.

CBO’s baseline incorporates the projected lifetime costs or savings to the federal government from the GSEs’ guarantees and assets when the guarantees are made or the assets are acquired, rather than when their various cash flows take place. If any legislation or administrative actions changed the cash flows associated with existing guarantees or assets, CBO would estimate how those changes would affect the deficit on a fair-value basis.

Specifically, CBO would calculate the net present value of the future cash flows for existing guarantees or assets under both current law and the proposed change and would record the difference between those values as spending or revenues in the year the action modifying the guarantees or assets occurred. For example, if legislation would require the GSEs to modify the terms of mortgages in a way that increased the cost of their existing guarantees, that legislation would have a budgetary cost that would be recorded in the year the modification would take place.

5. CBO’s baseline budget projections and cost estimates incorporate administrative and judicial actions

CBO’s cost estimates compare the projected budgetary effects of proposed legislation with a baseline that incorporates the effects of recent administrative and judicial actions, including any actions that affect Fannie Mae and Freddie Mac.12 When CBO accounts for recent administrative and judicial actions in its baseline and cost estimates, it must assess the likelihood, timing, and resulting budgetary impact of those actions.

If proposed legislation would require the FHFA and the Treasury to take certain actions, CBO’s estimate of the legislation’s costs would depend on whether the budgetary effects of those actions were already included in the baseline:

  • If the Administration had not yet taken those actions, the legislation would affect the deficit.
  • If the Administration had already taken those actions in a clear, public, and official way, CBO might estimate that the legislation would have no effect on the deficit.
  • If the Administration had proposed those actions—for example, by publishing a notice of proposed rulemaking in the Federal Register or announcing a specific plan—CBO’s baseline would incorporate 50 percent of the budgetary effects, and CBO would estimate that the legislation would have the remaining 50 percent of those effects. That treatment, which is consistent with how CBO treats other proposed rules, reflects uncertainty about whether a proposed rule will be finalized and what form the final rule might take. (Restructuring the Treasury’s shares in the GSEs and ending the conservatorships would not necessarily require proposing a new rule.)

6. CBO would incorporate the cost of the Treasury’s implicit or explicit commitments to the GSEs in cost estimates

If Fannie Mae and Freddie Mac were released from federal ownership, they might still benefit from a federal guarantee of their solvency, whether implicit or explicit. Before their conservatorships, the GSEs benefited from an implicit federal guarantee.13 Investors who bought mortgage-backed securities from the GSEs were willing to accept lower yields on them because they perceived that the government would make sure they received payment in a timely fashion if the GSEs faced insolvency—even though at that time any assistance required a future act of Congress. That perception turned out to be accurate. When the GSEs’ financial condition deteriorated in 2008, lawmakers enacted the Housing and Economic Recovery Act, which authorized the Treasury to provide financial assistance to the GSEs.

Once they were released from federal ownership, Fannie Mae and Freddie Mac could benefit from an explicit commitment if, for example, the commitments in their 2008 agreements with the Treasury remained in effect or if those commitments were replaced by another mechanism. The current commitment requires the Treasury to invest in the GSEs when their net worth falls below zero. In the past, Members of Congress have proposed different types of commitments, such as guaranteeing to cover investors against losses on MBSs above a threshold level that would be likely to occur only in catastrophic circumstances rather than guaranteeing the solvency of the GSEs themselves. Without an explicit commitment, the GSEs would probably benefit from an implicit federal guarantee—especially if they continued to be linchpins of the housing finance system and retained the advantages of their federal charters (such as exemption from state and local taxes).

If CBO no longer considered the GSEs to be government entities, it would incorporate the estimated cost of any explicit federal guarantee in its future cost estimates and baseline projections. To estimate that cost, CBO would assess the risk associated with the GSEs’ financial exposure and consider any private capital that would serve as a buffer to absorb losses before the federal guarantee came into play. The way in which CBO would measure the cost of the guarantee—whether it would use cash or accrual estimates and whether it would incorporate market risk—would depend on the nature of the commitment.

Before the GSEs were placed in conservatorships, CBO did not incorporate their implicit federal guarantee in its budget estimates. But the events of 2008 provide support for recording the cost of such a guarantee if the GSEs were released from government control without an explicit federal backstop. CBO would look to provide the Congress with comprehensive information about the estimated cost of an implicit guarantee. In particular, CBO would consult with the House and Senate Budget Committees about how to present information about the cost of the implicit guarantee along with, or as part of, its future baseline projections.

