Statement of
Douglas Holtz-Eakin
Director

Regulation of the Housing
Government-Sponsored Enterprises

before the
Committee on Banking, Housing, and Urban Affairs
United States Senate

October 23, 2003

Mr. Chairman, Senator Sarbanes, and Members of the Committee, thank you for this opportunity to discuss the Congressional Budget Office's (CBO's) work on the economics, costs, and regulation of the government-sponsored enterprises (GSEs) for housing--namely, Fannie Mae, Freddie Mac, and the Federal Home Loan Banks. Broadly speaking, that work leads to three main points:

• The federal government confers substantial benefits on GSEs through an implied guarantee of their debt and other financial obligations;
  
  
• In doing so, the government necessarily exposes taxpayers to risks; and
  
  
• Effective regulation can reduce but not eliminate the risks to taxpayers from the GSEs.

The Benefits of GSE Status

The principal benefit of having the status of government-sponsored enterprise is the ability to borrow at lower rates of interest than any fully private firm holding the same amount of private equity capital and taking the same risks is able to do. Sponsored status also enables the GSEs to borrow far larger sums than would be available to private borrowers. Low-cost capital and easy access to the market is the direct result of an implied federal guarantee of the GSEs' obligations.

The implicit guarantee is communicated to investors in capital markets through a number of provisions of law that create a perception of enhanced credit quality for the enterprises as a result of their affiliation with the government. Those provisions include a line of credit at the U.S. Treasury; exemption from the Securities and Exchange Commission's (SEC's) registration and disclosure requirements; exemption from state and local income taxes; and the appointment of some directors by the President of the United States. In addition, although federally chartered and federally insured banks face a limit on the amounts that they can invest in other types of securities, that limit does not apply to the GSEs' securities. Taken together, those statutory privileges have been sufficient to overcome an explicit denial of federal backing that the GSEs include in their prospectuses.

GSE status and the benefits it conveys are no longer necessary to the functions that Fannie Mae, Freddie Mac, and the Federal Home Loan Banks perform. Those purposes include ensuring a reliable source of funds to housing and increasing access to mortgage credit by low- and moderate-income borrowers so that more families can own their homes. Private financial institutions that lack GSE status, such as Washington Mutual and Bank of America, currently maintain a reliable link between the wholesale capital markets and retail lenders who originate home mortgages not eligible for financing from the GSEs. Moreover, the government has numerous more-direct policies to assist low-income home buyers, including mortgage insurance offered by the Federal Housing Administration and other more-targeted programs administered by federal agencies.

Private financial intermediaries, however, cannot match the low funding costs of the GSEs. To approach the GSEs' borrowing rates, they would have to raise more private equity capital and other private credit enhancements than do the housing GSEs. In short, they would need to convince lenders that they could replicate the federal guarantee through private means. However, private providers of risk-bearing or credit-enhancement services require compensation commensurate with the assumed risk. The requisite backing from private sources, therefore, is costly. By contrast, the government, provides the benefits of low-cost funding without charge.

Assisted by the implied federal guarantee, the housing GSEs have grown into some of the largest financial institutions in the world. Their outstanding securities now exceed $4 trillion--or more than the entire U.S. public debt. In the process, Fannie Mae and Freddie Mac have come to dominate the U.S. residential mortgage market, accounting for almost 57 percent of residential mortgage debt (see Table 1).

The value of the federal subsidy to the GSEs can be approximated by comparing the enterprises' actual funding costs with those they would face as private intermediaries. In May 2001, CBO estimated that difference--on the basis of a credit rating of AA- for the housing GSEs--to be $10 billion to $15 billion per year from 1998 to 2000. Adjusted for the growth of the enterprises (but with any increases in risk ignored), the current annual subsidy is, at a minimum, above the upper end of that range.
 

The Exposure of Taxpayers to Risks from the GSEs

By supporting the activities of the housing GSEs through an implied guarantee, the government has assumed, on behalf of taxpayers, the risk of losses that might exceed the enterprises' holdings of private equity capital. The housing GSEs offer public assurances that their assumed risks, especially for credit or default losses, are low in relation to their private capital. As a result, taxpayers may conclude that their own risk exposure is also low.

The housing GSEs appear to be principally exposed to interest rate, prepayment, and operations risks. Interest rate risk refers to the different effect that changes in interest rates can have on the value of a firm's assets and liabilities and thus on its net worth. For example, an increase in interest rates will reduce the value of both fixed-rate assets and fixed-rate liabilities, but the value of assets will be hit harder if the assets have a longer maturity than the liabilities do. A rise in interest rates, therefore, can wipe out a financial intermediary's equity capital.

Entities that hold portfolios of fixed-rate mortgages are also subject to prepayment risk. Specifically, the value of a portfolio of fixed-rate mortgages declines when borrowers exercise their option to refinance and prepay their existing mortgages in response to a decline in market rates. In combination, interest rate and prepayment risks mean that the housing GSEs are potentially vulnerable to losses from both increases and decreases in interest rates.

