Skip Navigation
| Skip Navigation |
|
|
|
|
| Assessing the Public Costs and Benefits of Fannie Mae and Freddie Mac May 1996 |
|
The Congress could adopt measures to increase public
benefits, reduce costs, or improve the ability of the government to obtain
timely, relevant, and reliable information about the Federal National Mortgage
Association and the Federal Home Loan Mortgage Corporation. Although many
of those changes in current policy would stop short of privatization, they
would withdraw some of the retained benefits of the status of a government-sponsored
enterprise or provide firmer estimates of the potential gains or losses
from privatization.
Increase Public Benefits
Increasing public benefits would redirect subsidies from Fannie Mae and Freddie Mac to public beneficiaries, including taxpayers. If the subsidy was to be redirected to home buyers, a policy of increasing competition in the secondary market for conforming mortgages would be required. If the subsidy was to be retargeted toward low-income and other high-risk borrowers, more intrusive regulation of GSE mortgage purchases and additional mandated activities could be necessary.
Increase the Subsidy Pass-Through to Home Buyers
The lack of competition in the market for securitizing conforming mortgages limits the share of the subsidy now passing through to home buyers. Fannie Mae and Freddie Mac are able to sustain a duopoly because GSE benefits are provided exclusively to them. One solution is for the government to withdraw its implicit guarantee of all GSE securities. Doing so, however, would cause mortgage interest rates to rise. Alternatively, the government could explicitly guarantee all mortgage-backed securities, whether issued by a GSE or a fully private company. To qualify, fully private conduits would have to agree to meet the safety and soundness requirements of the Office of Federal Housing Enterprise Oversight under the same terms as Fannie Mae and Freddie Mac. The government would not make any payments under that guarantee, except in the extreme circumstance--as is the case with the GSEs--of an issuer failing to make payments on MBSs.
Under this option, mortgage rates could fall by the portion of the subsidy on MBSs that the GSEs retain--that is, five basis points (0.05 percentage points). The government's liability, however, would increase in scope. Moreover, under current budget concepts, the expected cost of those guarantees would be recorded in the budget as outlays when the federally guaranteed MBSs were issued. Some of that increase in outlays, however, would simply recognize the costs of the GSEs that are now excluded from the budget.
Eliminate the Debt Subsidy
If explicit MBS guarantees were available, no justification would exist for the special privileges now granted to the GSEs, including the exemption from state and local income taxes and Securities and Exchange Commission registration fees. Those special provisions of the law could be repealed. When taking that action, the government could also disavow any responsibility for subsequent debt issues of the GSEs. If the government's disavowal of credit enhancement for future debt issues was credible, the market interest rate on new issues would rise to reflect the intrinsic quality of credit of the GSEs. The explicit guarantee on MBSs, along with withdrawing all special privileges for GSEs, should eliminate the subsidy on GSE debt. If, despite those measures, the market continued to regard the debt of GSEs as being enhanced in quality by the federal government, then the GSEs would emerge with most of their subsidy intact, unthreatened by the entry of private competitors that would not have the option to issue subsidized debt. In such circumstances, the GSEs might leave the MBS market entirely to private intermediaries and specialize in debt-financed portfolio lending, whose high profitability could survive this policy change.
An exclusive or dominant focus by Fannie Mae and Freddie Mac on higher-risk portfolio lending would increase the importance of effective safety and soundness measures. The scope of OFHEO's regulatory responsibilities would also increase under this policy as the number of firms within its jurisdiction rose. The exposure of taxpayers to risk could also widen if a larger share of home mortgages was securitized.
If those policies were successful in terminating the subsidy on GSE debt, substituting explicit for implicit guarantees, and forcing the complete pass-through of the MBS subsidy, then they would have achieved successful privatization by most standards. Competition would increase, mortgage interest rates would fall, and home ownership would rise. Eventually, the Congress could repeal the GSE charter acts and recharter the enterprises under state law as a formality. Breaking the GSEs' monopoly and replacing the exclusive implicit guarantee with an inclusive explicit one would strengthen Congressional control of the subsidy. For example, the government's guarantee might eventually be reduced from 100 percent of loss to 95 percent. Alternatively, the government could provide the current volume of federal guarantees of MBSs without charge and meet the growth in demand for guarantees by auctioning additional credit enhancement. Those policy changes could result in budget savings.
Target the Retained Subsidy
If Fannie Mae and Freddie Mac were to be permitted to continue receiving subsidies at current rates, policy could attempt to redirect a portion of the retained benefit to low-income and other high-risk home buyers. Raising the affordable-housing goals authorized in the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 would be a natural approach to that policy.