7. CBO’s estimate of the proceeds from selling the Treasury’s shares in the GSEs would depend on many factors

CBO’s projection of how much the Treasury would earn from selling its financial stake in Fannie Mae and Freddie Mac would depend on the details of the proposed policy. Some of the most important factors include possible changes in the GSEs’ obligations that would reduce the value of the Treasury’s shares, policy changes that could affect the potential profitability of the GSEs’ mortgage guarantees and other assets, and the extent of any federal guarantee of the solvency of the GSEs as private companies.

The Value of the Treasury’s Shares

Legislation or further actions by the Treasury could reduce the GSEs’ obligations to the Treasury. Such a change would allow the GSEs to raise more capital and thereby leave their conservatorships sooner, but it would also reduce the value of the Treasury’s shares.

For example, decreasing the seniority of the Treasury’s ownership stake in the GSEs would enable them to raise funds to meet their capital requirements by selling new common stock. To encourage private investors to buy the new shares, the GSEs would need to be able to pay dividends on them in the foreseeable future. But the current ownership structure of Fannie Mae and Freddie Mac makes that difficult. Today, the GSEs do not pay dividends on any of their shares. If they began doing so, the senior preferred stock held by the Treasury would have the highest priority for those payments. Outstanding junior preferred stock in the GSEs, which is held by private investors, would have the next highest priority for receiving dividends (see the appendix). Common stock would have the lowest priority. As long as the Treasury’s senior preferred shares remained outstanding and the GSEs’ profits were not sufficient to pay dividends on all preferred stock, holders of common shares would receive no dividends, and investors might be unwilling to purchase such shares.

Various ways exist to reduce the seniority of the Treasury’s claim on the GSEs to enable them to sell common stock more easily. In a 2020 report and a 2024 update, CBO analyzed a scenario in which the Treasury would modify its agreements with the GSEs so they could buy back the Treasury’s senior preferred shares at the face (par) value.14 According to the central estimates in that update, if the GSEs were recapitalized and released in 2027, the Treasury would need to accept less than the par value of its preferred shares to allow the GSEs to raise the funds they would need from new investors.

The Treasury could modify its shares in other ways, such as by converting its senior preferred stock to common stock—as it did when resolving the bankruptcy of the insurance and financial company AIG after the 2007−2009 financial crisis.15 Reducing the seniority of the government’s claim in that way would decrease the value of the Treasury’s shares and increase the value of outstanding preferred and common shares held by private investors. As an alternative, placing Fannie Mae and Freddie Mac in receivership could avoid lowering the value of the Treasury’s shares, but it would involve other trade-offs (see the appendix).

Mortgage Guarantees and Other Assets

CBO’s estimate of the market value of the GSEs would depend heavily on its estimate of the future profitability of their core business—guaranteeing mortgage-backed securities for a fee. Because CBO estimates that the GSEs charge slightly lower fees for their new MBS guarantees at origination than a private investor would charge, every new cohort of MBSs they originate subtracts value from the GSEs rather than adding value. By contrast, the GSEs’ portfolios of outstanding MBSs are expected to generate positive net income because they tend to face less default risk over time than they did at origination. Moreover, the GSEs might raise their guarantee fees if they were released from federal control. In addition to mortgage guarantees, the GSEs’ other assets—such as cash held in case of future losses—produce interest and other income for the enterprises.

Solvency Guarantee

The price at which the Treasury could sell its shares in the GSEs would depend on any guarantee of future solvency that the GSEs might receive from the government, net of the cost that the government would charge for that guarantee. If everything else was unchanged, the sale price would be higher if the GSEs received a guarantee and lower if they had to pay a fee for it. The net effect would depend on whether the fee was higher or lower than the value of the guarantee.

If the government charged less for its guarantee than a private firm would, that underpricing would make the GSEs more profitable, and their stock would be likely to sell at a higher price. In addition, anyone who lent money to the GSEs, including buyers of their corporate bonds, would require lower interest rates if the government stood behind the two firms. And private investors would pay more for MBSs guaranteed by the GSEs if those securities had a government backstop. If everything else was unchanged, those lower costs of funding and greater proceeds from the sale of MBSs would translate into higher profits for the GSEs and thus a higher price for their common stock.