Even those firms that appear to be well managed are subject to operations risk, or the adverse effects of errors in judgment by management in protecting the value of a firm. That threat can manifest itself in lapses in the integrity and performance of existing controls, systems, and practices.

Private equity holders and other stakeholders in the housing GSEs have some incentive to manage and control risk, but overall those incentives are weaker than those for investors in other entities. Market discipline is weakened by the federal guarantee, which reduces the need for bondholders to monitor and restrict the enterprises' risks. Further, equity holders have diminished incentives to resist risk taking to the extent that they believe that the government would intervene to sustain the GSEs. Member institutions holding equity in the Federal Home Loan Banks may undervalue the enterprises' risks because they can withdraw some of their equity from a financially troubled bank to reduce their potential losses. Following severe losses, equity holders who cannot withdraw their capital can have an incentive to accept increased risks by the enterprises because that approach may be their only means of recovering those losses. In sum, the federal government cannot count exclusively on nonfederal stakeholders to limit the risks to taxpayers from the housing GSEs.

Nonetheless, the housing GSEs are managing prepayment risk and interest rate risk through such means as issuing debt securities that can be redeemed at par before maturity and using derivatives, including interest rate swaps. Also, the GSEs' internal monitoring and safeguards reduce operations risk. Finally, the housing GSEs are limiting their exposure to credit risk by requiring private mortgage insurance on loans with less than a 20 percent down payment and by leaving some of that risk with the loan originators.

As a practical matter, however, the enterprises' risks cannot be eliminated, nor would doing so be in the interests of equity investors. The risks of financing and holding a portfolio of mortgages are simply too varied and complex to permit management to identify them all and to find another party willing to accept them at a reasonable cost. The more feasible objective of holding interest rate and prepayment risks within acceptable bounds is among the most complex and difficult tasks facing the managers of mortgage portfolios. At the housing GSEs, risk management is assigned a high priority and is reported to be vigorously pursued with state-of-the-art systems and analytical procedures. Even so, best practices intended to achieve vital objectives occasionally fail and produce unpleasant surprises.

Matters are complicated further by shareholders' desire to retain some risks. The return on riskless financial activity is close to the return on U.S. Treasury securities. In competitive markets, investors can obtain high rates of return only by assuming risks. Fannie Mae and Freddie Mac have consistently earned high rates of return on equity. For example, the average annual return on their equity from 1990 to 2002 was over 23 percent. A comparison group of large financial services firms averaged returns of less than 14 percent during that period. One essential operating difference between those two GSEs and private firms is that the GSEs hold less than half as much private equity capital per dollar of assets as the comparison firms do (3.70 percent versus 9.14 percent). If Fannie Mae and Freddie Mac retained about the same risks as private financial services firms, then their higher rates of return on capital could be explained by their lower levels of capital.

Future losses from risks retained by the housing enterprises would be borne by the enterprises' equity investors up to the limit of the GSEs' equity and reserves. Creditors could then look to the federal government to cover losses above those amounts. Some observers claim that the government's commitment is only conjectural and therefore potentially illusory. However, when another GSE, the Farm Credit System, suffered threatening losses in the 1980s, the Congress authorized up to $4 billion in federal financial assistance to avoid a default on bonds that carried a similar guarantee. In that case, at least, the implied federal guarantee became real. In the event of future losses by the housing GSEs in excess of their private capital, the government would face a choice between ignoring a financial shock of unknown magnitude or confirming that its guarantee would be honored. The significant difference in the expected short-term costs of those alternatives suggests that the capital markets are likely to be correct in supposing that the government will not walk away from its implied guarantee when the need for federal support arises.

A rough indication of the likelihood of such an event is provided by the cumulative average historical default rate for corporate debt with a credit rating comparable to that of Fannie Mae and Freddie Mac. Standard & Poor's reports that for debt rated AA-, the cumulative average default rate over 15 years is 1.92 percent. By that indication, a default by Fannie Mae or Freddie Mac is highly unlikely over the next 15 years. But it is not an impossibility.
 

The Role of Regulation in Limiting Taxpayers' Risks

By enhancing the housing GSEs' credit quality, the federal government gives the enterprises substantial control over the risks faced by taxpayers and over the amount of the federal subsidy. The enterprises can increase that subsidy by expanding their volume of guaranteed debt, by engaging in riskier activities, by reducing their efforts to hedge existing risks, and by diverting income to activities outside their missions or distributing it to shareholders.

Fannie Mae and Freddie Mac have two means of channeling funds from the capital markets to retail lenders: investing in mortgages and guaranteeing mortgage-backed securities (MBSs). To invest in mortgages, the enterprises issue debt obligations and purchase mortgages. Alternatively, they pool individual mortgages, insure the pools against credit risk, and sell undivided interests in the pools directly to investors in the form of mortgage-backed securities. Purchasing and holding mortgages as investments entails greater risks and returns for the GSEs than guaranteeing MBSs does. Fannie Mae and Freddie Mac have dramatically increased the size of their investment portfolios relative to their guarantees of MBSs since 1990 (see Table 2). In fact, Fannie Mae and Freddie Mac now hold in portfolio about one-third of their guaranteed MBSs. Similarly, the Federal Home Loan Banks have increased their portfolio holdings of mortgages from less than $1 billion in 1998 to more than $60 billion in 2002 and to $90 billion by the middle of 2003.

