Government Sponsored Enterprises:

INTRODUCTION  

by, SENATOR PETE V. DOMENICI

The savings and loan debacle has taught us some expensive lessons. First, a government guarantee (even an implicit or indirect guarantee) is no free lunch. People rely on these guarantees and expect to be compensated if an institution fails. Second, we've learned that we need to plan ahead. We cannot afford to issue trillions of dollars of federal guarantees and just  wait until something goes wrong .... Contingent liabilities, as we are all painfully aware, can turn into heavy taxpayer losses.

This report points to potential financial exposure government sponsored enterprises pose to the American taxpayer. Most Americans are probably not even aware of the existence of government sponsored enterprises (GSEs), let alone the fact that the federal government implicitly guarantees over a trillion dollars of their debt and securities.

GSEs are a hidden contingent liability. They are off-budget entities which are not included in the deficit or the budget; yet they carry with them the status of implied government backing. While it is unclear how "implied backing" is fmancially defined, the market.has taken the view that if GSEs get into financial trouble, Uncle Sam's deep pockets would bail them out.

Even though several GSEs are larger than the largest American financial institutions, some of these institutions are thinly capitalized and poorly supervised by the federal government. And while most of the enterprises seem to be in relatively good financial condition today, the federal government once again finds itself without the financial oversight taxpayers have a right to expect.

The report underscores the significance of federal support of GSEs and examines some of the problems with trying to enact necessary reforms before financial risk actually turns into taxpayer losses. As is often the case, the benefits of an implicit federal guarantee are immediately apparent - in the form of easy credit for homebuyers, farmers and students, for example - while the potential risks remain hidden from view until losses actually occur.

Federal regulatory policy toward GSEs was fashioned at a time when their activity was small. Today, however, both the enterprises and the risks associated with their activities are  enormous. Recognizing these risks, at my insistence, Congress adopted as part of the 1990  budget agreement a requirement that Congress produce this year legislation to strengthen the  safety and soundness of all GSEs. To date, no such legislation has been agreed to. Congress  cannot ignore this issue; it must act.

I commend the author and the Institute for Policy Innovation for this timely and insightful report, and would urge my colleagues to study it closely as we inevitably deliberate improved financial accountability of government sponsored enterprises.

 

EXECUTIVE SUMMARY

The S&L debacle painfully illustrates the potential problem government guarantee and insurance programs can hold for taxpayers. The dramatic growth of government sponsored enterprises (GSEs) raises a number of serious policy issues, including the potential for another multi-billion dollar taxpayer bailout.

Today, six GSEs are under increasing scrutiny because of the enormous taxpayer exposure involved in their activities. The six enterprises are the Farm Credit System (FCS), the Federal Agricultural Mortgage Corporation (Farmer Mac), the Federal National Mortgage Corporation (Fannie Mae), the Federal Home Loan Mortgage Corporation (Freddie Mac), the  Student Loan Marketing Association (Sallie Mae), and the Federal Home Loan Bank System (FHLBS).

Together these six GSEs represent a contingent liability of over one trillion dollars ($1,000,000,000,000) for u.s. taxpayers.

GSEs have a combination of public and private characteristics that make them difficult to describe as either federal entities or truly private corporations. On the one hand, GSEs are privately owned. They issue and sell stock to private individuals and institutions; their employees are exempt from federal civil service and procurement laws; they are free to make a profit; and their operations do not depend upon regular federal appropriations. On the other hand, GSEs are created by Congress; their debt obligations have many of the characteristics of federal Treasury securities; they may borrow from the Treasury; and, most importantly, their
obligations are implicitly guaranteed by the federal government. This last characteristic creates potentially open-ended claims on federal/taxpayer funds.

The implicit federal guarantee associated with GSE activity carries economic consequences that are comparable to federal deposit insurance. The implicit guarantee accorded GSE obligations is a product of special characteristics that may include the following:

• The U.S. Treasury Secretary approves their issue, interest rate, and maturity.

• They are exempt from full state and local taxation.

• They are exempt from SEC regulations, except as US government securities are regulated.

• GSEs are federally chartered instrumentalities of the United States.

• GSEs have lines of credit with the U.S. Treasury.