Other Factors

Various other factors could affect CBO’s projection of the proceeds from selling the Treasury’s shares. If lawmakers introduced competition for the GSEs—such as by issuing charters to other enterprises to compete with Fannie Mae and Freddie Mac—CBO would assess that mortgage interest rates would be lower than they would be otherwise but that proceeds from the sale would also be lower. If the GSEs were required to allocate a large portion of their resources to buying or guaranteeing mortgages from borrowers with low income, or if they were restricted from offering new financial products, their future profitability would be lower, as would the sale price. Conversely, if restrictions on risk-taking by the GSEs—such as limits on the size of their retained portfolios or down payment requirements for the mortgages they were allowed to buy—were relaxed, the sale price would be higher.

When estimating the cost of legislation that would involve selling the Treasury’s shares in the GSEs, CBO would first consider whether the proceeds from the sale should be included in the cost estimate. In keeping with the Congress’s scorekeeping guidelines for asset sales, CBO makes such a decision by comparing the expected sale price of an asset with the value of the future income the government would derive from the asset if it kept it. Then, if the projected sale price is at least as much as the present value of that future income, CBO includes the proceeds from the sale in its cost estimate. If the sale price is less than the present value of that future income, CBO does not include the proceeds in its cost estimate. In the case of selling the Treasury’s shares in Fannie Mae and Freddie Mac, CBO expects that the proceeds would meet the required threshold as long as the shares were sold through a competitive process.

Appendix: Illustrative Policies to Recapitalize the GSEs and Release Them From Federal Control

This appendix explores the budgetary effects of two illustrative approaches to recapitalizing Fannie Mae and Freddie Mac and ending their federal conservatorships. Those formerly independent government-sponsored enterprises (GSEs) are currently controlled by the Federal Housing Finance Agency (FHFA) and effectively owned by the Treasury, their majority shareholder.

Outline of the Illustrative Policies

Many options exist for rebuilding the capital reserves of the GSEs and releasing them from government control. This appendix examines two hypothetical policies: one in which the Treasury would convert its senior preferred stock in the GSEs to common stock and one in which the FHFA would place the GSEs in receivership (a condition that would involve having their assets transferred to other entities).

Conversion During Conservatorship

In the first approach, the process of recapitalization and release would begin with the Treasury changing its senior preferred shares in the GSEs into common shares. Next, the Treasury would exercise the warrants it received in exchange for agreeing to keep the GSEs solvent. Those warrants allow it to buy additional common shares in the GSEs. By converting its senior preferred shares, the Treasury would give up the preference for dividend payments that those shares command; instead, the Treasury’s stock would have a status equal to that of the common stock held by other investors.

Next, Fannie Mae and Freddie Mac would issue additional common stock and sell it to the public. The cash raised from those sales would create capital reserves that would serve as a buffer to absorb larger-than-expected losses in the future and to meet the FHFA’s capital requirements for the GSEs—a precondition for releasing them from federal control. After their release, the Treasury would gradually sell its holdings of common stock. Each step in that illustrative process would have distinct effects on the proceeds the Treasury would be likely to receive from that sale (see Table 1).

Table 1.

Steps in an Illustrative Approach for Recapitalization and Release in Which the Treasury Converts Its Senior Preferred Stock to Common Stock

Notes

Data source: Congressional Budget Office.

This illustrative approach is the one labeled “Scenario 1: Conversion during conservatorship” in Table 2.

FHFA = Federal Housing Finance Agency; GSEs = government-sponsored enterprises (specifically, Fannie Mae and Freddie Mac).

Transfer During Receivership

In the second approach, the FHFA would change the GSEs’ legal status from conservatorship to receivership. Putting the GSEs in receivership would generally involve transferring their assets to other entities. In this illustrative example, the FHFA would create new legal entities to continue the operations of Fannie Mae and Freddie Mac (under the same names) and would transfer all of their existing assets and liabilities to those new corporations.