However, clarity of objectives is vital in evaluating this option. The goal is to increase loans to borrowers who would otherwise not get credit but who are good credit risks. The policy does not intend to force issuers to originate loans that cannot be repaid and result in foreclosure and sale of property. That outcome would leave the borrower worse off than if a loan had not been granted, impose needless costs on the GSEs, and increase taxpayers' risks.
Given the limited understanding of the factors that lead to mortgage default, the desired goal is unlikely to be achieved by simply using arbitrary numerical targets for the purchase of mortgages by borrowers with specific characteristics. Use of the automated underwriting systems developed by the housing GSEs is increasing and might be helpful, but its success is not yet confirmed. More important, if the automated systems do succeed in identifying good credit risks who would otherwise be rejected, Fannie Mae and Freddie Mac would find it profitable to purchase those mortgages without binding goals for affordable housing.
The GSEs are expert at funding and generating large profits. The last
characteristic suggests that a more direct way to ensure affordable housing
would be to require the GSEs to contribute cash to a fund that would directly
assist low-income borrowers through mortgage interest rate buydowns and
contributions to down payments. A precedent for such a requirement is the
Federal Home Loan Bank System's Affordable Housing Program, which requires
the banks to contribute 10 percent of their net income from the previous
year for acquiring and rehabilitating affordable rental housing.(1)
That measure would more effectively target the retained subsidy toward
the intended beneficiaries.
Reduce Public Costs
The strategy of reducing public costs includes making policy changes
that would limit either the subsidy rate or the total amount of subsidy
accruing to the housing GSEs or both. Four examples would be to raise the
equity requirements for GSE shareholders, lower the ceiling on conforming
mortgages, cap and reduce the size of the GSEs' mortgage portfolios, and
impose a cost-of-capital equalization fee on GSE debt issues. In those
cases, the social gains from reducing the scope of GSE activity could be
affected by the extent to which the benefits of GSE status are "value added"
by government (see Box 5).
|
The housing enterprises have claimed that some of their funding cost savings is value added by the government in excess of the expected outlays from a GSE insolvency (see Box 4 in Chapter 2). If that is true, then cutting public cost by reducing the enterprises' activities (or moving to full privatization) could reduce the government's value added and cause the loss of value to exceed the gains to taxpayers. This potentially adverse effect, however, overlooks explicit guarantees as an alternative means of creating that value. Replacing free implicit guarantees with explicit ones and selling them at competitive fees would have several advantages over current policy. First, an explicit guarantee would provide an unconditional guarantee to investors. Second, the cost of the guarantee would be recognized and controlled in the budget. And third, receipts from guarantee fees would provide the government with resources to target subsidies toward special needs such as first-time or low-income home buyers or for any other public purpose. |
Increase Shareholder Equity
For a given level of risk assumed by a GSE, the higher the shareholder equity is, the less the need for credit enhancement by taxpayers. Thus, a requirement that shareholders put more of their capital at risk could reduce the cost of GSE operations to taxpayers. That requirement would be an extension of the policy of imposing risk-based capital requirements on the GSEs.
One of the disadvantages of reducing the taxpayer subsidy is that the GSEs might lower the portion of the subsidy passed through to home buyers, particularly if they had no additional competition in the marketplace.(2)
Lower the Ceiling on Conforming Mortgages
Decreasing the ceiling on conforming mortgages would reverse the direction of annual change in that market. Instead of increasing, the maximum-size loan eligible for purchase by the housing GSEs would decrease each year starting from the current level of $207,000. This policy would reduce management's discretion to determine the GSEs' rate of growth and the call on taxpayer resources. In time, it would also produce smaller GSEs.
Under the policy of downsizing loans, the dividing line separating the conforming and jumbo markets would move steadily in the direction of mortgages within reach of low-income households. As the limit on conforming mortgages receded and on jumbo loans expanded, competitive fully private intermediaries would securitize a wider range of mortgages. All mortgage markets would retain their access to the capital markets.
Interest rates at the conforming/jumbo boundary would rise. Those interest rate increases, however, would apply only to the largest mortgages, where a 25 to 35 basis-point rise in rates would have only a small effect on the decision to become a homeowner. In time, the declining ceiling on conforming mortgages would reach more interest-sensitive, low-income borrowers. When that occurred, the government could target cash subsidies toward those borrowers. Those subsidies could be financed from fees charged for explicit federal guarantees of privately issued MBSs.