• GSEs are controlled by boards that include government appointed directors.

As with federal deposit insurance, federal support of GSEs weakens market discipline by permitting creditors to rely on the federal government for assurance that each GSE will meet its financial obligations. The implicit guarantee also provides considerable incentive for managers to seek extra returns by taking excessive risks. Heads, the corporation and its shareholders win; tails, the U.S. taxpayer is called upon to pay for any large mistakes. 

As with other institutions that deliver federal credit subsidies, GSEs also raise issues of competitive equity. The private business corporation faces competition from potential entrants who can decide to engage in the same business. By contrast, GSEs operate under unique federal laws that preclude other companies from receiving the implicit federal guarantee or the other competitive benefits that are granted to the GSEs. One result of this arrangement is that GSE growth can be quite rapid - leaving the GSEs often dominating the markets prescribed for them by federal law.

In fact, GSE growth has been phenomenal. The Congressional Budget Office calculates that GSE securities have grown from $38.9 billion outstanding in 1970 to $177 billion in 1980 to $980 billion in 1990. There are few constraints to prevent GSEs from issuing ever-increasing amounts of debt.

Several GSEs are already larger than the largest American fmancial institutions in their total volume of lending activity. Today, Fannie Mae and Freddie Mac fund over 25 percent of home mortgages in the U.S. and Sallie Mae funds over half of all guaranteed student loans. The $980 billion in outstanding GSE securities in 1990, for instance, compares with $726 billion in federally backed deposits in savings and loan institutions.

Because they are not fully subject to the forces of market discipline, and because their lending activities are implicitly guaranteed by the federal government, federal regulations are necessary to ensure their safety and soundness. To a large degree, federal regulatory policy toward GSEs is haphazard and outdated, conceived at a time when GSE activity was small and relatively insignificant. Today, however, the lack of an effective and comprehensive regulatory strategy leaves taxpayers potentially exposed to a multi-billion problem.

The federal government is beginning to assess ways of dealing with the trillion dollar taxpayer contingent liability involved in GSE activities. Ultimately, the question is a simple one: whether the clear dictates of public policy, to assure that GSEs are well-capitalized and effectively supervised, will outweigh the short-term political and financial interests of some GSEs and their shareholders. The policy debate has been joined, and it remains tobe seen how much federal policy makers have learned from the catastrophic failure of hundreds of thrift institutions. The question is whether policy makers are willing to take the needed action now, or whether they prefer to wait until financial risks turn into taxpayer losses.

 

I. GOVERNMENT SPONSORED ENTERPRISES: A TRILLION DOLLAR CONTINGENT LIABILITY

It is time for the federal government to come to grips with the immense contingent liability represented by government sponsored enterprises (GSEs). Today, GSEs are under increasing scrutiny because of the potential taxpayer exposure involved in their activities: the Farm Credit System (FCS) , the Federal National Mortgage Association (Fannie Mae), the Federal Home Loan Mortgage Corporation (Freddie Mac), the Student Loan Marketing Association (Sallie Mae), the Federal Home Loan Bank System (FHLBS) , and the Federal Agricultural Mortgage Corporation (Farmer Mac). Together, they represent a contingent liability of approximately one trillion dollars for U.S. taxpayers.

"GSE" can be defined as a privately owned, federally-chartered financial institution  with nationwide scope and specialized lending powers that benefits from an implicit federal guarantee to enhance its ability to borrow money. This definition helps to clarify the many common characteristics that GSEs have with banks and thrift institutions.

Like banks and thrifts, GSEs are privately owned. The largest GSEs are investor-owned with stock that trades on the New York Stock Exchange. Like banks and thrifts, GSEs operate under conditions set by federal law. Federal law defines and limits their lending powers. Fannie Mae and Freddie Mac purchase home mortgages in the secondary market, Sallie Mae funds student loans, the Farm Credit System provides agricultural loans, the Federal Home Loan Banks lend to thrift institutions and Farmer Mac will purchase or guarantee pools of agricultural mortgages in the secondary market.