The receivership would then sell common shares in the new Fannie Mae and Freddie Mac corporations to private investors. Part of the proceeds from those sales would be kept by the corporations to satisfy the FHFA’s capital requirements. The rest would be used to liquidate shareholders’ claims in the original GSEs. In CBO’s assessment, the amount available to liquidate all claims would almost certainly be less than the liquidation preference of the Treasury’s senior preferred shares (the amount of money the Treasury is eligible to receive before more junior investors are paid). Thus, the Treasury would receive all of the proceeds of the common stock sale not kept as capital reserves.16

Effects on the Value of Fannie Mae and Freddie Mac

Those two approaches would result in identical projections of future earnings for Fannie Mae and Freddie Mac after recapitalization and release. Whether the conservatorships were ended through the conversion of preferred shares and sale of additional common shares or through receivership and transfer, Fannie Mae and Freddie Mac would have the same assets and liabilities afterward. The total market value of their equity should also be the same, but that value would be divided differently among stakeholders under the two approaches.

To illustrate that difference, the Congressional Budget Office set the combined market value of Fannie Mae and Freddie Mac after recapitalization and release at $368 billion in both scenarios and calculated how that value would be divided among various stakeholders (see Table 2).17 In that example, at the beginning of the recapitalization process, the GSEs’ combined capital would be $162 billion less than the amount required for release. The GSEs would need to raise the additional capital from new investors.

Table 2.

Distribution of the Equity Value of Fannie Mae and Freddie Mac Under Two Illustrative Scenarios for Recapitalization and Release

Billions of dollars

Notes

Data source: Congressional Budget Office.

n.a. = not applicable.

The mechanics of raising the additional capital would differ between the two scenarios, but both would result in new investors providing $162 billion to the companies and owning $162 billion of equity in exchange. Under the scenario of conversion during conservatorship, that $162 billion would come from selling shares in the existing enterprises. Under the scenario of transfer during receivership, shares for the entire ownership of the new corporations would be sold, and the $162 billion would be retained from the proceeds of those sales (the rest would be distributed to claimants in the original GSEs).

Conversion During Conservatorship

In the case of conversion, the new investors who provided $162 billion in new capital and the current stakeholders in the GSEs would own companies worth a total market value of $368 billion. With the new investors owning $162 billion, the remaining $206 billion would represent the value of the GSEs’ existing stock. Of that $206 billion, the Treasury would own $170 billion after converting its preferred stock to common stock and exercising its warrants, CBO estimates.

To calculate that amount, CBO applied an illustrative conversion rate of $2 per share, meaning the Treasury would receive 95 billion common shares for the $190 billion face value of its preferred stock.18 The warrants entitle the Treasury to an additional 7.2 billion common shares (four times the 1.8 billion common shares currently outstanding). The resulting 102 billion shares of common stock that the Treasury would own would have a value of $170 billion, CBO estimates, given the total value of common stock and the need to issue enough additional shares to raise $162 billion from new private investors. Under those estimates, nearly 97 billion shares would be issued to new private investors, with an implied share price of $1.67.19

Transfer During Receivership

In the case of transfer, the Treasury would receive the entire $206 billion value of the GSEs’ existing stock, because the liquidation preference of its senior preferred stock exceeds that amount. The difference between the Treasury’s receiving $206 billion for its shares in the transfer scenario but owning $170 billion in common stock in the conversion scenario results from converting the senior preferred stock to common stock. If the Treasury gave up the preference associated with its senior shares, that conversion would create $36 billion of value for private shareholders in the GSEs—$33 billion for the owners of junior preferred stock and $3 billion for the owners of existing common stock. That $36 billion would essentially be transferred from the Treasury to those shareholders.

Glossary of Terms Related to Fannie Mae and Freddie Mac

accrual accounting: A system of accounting in which the estimated value of expenses and related receipts is recorded when the legal obligation for those expenses and receipts is first made rather than when subsequent cash transactions occur. Accrual accounting is used for a limited set of programs in the federal budget—mainly programs that make or guarantee loans.

administrative actions: Actions taken by federal agencies to implement laws and programs under their authority. Such actions include rulemaking, guidance, agency letters, directives, notices, and similar activities.

backstop: See solvency guarantee.

basis point: One one-hundredth of a percentage point.

capital requirements: The minimum amount of financial resources that regulators require a financial institution to maintain to absorb potential losses.

capital shortfall: The amount of additional resources that a financial institution needs to amass to meet its regulators’ capital requirements.

cash-based accounting: A system of accounting in which revenues are recorded when cash is received and expenses are recorded when payment is made, regardless of the accounting period in which the revenues were earned or the costs were incurred. Cash-based accounting is used for most programs in the federal budget.