A gradual downsizing of the GSEs would reduce the amount of shareholder capital required to protect taxpayer equity and would free equity capital in the housing GSEs. That effect would permit Fannie Mae and Freddie Mac to buy back existing equity shares without exposing taxpayers to increased risk. By refunding equity to stockholders, the housing GSEs would be providing capital to investors, which could be placed with the private intermediaries that would be expanding into the market formerly dominated by Fannie Mae and Freddie Mac.
Reducing the limit on conforming loans would also slim the housing GSEs to the size of fully private intermediaries. That reduction in the scale of Fannie Mae and Freddie Mac would address the general concern that privatization would not be effective in withdrawing the implicit federal guarantee because the housing enterprises would be too big for the government to permit them to fail.
Cap the GSEs' Mortgage Portfolios
The subsidy rate to the GSEs is substantially higher on debt issued to finance portfolio holdings of mortgages than on MBSs because the taxpayer capital required to back portfolio lending is higher. If the GSEs shifted their funding from debt to mortgage-backed securities, subsidies by taxpayers would be reduced without losing market integration or other benefits that the GSEs may provide to the public. This option places a dollar-volume cap on portfolio assets, although it could also be framed as a dollar cap on the volume of outstanding debt securities. In either case, the cap could be reduced gradually to the point at which the volume of risky assets held by the GSE was no more than 5 percent of the volume of MBSs outstanding. That policy would make the GSEs more like the jumbo conduits. Moreover, a portfolio of that size would be large enough to permit the GSEs to hold mortgages in inventory as a part of the process of securitizing mortgages. But it would not be large enough for the GSEs to take on substantial interest rate risk. A limit of 5 percent would be comparable with the size of Freddie Mac's owned portfolio when its operating focus was almost exclusively on MBS funding.
The housing GSEs argue that if their role in the secondary market was reduced, the volume of mortgages held by federally insured depositories or on-budget government agencies would increase. Thus, they argue, the taxpayer's exposure to risk would not be reduced by scaling back or privatizing Fannie Mae and Freddie Mac. The government would have an increased exposure to loss from the failure of insured banks and thrifts and from its on-budget direct loans and guarantees. But the distribution of mortgage risk among institutions after the market has been privatized is unknown. Capital requirements are significantly higher, however, for insured depository institutions than for the housing GSEs. Those requirements provide the federal government with some protection from depository risk. In addition, the government already recognizes and reserves funds for losses in its on-budget programs. Finally, the claim that deposit-insurance and federal-guarantee policy needs to be improved is a call for their reform; it is not an argument against reforming GSE policy.
Impose a Cost-of-Capital Equalization Fee on Debt
A capital-cost equalization fee might be levied on the average volume
of debt that each GSE had outstanding each year. Such a fee would recover
some of the benefit on the most deeply subsidized activity of the GSEs.
It would target benefits as well as encourage the GSEs to focus more on
their MBS line of business. A fee of 20 basis points would yield more than
$800 million per year based on the currently outstanding debt of the GSEs.
At the direction of the Congress, some or all of those collections could
be earmarked for targeted subsidies to low-income home buyers or for other
purposes. Exempting MBSs from the fee would provide an incentive for the
housing GSEs to shift funding toward less deeply subsidized forms of financing
but avoid the need to raise mortgage interest rates.
Improve the Ability of Government to Monitor the GSEs
The aim of improving the government's ability to monitor the GSEs is to reduce their ability to control the size of their federal benefit and to narrow the range of disputed subsidy estimates. That approach includes increasing required disclosures by the GSEs and conducting several market transactions to obtain more objective information about the GSEs and the subsidies they receive.
Increased Disclosures
Under this option, the GSEs would be required to report such information as:
The above information could be subject to audits and detailed review by the Office of Federal Housing and Enterprise Oversight.
Market Tests
This option would narrow the estimates of the GSE subsidy and subsidy pass-through rates by revealing them in market prices. It includes requirements for:
Privatize
Inasmuch as the GSEs are already privately owned, it seems odd to speak of privatization as a policy option. "Restructuring" is the preferred term used by one study.(3) Withdrawing federal sponsorship, or defederalization, is close to the essence of this option. However achieved, this policy would effectively eliminate the implied federal guarantee of GSE debt and MBSs.