Most importantly, GSEs benefit from an implicit federal guarantee with financial consequences comparable in essential respects to federal deposit insurance. (The nature of that implicit guarantee is explored in the appendix to this paper.) This federal support weakens market discipline by permitting creditors to rely on the federal government for assurance that each GSE will make good on its financial obligations. Reliance upon the federal government rather than upon a GSE' s creditworthiness raises the issue of moral hazard: a GSE can take much greater risks than could a completely private company in its line of business. A GSE would have an extensive market for its debt obligations and guarantees even if, for example, it were undercapitalized or did not adequately police the credit quality of the loans it purchased or guaranteed.

It should be emphasized that GSE managers, and managers of other fmancial institutions with federal backing, may in fact develop prudent business policies. However, the implicit federal guarantee does provide considerable incentive to seek extra returns by taking excessive risks. The financial distress of Fannie Mae in the early 1980s and the financial failure and federal bailout of the Farm Credit System in the mid-1980s shows that there is ample room for managers to take risks if they are so inclined.

As with banks and thrift institutions, GSEs raise two kinds of issues for the federal government: (1) the need to assure their safety and soundness despite operation of a federal guarantee that tends to undercut market discipline and distort capital markets; and (2) the need to specify the proper scope of their permitted lending and other financial activities. This report concentrates on the first issue, although the importance of the second is seen in its contribution to the failure of hundreds of thrift institutions.¹

GSEs also raise issues of competitive equity. The ordinary private corporation faces competition from potential entrants who can decide to engage in the same business. By contrast, GSEs operate under unique federal laws that preclude other companies from serving the same markets while receiving the same implicit federal loan guarantees and other competitive benefits granted to the GSEs.

Under their generous charters, GSEs are able to borrow virtually unlimited amounts of money in the federal agency credit market on favorable terms. They are exempt from state laws and taxes applicable to many competitors and some are exempt from federal taxes as well. As federal instrumentalities they are not subject to burdensome requirements such as state doing-business laws. They are also free from federal securities registration requirements and, indirectly, are permitted to avoid most state securities requirements. These benefits increase the ability of a GSE to operate without many of the constraints and costs facing private competitors.

One result of this freedom of GSEs from effective competition is that their growth can be quite rapid. They often dominate the markets prescribed for them by federal law. Sallie Mae, for example, grew from $2.8 billion in assets in 1980 to $41.1 billion in 1990. Sallie Mae today funds over half of all guaranteed student loans. Fannie Mae and Freddie Mac today are huge institutions. At year-end 1990, for example, Fannie Mae had $133 billion of assets plus $300 billion in guaranteed mortgage-backed securities, and Freddie Mac had $41 billion of assets plus $316 billion in guaranteed securities. This makes them larger than the largest U.S. banking institutions in their total volume of lending activity (assets plus guarantees). Today, Fannie Mae and Freddie Mac fund over one out of every four residential mortgages in the United States. Fannie Mae grew by $80.4 billion and Freddie Mac by $48.6 billion in 1990 alone.

Table I shows a slightly different measure of GSE growth, the volume of GSE securities outstanding over the years. The Congressional Budget Office calculates that GSE securities have gone from $38.9 billion outstanding in 1970 to $176.9 billion in 1980 and $980.1 billion in 1990. That 25-fo1d growth over 20 years includes creation of new GSEs, notably Freddie Mac in 1970 and Sallie Mae in 1972.

Except for limitations imposed by Congress or a regulator, there seem to be few constraints to prevent GSEs from issuing ever-increasing amounts of debt and other securities. On the lending side, GSEs seem limited mainly by the size of their statutorily permitted markets. By the end of 1990, outstanding GSE securities amounted to $980 billion, compared to $726 billion in federally backed deposits in savings and loan institutions. Table 1 shows how several of the GSEs - Fannie Mae, Freddie Mac, and Sallie Mae - have more than doubled in size in five years. The rate of GSE growth is likely to present  a serious policy issue for Congress, 'especially because of the taxpayer financial exposure that is involved in GSE activities.

 

II. SAFETY AND SOUNDNESS OF GSEs: AN ANALYTIC FRAMEWORK

The Risks Involved in GSE Lending

When the federal government extends its credit support to a privately owned and controlled company - such as a bank, thrift, or GSE - the government must protect itself against excessive risk-taking that could increase the financial exposure of taxpayers. It is useful to analyze the kinds of financial risks that financial institutions can take, the way that federal backing weakens the market discipline that otherwise would limit such risk-taking, and the need for effective federal supervision.