CBO baseline: Projections of the federal spending and revenues that would result over the next 10 years if current laws generally remained in place. CBO constructs its baseline in accordance with rules specified in law or developed jointly with the House and Senate Committees on the Budget and the Administration’s Office of Management and Budget. The baseline is not intended to be a forecast of future budgetary outcomes; rather, it serves as a benchmark that policymakers can use to evaluate the anticipated effects of proposed legislation.

CBO cost estimate: An estimate of how proposed legislation would change federal spending and revenues over the next 5 or 10 years in relation to CBO’s projections of budgetary outcomes under current law. CBO is required by law to produce a formal cost estimate for nearly every bill that is approved by a full committee of either the House or the Senate.

commitment fee: A fee charged in exchange for a legally binding promise to do something. In this report, the term refers to an amount the Treasury could charge Fannie Mae and Freddie Mac to continue guaranteeing their solvency after they were released from federal control.

common stock: Ownership shares in a company that have the lowest priority for receiving dividend payments from the company or a share of the company’s assets if it is liquidated.

conservatorship: A legal arrangement in which a person or entity is appointed to control and oversee a company to put it on a sound financial footing. Fannie Mae and Freddie Mac have been under the conservatorship of the Federal Housing Finance Agency since September 2008.

conversion rate: The price that determines how many shares of one security an investor receives for another security. In this report, the term refers to the price at which the Treasury would convert its senior preferred shares in Fannie Mae and Freddie Mac to common shares under one possible approach for recapitalizing the GSEs and releasing them from federal control.

default: The failure to make required repayments on a debt obligation, such as a mortgage.

dividends: Payments that a company makes to its shareholders from its profits.

equity value: The total value of a company available to its owners or shareholders.

face value: See par value.

fair-value accounting: A type of accrual accounting that incorporates market risk, the risk that adverse movements in overall market conditions will reduce the value of a financial asset. Fair-value accounting takes into account the fact that financial assets tend to perform poorly when the economy is weak. At those times, borrowers are more likely to default on their debt obligations, and amounts recovered from defaulting borrowers are smaller.

Federal Credit Reform Act of 1990 (FCRA): A law requiring that certain federal credit programs be accounted for on an accrual basis in the federal budget. FCRA accounting differs from fair-value accounting by not incorporating market risk.

federal credit subsidy for mortgage guarantees: The amount the federal government effectively spends on mortgage-backed securities guaranteed by Fannie Mae and Freddie Mac. That subsidy equals the difference between the present value of losses from defaults (net of recoveries) expected to occur on those guarantees over their lifetime and the present value of the fees expected to be collected for those guarantees over their lifetime.

Federal Housing Finance Agency (FHFA): The federal agency that regulates Fannie Mae and Freddie Mac and serves as their conservator.

government-sponsored enterprises (GSEs): Private companies created by federal law to fulfill a particular purpose, such as facilitating the flow of funding for home loans. In this report, “GSEs” refers specifically to Fannie Mae and Freddie Mac.

junior preferred stock: Ownership shares in a company that have a higher priority than common stock—but a lower priority than senior preferred stock—for receiving dividend payments from the company or a share of the company’s assets if it is liquidated.

liquidation: The process of ending a company by selling its assets and distributing the proceeds to various claimants, such as creditors and shareholders.

liquidation preference: The amount of money a class of investors is eligible to receive from a company being liquidated before more junior investors are paid.

market risk: 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. The government is exposed to market risk through various activities, such as making or guaranteeing loans, providing insurance, or investing in financial assets.

mortgage-backed securities (MBSs): Financial securities whose payments of interest and principal are backed by the payments from a pool of mortgages. Fannie Mae and Freddie Mac provide funding to mortgage markets by purchasing home loans from lenders and pooling them into MBSs, which they sell to investors.

mortgage guarantee: Fannie Mae’s and Freddie Mac’s guarantee that investors who buy MBSs from them will receive timely payments of principal and interest on the pools of mortgages underlying those securities, even if borrowers default on the loans.

