The GSE Experiment Has Failed–Congress Should End It

Throughout the 16-year conservatorship, many people have proposed plans to rework the Preferred Stock Purchase Agreements (PSPAs) that specify the terms of the support the US Treasury provided to the GSEs.³ Although renegotiating the PSPAs may be necessary to end the GSEs’ conservatorships, any new deal should be structured to facilitate receivership and liquidation, accelerating an end to America’s failed GSE experiment.

The federal government’s dominance in the housing finance market—both before and throughout the GSE conservatorships—has stifled private competition and worsened Americans’ financial condition, but it has done virtually nothing to increase the homeownership rate. It is time to wind down the affairs of the GSEs through receivership, a process that the federal government should have undertaken in 2008.

On September 6, 2008, the Federal Housing Finance Agency (FHFA) placed the GSEs under conservatorship rather than receivership. In accordance with its broad conservator powers, the FHFA secured two $100 billion credit lines (one for each company) from the Treasury to bail the GSEs out of trouble. As part of this PSPA, each GSE issued a million shares of a new class of senior preferred stock to the Treasury (with an initial valuation of $1,000 per share) in exchange for its line of credit.

The PSPAs specified various other terms and conditions of the bailout, and the initial agreement contained several taxpayer protections, including the following four items.⁴

1. Absent Treasury approval, dividend payments on classes of stock other than the specially created senior preferred stock were suspended until the GSEs repurchase the preferred stock from the Treasury.
2. Treasury holds the right (through warrants) to purchase up to 79.9 percent of the GSEs’ common stock. The warrants do not require the Treasury to make this purchase.
3. A “liquidation preference” specifies that any funds derived from either new capital infusions or the liquidation of assets must first be used to compensate taxpayers for the bailout, and the GSEs cannot emerge from conservatorship without paying this liquidation preference in full. The liquidation preference was initially set at $1,000 per share ($1 billion for each GSE’s senior preferred stock). It was also designed to increase as the GSEs drew on their lines of credit and if the GSEs did not pay (to the Treasury) the periodic commitment fees due on the senior preferred stock. In the event the GSEs are dissolved, the liquidation preference takes priority over liabilities due to other investors or creditors.
4. The GSEs were required to pay quarterly dividends on the Treasury’s senior preferred stock. The initial PSPAs set these dividends equal to 10 percent (annualized) of the value of the liquidation preference if paid in cash. Dividend payments were also set to increase if the Treasury extended additional credit. Notably, dividend payments on the senior preferred stock do not diminish the value of the liquidation preference—the dividend payments function as interest on a loan, and the balance of the liquidation preference reflects the unpaid principal.

Between 2008 and 2012, the companies struggled and had difficulty meeting their financial obligations under the PSPAs. While the FHFA could have placed the GSEs into receivership, the FHFA and the Treasury chose instead to amend the PSPAs three times. Each rendition effectively forced the taxpayers to bail out the GSEs again. Thus, in total, the GSEs were bailed out four times, as follows.⁵

1. In September 2008, the Treasury bailed out both GSEs, promising to provide each with up to $100 billion in credit. In return, the companies each gave the Treasury one million shares of senior preferred stock, worth $1 billion each. In just the first three quarters of the conservatorship, Fannie Mae exhausted one-third of its Treasury credit commitment, and Freddie Mac exhausted more than half of its commitment.
2. In May 2009, with losses rapidly piling up, the Treasury doubled its prior commitment, promising to provide each GSE with up to $200 billion in credit.
3. In December 2009, with the companies still struggling, the Treasury changed its commitment formula, allowing it to provide more than $200 billion in credit. The GSEs drew on these lines of credit in 2010, 2011, and 2012. The $200 billion caps adjusted upwards by the amount of the draws throughout those three years, less any positive net worth of the GSEs at the end of 2012. The GSEs’ poor financial performance between 2009 and 2012 resulted in the commitment expanding from $400 billion in 2009 to $445.5 billion by the end of 2012.
4. At the end of 2011, the GSEs’ combined liquidation preference was more than $187 billion. Moreover, the required 10 percent senior preferred stock dividends of nearly $19 billion per year exceeded the net income ever earned by the GSEs in a single year. It had become obvious that the GSEs might be unable to meet their obligations to the Treasury—from December 2008 through December 2011, the GSEs had borrowed $36 billion from the Treasury to pay the dividends they owed the Treasury. So, in August 2012, the Treasury changed the dividend formula to prevent the GSEs from further drawing on the Treasury commitment just to pay the dividends. Specifically, the Treasury amended the agreement in August 2012 so that beginning in August 2013, the Treasury would take (“sweep”) any GSE profits to satisfy the quarterly dividend payments.