Banks, thrifts and GSEs are subject to a variety of risks. Some of these are associated with lending activities generally, and others relate to specialized kinds of lending by a particular institution. Different analysts may categorizerisks somewhat differently, and new forms of risk are recognized as markets change. Two kinds of risk involve more fundamental factors than the others: the institution's management, and its market.

• Management Risk: Management risk includes the problem that some GSE managers
  may not be of the highest quality. While managers of most GSEs today appear to be
  of high quality, unforeseen problems can arise because of increasing financial
  pressures or through changes in control. Management risk may be the single most
  important factor in the performance of a financial institution.

• Market Risk: GSEs are specialized lenders, and this can create vulnerability if the .
  particular sector they serve suffers economic adversity. The Farm Credit System, for
  example, was harmed by the decline in American agriculture in the early 1980s.

Financial institutions are also susceptible to other kinds of risk. These include:

• Credit Risk: Credit risk is the risk that a borrower will fail to make timely payments
  of principal and interest on a loan held or guaranteed by a GSE or other lender. If
  a borrower defaults on a loan, the GSE must supervise the process of foreclosing on
  collateral or otherwise reducing its losses.
  
  
• Interest Rate Risk: Interest rate risk involves the risk of losses when interest rates
  change. If an institution is funding a portfolio of long-term loans with short-term
  borrowings, or vice-versa, the institution can either profit or lose when there are
  changes in the so-called yield curve - the relationship between long- and short-term
  interest rates. This kind of interest rate risk affects a GSE's portfolio lending, but not
  its guarantees of mortgage-backed securities. 
  
  
• Operations Risk: Operations risk includes the ability of a lender to manage its
  business effectively. GSEs are multi-billion dollar institutions that often grow by
  many billions of dollars a year in net new business. Sophisticated computer operations
  are required to assure that the lender manages the literally millions of transactions that
  may occur every month to process loan payments and distribute payments in turn to
  holders of guaranteed mortgage-backed securities and GSE obligations. Another part
  of operations risk is the problem that has occurred for some institutions, for example
  the Continental Illinois National Bank, when their portfolios contain assets of
  significantly different quality from what their management information systems lead
  them to believe.

The Implicit Federal Guarantee and its Effects on GSE Risk Taking

While managers in fact may be prudent, it must be recognized that an incentive to take excessive risk is inherent in the implicit federal government backing for banks, thrifts, and GSEs. Just as became clear in the thrift industry debacle, the federal government guarantee removes much of the usual market discipline from GSEs.

Significant market discipline normally comes from investors when a company issues debt obligations or guarantees securities. For the ordinary company, without federal backing, debt holders and purchasers of guaranteed securities will limit the degree of risk by demanding increasing returns for themselves as shareholders increase risk and attempt to increase leverage. Before investors rely upon the corporation's obligation or guarantee, they require assurances about the company's fmancial backing. Investors check not only the balance sheets of the corporation providing the obligation or guarantee, but also the credit quality of the assets standing behind the obligation or guarantee. Before purchasing unsecured debt obligations of a private corporation, investors require assurances about the corporation's balance sheet, including the quality of assets it holds, the corporation's debt-to-equity ratio, and the corporation's general creditworthiness.

Investors in GSE securities are unlikely to provide such market discipline because they can simply rely upon the implicit federal backing for the GSE and not on the company's  balance sheet and general creditworthiness. ²

This reliance upon the federal government rather than upon the GSE's creditworthiness means that a GSE can take much greater risks than could a private company in its line of business. A GSE would have an extensive market for its debt obligations and guarantees even if, for example, it were undercapitalized and did not adequately police the credit quality of assets it bought or securities it guaranteed. 

These concerns are not merely hypothetical. The banks of the Farm Credit System adopted an average-cost pricing policy for loans that permitted FCS institutions to use long-term debt to fund variable rate loans on the basis of FCS average debt funding costs. As interest rates rose during the 1970s, new debt tended to be priced somewhat higher than the average price of outstanding debt. By pricing loans based on the average rather than the new debt costs, FCS institutions could offer borrowers lower priced loans than were available from competing private lenders. By 1981, FCS banks were pricing their real estate loans almost five percentage points below the rates of their commercial competitors.