When the enterprises buy and hold mortgage assets in portfolio, they are retaining interest rate, prepayment, and credit risks on those loans. But when the GSEs sell mortgages to investors through guaranteed MBSs, they transfer interest rate and prepayment risks, retaining only the more transparent, manageable credit risk. As the GSEs move mortgages into their portfolios, they increase both the expected returns and risks to shareholders; for taxpayers, only the risks increase. The increase in risk is reflected in the statutory minimum for private capital to be held by Fannie Mae and Freddie Mac of 2.5 percent for mortgages in portfolio and 0.45 percent for MBSs. Whether those differences in capital requirements accurately reflect true differences in the level of risk, however, is impossible to know because the enterprises can vary the extent to which they hedge portfolio risks. Determining the adequacy of the Federal Home Loan Banks' capital is further complicated by the ability of members to redeem some capital at par. Redeemable capital is unlikely to be available to absorb the banks' losses or to protect taxpayers.

An important purpose of the regulation of GSEs is to limit taxpayers' risks and the size of the subsidy. To do so, regulators must understand, monitor, and assess the risks of the enterprises virtually to the same extent that their management does. But some dimensions of risk are not easily transparent. Even world-class regulators--well funded, well staffed, and politically independent--are unlikely to be able to maintain a complete understanding of the extent to which taxpayers are exposed to risks.

Nonetheless, regulators can limit the GSEs' ability to leverage the value of the federal guarantee. To do that, they need a range of capabilities to address the varied means by which the GSEs can increase the risk exposure of taxpayers. Those capabilities include being able to adjust capital requirements, to assess the extent to which the GSEs have retained interest rate and prepayment risks and the effectiveness of hedges against those risks, to hold management responsible for the adequacy of internal systems and controls, and to prevent a failed GSE from continuing to use the federal guarantee.

The regulators also need enough public support to enable them to exercise their authority to compel changes in risky behavior by the housing GSEs. Toward that end, increased public disclosure of the findings of regulatory oversight of the enterprises could be useful. Freddie Mac has agreed to publicly report its fair-value, or mark-to-market, net worth quarterly. That practice increases transparency and might be usefully adopted by all of the GSEs.

The Congress could facilitate the regulators' difficult task by setting statutory boundaries on the GSEs' ability to increase the value of the federal subsidy. For example, the Congress could legislate a higher margin of safety in the minimum capital standards. It could also act to limit the growth (or profitability) of GSEs' portfolio investments and move toward more-equal treatment of the enterprises and their potential competitors. Some Members of Congress have proposed requiring SEC registration of GSE securities, for example. A May 2003 CBO report on that topic found that such a requirement would be unlikely to have a significant adverse effect on the GSEs or on the mortgage markets. Similarly, in the absence of evidence that Presidentially appointed directors have a unique advantage in defending taxpayers' interests, the selection of directors might be left entirely to private shareholders.

Action by the Congress to bolster regulators' ability to ensure safe operation by the GSEs would better protect taxpayers. Furthermore, the GSEs' public mission does not appear to require them to sacrifice safety and soundness. Certainly, from the taxpayers' perspective, having the GSEs pursue a low-risk strategy is strongly preferable to tolerating a risky one.
 

Related CBO Publications

Effects of Repealing Fannie Mae's and Freddie Mac's SEC Exemptions, May 2003

Letter to the Honorable Paul S. Sarbanes regarding the new-business assumption in the risk-based capital rule for Fannie Mae and Freddie Mac, January 2003

Letter to the Honorable Richard H. Baker regarding CBO's May 2001 report on the housing GSEs, July 2001

Federal Subsidies and the Housing GSEs, May 2001

Testimony of Dan L. Crippen on Federal Subsidies for the Housing GSEs, May 2001
(Before the Subcommittee on Capital Markets, Insurance, and Government Sponsored Enterprises of the House Committee on Financial Services)

Interest Rate Differentials Between Jumbo and Conforming Mortgages, 1995-2000, May 2001

Remarks of June E. O'Neill before the Conference on Appraising Fannie Mae and Freddie Mac, May 1998

Testimony of June E. O'Neill on Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac, June 1996 (Before the Subcommittee on Capital Markets, Securities and Government Sponsored Enterprises of the House Committee on Banking and Financial Services)

Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac, May 1996

The Federal Home Loan Banks in the Housing Finance System, July 1993

Controlling the Risks of Government-Sponsored Enterprises, April 1991

Government-Sponsored Enterprises and Their Implicit Federal Subsidy: The Case of Sallie Mae, December 1985

The Housing Finance System and Federal Policy: Recent Changes and Options for the Future, October 1983