In addition to these bailouts, the Treasury and Federal Reserve purchased more than $3 trillion in mortgage-backed securities (MBSs) and GSE bonds, staving off further losses. From September 2008 through December 2009, the Treasury purchased more than $220 billion of GSE MBSs.⁶ The Fed purchased $134.5 billion of GSE bonds and more than $1.1 trillion of GSE MBSs from December 2008 through March 2010 and an additional $2.2 trillion in MBSs from October 2011 through June 2019.⁷ As of September 2024, the Fed still holds $2.3 trillion in MBSs on its balance sheet.⁸

While the fourth (2012) amendment was controversial, the GSEs would not have survived without these federal bailouts. As FHFA Director Edward DeMarco explained after this 2012 amendment, “The continued payment of a fixed dividend could have called into question the adequacy of the financial commitment contained in the PSPAs.” Demarco also reaffirmed the FHFA’s “commitment to move forward with the components of the Strategic Plan for the Conservatorships of Fannie Mae and Freddie Mac,” a plan that included (among other items) “gradually contracting their operations.”⁹

Nonetheless, the companies are larger today than they were in 2012. Aside from normal growth, in 2021, the FHFA removed restrictions on the GSE’s ability to acquire single-family mortgages, certain loans with higher risk characteristics (including second homes and investment properties), and multifamily loans.¹⁰ As a result, Fannie Mae’s mortgage loans (net of allowances) increased from $2.9 trillion in 2012 to $4.1 trillion in 2023.¹¹ Freddie Mac’s net mortgage loans increased from $1.7 trillion in 2012 to $3 trillion in 2023.¹² Despite their growth, both GSEs have negative capital balances and remain in conservatorship.¹³

The FHFA suspended the GSEs’ capital requirements for the duration of the conservatorships because the GSEs had no capital, but these must be reinstated upon release.¹⁴ Although the FHFA allowed the GSEs to retain profits and build additional equity in 2019, their capital balances remain negative, as they have been since 2008 (see Figure 1).¹⁵ Importantly, the current negative capital balances of both GSEs would result in classification as “critically undercapitalized” if the FHFA reinstated their capital standards.¹⁶ Under the Housing and Economic Recovery Act of 2008 (HERA), the FHFA director has the discretionary power to place the GSEs into receivership and liquidate their assets if they are classified as critically undercapitalized.¹⁷

As of 2019, the GSEs needed approximately $200 billion to avoid being classified as critically undercapitalized under the FHFA’s proposed capital rules.¹⁸ However, since 2021, the GSEs have improved their capital position by approximately $100 billion, suggesting that they may need to raise $100 billion to exit conservatorship and meet the minimum capital requirements.¹⁹ However, this (still quite large) figure ignores the additional amount that would be needed to pay the nearly $200 billion liquidation preference on the senior preferred stock.

Any write-down of this liquidation preference would be another bailout, but supporters argue that such a write-down is justified because the GSEs paid back more in dividends than the amount they borrowed. This argument, however, focuses strictly on cash-flow accounting. It ignores the principal component of the original debt agreement, the ongoing risks borne by taxpayers, and the opportunity costs associated with the cash infusions. In other words, any such amendment would deprive taxpayers of proper compensation for the risks they undertook during the past 16 years for bailing out the GSEs.