This policy provided immediate rewards to FCS managers and their cooperative borrower/shareholders. Managers were rewarded for increasing market share and were able to construct lavish new office buildings, hire staff, and generally expand the organization. FCS cooperative shareholders, unlike shareholders in corporate institutions, received their benefits largely in the form of lower loan rates, rather than from stock dividends, thus giving them an incentive to take out a greater than optimal volume of loans from FCS banks.

As the agricultural economy went into depression, loan delinquencies and defaults rose significantly. Then interest rates came down, and average cost pricing became impossible to sustain. The high volume of outstanding long-term debt issued by the FCS at times of high interest rates soon reflected itself in higher rates for borrowers, driving credit worthy borrowers to refinance their loans and take their business away from the FCS and to competing lenders.

Starting in 1985, the FCS began reporting huge annual losses, and the Governor of the Farm Credit administration announced that the Farm Credit System required a massive infusion of federal funds within 18-24 months to remain in business. Even after this announcement, the implicit federal guarantee was perceived as being so strong that FCS could continue selling its debt less expensively than most private corporations. In 1987 legislation, Congress created a $4 billion loan fund to pay for FCS losses resulting directly from excessive FCS risk-taking that, initially at least, provided substantial benefits to FCS managers and shareholders.

Fannie Mae underwent a similar experience, but fortunately with a happier ending.In the 1970s when short-term interest rates were lower than long-term rates, Fannie Mae was able to purchase billions of dollars of long-term mortgages and fund them with short-term debt. Because short-term debt was less expensive than long-term debt, at least for the moment, shareholders received immediate returns, above the normal returns that could be expected if funding had been done on a more matched maturity basis. This funding system was only possible because of the implicit government guarantee that allayed concerns of Fannie Mae's debt holders about the extra risk involved. Without the implicit guarantee, these debt holders would have become increasingly unwilling to lend money inexpensively to Fannie Mae because of the growing interest rate risk in its operations.

Starting in 1979, short-term borrowing costs rose dramatically relative to long-term rates. Fannie Mae's debt costs soon exceeded returns on its portfolio oflong-term mortgages bearing relatively low interest rates. By 1981, Fannie Mae had a market-value negative net worth of $10.8 billion. Fortunately, with energetic new management strategies and declining interest rates, Fannie Mae avoided the fate of the Farm Credit System.

It is easy enough for an ordinary company to fail in the marketplace. Without  marketplace discipline fully imposed through higher costs and an ultimate limit on the ability to sell debt obligations and other securities, GSEs are at risk in this regard. The government guarantee distorts market signals, permitting even GSEs with poor balance sheets and management strategies to continue selling obligations and guarantees at favorable interest rates.

Both the Farm Credit System and Fannie Mae consistently recorded substantial profits in the years before the potential risk exposure turned into actual losses. If a high-risk strategy for such a GSE is successful, benefits go primarily to the shareholders. If the high- risk strategy fails, the federal government is under virtually irresistible pressure to make good on its guarantee. The failure of hundreds of thrifts with federally insured deposits  provides perhaps the most stark example of the way federal backing undermines usual market discipline that otherwise could limit losses when an institution falters or fails.

Regulation of Safety and Soundness to Compensate For Missing Market Discipline

Because the federal government's guarantee removes much of the usual market discipline, effective government regulation may help contain risk-taking and assure long-term viability of GSEs and other institutions with federal backing.

Regulation is certainly not an assured solution for problems that arise when the government's implicit guarantee distorts normal market incentives. As can be seen in the regulatory history of the Farm Credit System or of thrift institutions governed by the Federal Home Loan Bank Board, regulators live in the same political and psychological environment as the institutions they regulate.

In particular, Congress may be sensitive to special interests eager to increase father than reduce the federal government's risk exposure. For example, even as the Farm Credit System Banks struggled to keep their portfolios above water, desperate farm borrowers were able to persuade Members of Congress to pressure Farm Credit Banks to be lenient in handling delinquent loans. Similarly, the Federal Home Loan Bank Board came under considerable political pressure to refrain from promptly closing thrift institutions with negative net worth.