par value: The nominal value of a financial security, which determines the amount that the issuer promises to pay if it buys back the stock or is liquidated. In the case of preferred stock, such as that issued by Fannie Mae and Freddie Mac, the obligation to pay dividends is equal to the par value times a dividend rate.

pool of mortgages: The set of home loans underlying a mortgage-backed security.

preferred-stock purchase agreements: Agreements in which the Treasury has committed to keeping Fannie Mae and Freddie Mac afloat financially by purchasing additional senior preferred shares in the GSEs whenever their net worth falls below zero.

present value: A single number that expresses a flow of current and future income or payments in terms of an equivalent lump sum received or paid at a specific time, generally the present. The present value of future cash flows is determined by applying a discount rate, which reflects the time value of money and the risk associated with those cash flows. That rate is used to convert future cash flows into their equivalent value at a given time.

recapitalization: The process of enabling Fannie Mae and Freddie Mac to rebuild their capital reserves by retaining more of their earnings and possibly by issuing additional shares.

receivership: A legal arrangement in which a person or entity is appointed to liquidate and resolve a failing institution (rather than correct its financial problems, as in a conservatorship).

release: Ending federal control of Fannie Mae and Freddie Mac and returning them to private ownership.

senior preferred stock: Ownership shares in a company that have the highest priority for receiving dividend payments from the company or a share of the company’s assets if it is liquidated.

solvency guarantee: A commitment by the federal government to maintain the solvency of Fannie Mae and Freddie Mac. Such a guarantee could be explicit, as under the current preferred-stock purchase agreements with the Treasury, or it could be implicit, as many investors assumed before the GSEs were placed in conservatorships.

solvent: The condition of being able to pay all legal debts.

subsidy: See federal credit subsidy for mortgage guarantees.

underpricing: The extent to which, in CBO’s estimation, Fannie Mae and Freddie Mac charge lower premiums for their mortgage guarantees than private-sector investors would charge to assume the same risk. Under current policies, that underpricing is projected to average about 0.06 to 0.10 percentage points in each of the next 10 years.

warrants: Securities that give the holder the right (but not the obligation) to purchase stock for a fixed price. In this report, the term refers to warrants that give the Treasury the right to buy common stock in Fannie Mae and Freddie Mac—equal to 79.9 percent of their total outstanding shares—for a nominal amount. Those warrants expire on September 7, 2028.


  1. 1. Federal Housing Finance Agency, “FHFA and U.S. Treasury Announce Amendments to the Preferred Stock Purchase Agreements” (news release, January 2, 2025), https://tinyurl.com/j5wv4xvv

  2. 2. Congressional Budget Office, The Budget and Economic Outlook: 2025 to 2035 (January 2025), Table B-4, www.cbo.gov/publication/60870.

  3. 3. By comparison, the GSEs’ guarantee fees averaged 66 basis points per loan in 2023. See Federal Housing Finance Agency, Fannie Mae and Freddie Mac Single-Family Guarantee Fees in 2023 (January 2025), https://tinyurl.com/yck567pv.

  4. 4. Congressional Budget Office, How CBO Determines Whether to Classify an Activity as Governmental When Estimating Its Budgetary Effects (June 2017), p. 7, www.cbo.gov/publication/52803.

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

  6. 6. The proceeds from the sale of common stock would go first to the Treasury, up to a limit equal to the liquidation preference of the senior preferred shares (the amount of money the Treasury is eligible to receive before more junior investors are paid). Any remaining proceeds would go to the private holders of junior preferred shares, up to the value of their liquidation preference, and then to the holders of common stock.

  7. 7. Congressional Budget Office, How the Housing Trust Fund and Capital Magnet Fund Support Affordable Housing (November 2022), www.cbo.gov/publication/58427.

  8. 8. Similar to its treatment of mortgage guarantees, CBO reflects the costs or savings from the GSEs’ assets in its baseline on a fair-value basis (see section 4 below). When the Treasury revised its agreements with the GSEs to allow them to retain interest and other income from those assets, CBO’s baseline accounted for the projected lifetime income to the federal government from the GSEs’ ownership of the assets when the policy change occurred rather than when the income was projected to be received over time. Because the value of those earnings has already been recorded in CBO’s baseline, the future cash flows from those earnings are not included in the baseline—but a loss of that income would be reflected in a cost estimate.