Privatization could be undertaken abruptly by repealing the federal GSE charters and all the special provisions of law and regulation that convey the implicit guarantee. A sudden withdrawal of sponsored status--though it would make the decision more difficult to reverse--runs the risk of creating enterprises "too big to fail" and of subjecting the financial system to a shock from changes in the prices of many securities. A more gradual approach could address those difficulties.(4)
One of the thorniest issues facing privatization is the need to win the support of shareholders and management. The magnitude of the subsidy going to those interests makes it unlikely that stakeholders could avoid a loss if federal sponsorship was withdrawn. Accordingly, strong resistance to privatization is expected from the GSEs.
Based on recent experience with another GSE, the Student Loan Marketing Association, policies that reduce the federal subsidy can overcome such resistance.(5) Policies that would produce that result include reducing the size of the loan ceiling for conforming mortgages, imposing a cost-of-capital fee, limiting the ability of the GSEs to issue debt to finance their mortgage portfolios, mandating contributions to a low-income housing assistance fund, and imposing higher capital requirements for shareholders. Those policies could help the owners and managers of Fannie Mae and Freddie Mac to anticipate a net benefit from privatization.
Of course, such options beg a question: why would the GSEs agree to
those policies as a first step toward the withdrawal of their subsidy?
That admission simply acknowledges that once one agrees to share a canoe
with a bear, it is hard to get him out without obtaining his agreement
or getting wet. If the GSEs were to support privatization, they and the
Congress could certainly carry it out without financial disruption.
Fairness for the Housing GSEs
Although unilateral action by the government is envisioned under all
of the options, most of the proposals require tacit approval by the government-sponsored
enterprises. Fannie Mae and Freddie Mac's agreement is necessary because
they always have the option of seeking fully private status. Indeed, if
any policy imposed more costs on the GSEs than it provided in government
benefits, the GSEs' responsibility to their shareholders would require
them to pursue privatization rather than oppose it. Retaining the exit
option ensures that management and shareholders can avoid costs that exceed
benefits.
Remaining Questions
The Housing and Community Development Act of 1992, which mandated this study, directed that it examine the effects of privatization on six areas:
Although this report has supplied answers to most questions at various points, offering explicit responses provides a convenient summary of the Congressional Budget Office's principal findings.
Enterprise Operating Costs
The effect of privatization is difficult to predict. Repealing the exemption from state and local income taxes and SEC registration fees would raise GSE costs by perhaps $300 million per year. Yet the GSEs could presumably reduce spending for many purposes, especially lobbying and political risk-hedging. At the same time, competition on a level playing field with the private intermediaries should shed light on current expenditures that are not cost-effective. The net effect on operating costs of the GSEs could be either positive or negative.
Funding Costs
The effect of privatization on this area is unambiguous: capital costs of the housing GSEs would increase by at least 50 basis points on average--or by the amount of the federal funding subsidy.
Home Ownership
The effect on home ownership is ambiguous and depends on other policies adopted by the Congress. If, for example, the Congress decided to continue the subsidy to home buyers through other means, home ownership would most likely not be affected. Moreover, if some of the subsidy retained by the housing GSEs was effectively targeted toward low-income home buyers, home ownership among that group could rise.
Secondary Market Competition
All indications are that, after a period of adjustment, competition in the secondary market would increase in both the conforming mortgage and jumbo loan sectors.
Capital Requirements for the GSEs
On a per-dollar basis of assets and risk assumed, capital requirements would rise. GSE status implies an infusion of taxpayer equity that privatization would withdraw. However, the former GSEs would be significantly smaller after the adjustment to privatization was complete. Thus, their total dollar requirement for capital could be lower than under current policy.
Secondary Market Liquidity
The liquidity of a market refers to the ability to buy or sell large
quantities of securities without affecting price. Privatization should
not significantly affect the overall size of the secondary market. Big
volumes tend to promote liquidity. Yet the number of issuers of MBSs would
increase. That proliferation of issuers could increase the cost of evaluating
many separate security issues for investors. As a limit, the liquidity
of the secondary mortgage market should be no less than the liquidity of
the corporate bond market, in which large numbers of diverse debt securities
are bought and sold daily.
1. Congressional Budget Office, The Federal Home Loan Banks in the Housing Finance System (July 1993), pp. 21-24.
2. "GSE Chiefs Spurn Higher Capital Standards, Warning Costs Will Increase Mortgage Rates," Inside Mortgage Securities (April 19, 1996), pp. 9, 10.
3. Thomas H. Stanton, "Restructuring Fannie Mae and Freddie Mac: Framework and Policy Options," HUD Studies (May 1996), pp. 1-47.
4. Ibid.
5. Thomas H. Stanton, "Supplementary Analysis," HUD Studies (May 1996), pp. 78-79.