To start moving away from the GSE system, the FHFA must first remove them from conservatorship and place them into receivership (with the goal of liquidation). To place the GSEs into receivership, the FHFA must first reinstate the capital requirements and classify the GSEs as critically undercapitalized. Once the GSEs enter receivership, the FHFA director may begin the liquidation process and transfer GSE assets and liabilities into newly chartered limited-life regulated entities (LLREs).²⁰ Each GSE charter would then be immediately transferred to the LLREs, and each LLRE would assume the powers and attributes of the GSE being liquidated.²¹

The FHFA would then have a maximum of five years to wind down all the affairs of the LLREs.²² Once the FHFA sells at least 80 percent of the LLRE’s capital stock to third parties, the LLRE terminates automatically.²³ The FHFA must then divest any remaining capital stock of the former LLRE within one year.²⁴

To finally get rid of the GSEs, Congress, not the FHFA, would then need to revoke the GSEs’ charters to prevent them from being rereleased into the marketplace.²⁵ As former FHFA Director Mark Calabria has explained, “The law is quite clear. FHFA would continue to run the GSEs, with the option of a good/​bad bank model to resolve bad assets, and the only way FHFA can terminate the receivership is to sell the charters back into the marketplace.”²⁶ Without this type of wind down, any plan must fully compensate taxpayers for prior bailouts and ongoing credit risk by repaying the liquidation preference in full before resuming dividends to private shareholders.

Robust homeownership existed in the United States long before the government became heavily involved in the housing market, and a competitive, private market is not possible if the current government-guaranteed duopoly is allowed to continue. Liquidation of the GSEs—rather than recapitalization and release—is the most prudent way to create this competitive marketplace.

In the absence of the GSEs, fully private companies with no US Treasury credit lines or other special funding advantages can purchase and securitize mortgages. Rather than invest in the GSEs, investors can provide capital to new companies that do not have either implicit or explicit taxpayer backing, thus reducing taxpayer risk, ending the economic distortions caused by the government domination of the market, and gradually improving housing affordability.

Prior to the 2008 financial crisis, the federal government controlled a dominant share of the US housing finance system, and that share has expanded. As of 2023, Fannie and Freddie (both of which remain in government conservatorship) had a combined guaranty book of business of $7.6 trillion, representing approximately 47 percent of the nation’s outstanding mortgage debt.²⁷ From 2009 to 2023, Fannie and Freddie’s annual share of the total MBS issued averaged 70 percent. Including Ginnie Mae securities (those backed by Federal Housing Administration mortgages), the federal share of the MBS market averaged 93 percent per year.²⁸ Moreover, from 2008 to 2023, the Federal Reserve went from holding zero MBS to more than $2 trillion (combined Fannie, Freddie, and Ginnie MBS).²⁹ (See Figure 2.)

Even though the US homeownership rate has changed little since the 1960s, volatility of home prices and home construction in the United States were among the highest in the industrialized world from 1998 to 2009.³⁰ Overall, the evidence suggests that the expansive federal role has expanded mortgage ownership but has done little to expand homeownership. At the very least, federal housing finance policies have magnified economic instability by inducing higher home prices.³¹ Federal involvement expanded after the most recent financial crisis, for instance, and home prices rose to 43 percent more than where they peaked before their 2007 crash.³²

Inducing demand in supply-constrained markets can only serve to put upward pressure on prices, and housing markets are no exception. Nonetheless, inducing demand is precisely what federal policies have done for decades by making it easier to obtain home mortgages. Congress can start to reverse these trends by winding down the GSEs and reducing federal involvement in housing finance.³³

Without the commitment by the Treasury to extend up to $445 billion in credit to the GSEs, both would have long since dissolved. From 2008 through 2011, the GSEs relied heavily on this commitment to remain in business and to make timely payments to their debt holders and MBS investors. Though the GSEs drew less frequently on the line of credit after 2011, they could not have continued operating without the Treasury’s commitments.

Understandably, shareholders desire a release from conservatorship to enjoy a resumption of dividends and long-term growth in shareholder value. Yet release from conservatorship in accordance with the current PSPAs and reinstated capitalization standards would require a capital raise of approximately $300 billion. Liquidation of the GSEs—rather than recapitalization and release—is the most prudent way to create a more competitive market for housing finance and protect future taxpayers. The GSE experiment has failed, and it is long past time for Congress to end it.

Jerome Famularo contributed invaluable research assistance to this report.