  9. 9. 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. For more information, see Congressional Budget Office, Measuring the Cost of Government Activities That Involve Financial Risk (March 2021), www.cbo.gov/publication/56778.

  10. 10. To facilitate comparison with other federal credit programs, CBO also produces FCRA estimates of the annual costs of Fannie Mae’s and Freddie Mac’s activities. See Congressional Budget Office, Estimates of the Cost of Federal Credit Programs in 2025 (August 2024), www.cbo.gov/publication/60517.

  11. 11. A present value is a single number that expresses a flow of current and future income or payments in terms of an equivalent lump sum received or paid at a specific time, generally the present. The present value of future cash flows is determined by applying a discount rate, which reflects the time value of money and the risk associated with those cash flows. That rate is used to convert future cash flows into their equivalent value at a given time. For details, see Congressional Budget Office, How CBO Uses Discount Rates to Estimate the Present Value of Future Costs or Savings (October 2024), www.cbo.gov/publication/60284.

  12. 12. Congressional Budget Office, CBO Explains How It Incorporates Administrative and Judicial Actions When Updating Its Baseline Projections and Preparing Cost Estimates (December 2024), www.cbo.gov/publication/60846.

  13. 13. Congressional Budget Office, Federal Subsidies and the Housing GSEs (May 2001), www.cbo.gov/publication/13072.

  14. 14. Congressional Budget Office, An Update to CBO’s Analysis of the Effects of Recapitalizing Fannie Mae and Freddie Mac Through Administrative Actions (December 2024), www.cbo.gov/publication/60810, and Effects of Recapitalizing Fannie Mae and Freddie Mac Through Administrative Actions (August 2020), www.cbo.gov/publication/56496.

  15. 15. Federal Reserve Bank of New York, “Actions Related to AIG” (accessed May 14, 2025), www.newyorkfed.org/aboutthefed/aig.

  16. 16. Under this scenario, it would not matter whether the Treasury exercised its warrants, because common shares in the original GSEs would have too low a priority to have any value in liquidation, CBO estimates. The liquidation preference is equal to the face value of preferred stock, plus the value of dividends that the GSEs would have made if dividends had not been suspended because of changes made to the preferred-stock purchase agreements in 2019. That liquidation preference totaled $341 billion as of December 2024 and is projected to grow to nearly $400 billion by the end of 2026.

  17. 17. In this example, CBO used the total market value and amount needed for recapitalization from Scenario 2 of Table 1 in Congressional Budget Office, An Update to CBO’s Analysis of the Effects of Recapitalizing Fannie Mae and Freddie Mac Through Administrative Actions (December 2024), p. 9, www.cbo.gov/publication/60810. In actuality, the GSEs’ total market value and amount needed for recapitalization would depend on many factors and could differ from the amounts in CBO’s examples.

  18. 18. The conversion rate would be based on legal and other considerations. The rate CBO used is solely for illustration.

  19. 19. The fact that the implied share price is lower than the conversion rate reflects the loss that the Treasury would experience from agreeing to have a lower priority for dividends (relative to that of junior preferred stock) after the conversion.

This report was prepared to enhance the transparency of the Congressional Budget Office’s work in analyzing proposals to change the status of Fannie Mae and Freddie Mac. In keeping with CBO’s mandate to provide objective, impartial analysis, the report makes no recommendations.

Michael Falkenheim prepared the report with contributions from Mitchell Remy and guidance from Sebastien Gay. Chad Chirico, Justin Humphrey, Noah Meyerson, Zunara Naeem, Sam Papenfuss, Emily Stern, Robert Sunshine (a consultant to CBO), Aurora Swanson, David Torregrosa, and Byoung Hark Yoo offered comments. Wendy Kiska fact-checked the report.

Comments were also provided by Laurie Goodman of the Urban Institute and by Deborah Lucas of the Massachusetts Institute of Technology and Donald Marron of the Urban Institute (both consultants to CBO). The assistance of external reviewers implies no responsibility for the final product; that responsibility rests solely with CBO.

Mark Hadley and Jeffrey Kling reviewed the report. Christian Howlett edited it, and Jorge Salazar created the graphics and prepared the text for publication. The report is available at www.cbo.gov/publication/61374.

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

Phillip L. Swagel

Director