There is another problem with regulation. Regulators have not been very successful in detecting high risk problems in time. Regulators tend to be attuned to forms of risk which caused problems in the past rather than today's emerging problems. Indeed, management has an incentive to shift risk away from regulated and into unregulated areas. Failure of the Continental· illinois National Bank in 1984 provided a large example of the fact that regulation cannot protect against risk exposure of the federal government. Failure of hundreds of thrift institutions provides a sobering lesson about the potential costs of letting  regulators be dominated by the regulated institutions. Nevertheless, even with the emergence of possible new tools such as risk-related federal insurance premiums, prudent supervision of GSEs seems to be a limited but necessary substitute for the market discipline that has been weakened by governmental credit support.

In their 1978 article, Black, Miller, and Posner argue that when the government acts as a guarantor, government regulation is called for. The government should try to control unnecessary risk-taking just as a private guarantor imposes control upon firms whose obligations it guarantees and the private lender oversees the creditworthiness of borrowers.

"[E]fficient government regulation of banking would resemble the measures adopted by a private lender in a comparable position and . . . deviations from the private model are prima facie questionable." ³

A private lender is concerned about capital and leverage of the borrower and about the borrower's general management ability. The lender will require disclosure of material events and may even insist on direct supervision of the borrower's business. The lender does not try to substitute for the borrower's judgment in general business decisions, but tries to prevent the borrower from increasing the risks of business activities to benefit the borrower at the lender's expense.

The problem is, of course, that federal supervision is an inadequate substitute for the market discipline that governs private companies that operate without a federal guarantee. As was seen with the thrift debacle, the political process may compound taxpayer exposure once an institution gets into fmancial trouble. Unlike the marketplace, the government is likely to try to maintain rather than close a failing institution. 

III. FEDERAL SUPERVISION OF GSE SAFETY AND SOUNDNESS TODAY

It turns out that federal supervision of safety and soundness of most GSEs today falls far short even of the quality of bank: and thrift supervision. Information is generally limited and regulators often lack resources, the mandate, and the full complement of statutory powers to address safety and soundness issues.

The Quality of Federal Information

There is no single source of information within the federal government about GSEs.This is unfortunate, because GSEs involve issues oflaw, economics, and finance that corne together in unusual ways. Statistics are often unavailable except from the GSEs themselves.

Information about various GSEs is collected by a variety of federal agencies. HUD collects information from Fannie Mae and Freddie Mac; the Farm Credit System receives reports of condition from farm credit institutions; and the Federal Housing Finance Board collects information from the Federal Horne Loan Banks. Within the legislative branch, the Congressional Budget Office and General Accounting Office are increasingly allocating resources to analyzing GSEs and their activities.

The Supervision of GSE Safety and Soundness Today

Even with the regulatory failures of the Federal Horne Loan Bank Board and the Farm Credit Administration in the rnid-1980s, the lessons of those failures have not been applied to regulation of GSE safety and soundness.

One place where the lesson has been learned is the newly strengthened Farm Credit  Administration (FCA). FCA regulates the farm credit institutions and Farmer Mac under statutory powers similar to those of federal bank and thrift regulators. It can bring cease and desist proceedings, suspend or remove FCS institution officers and directors, impose civil fines, set and enforce capital adequacy standards, and appoint conservators or receivers. The agency is permitted to assess the regulated FCS institutions for supervisory costs, including the cost of examinations, and to pay FCA expenses from those assessments. 

By contrast, the Department of Housing and Urban Development has authority, but, until recently at least, has shown little inclination to supervise safety and soundness of Fannie Mae. HUD has general regulatory authority to see that the purposes of the Fannie Mae and Freddie Mac charters are carried out, but it lacks a clear mandate to supervise Fannie Mae's or Freddie Mac's fmancial soundness. HUD has never used its examination authority over Fannie Mae, even though that authority has been available since 1968. HUD has tended to concentrate on national housing policies, such as Fannie Mae's support of low-income housing in the late 1970s, rather than on narrower matters such as Fannie Mae's financial safety and soundness. Unlike other financial institution regulators, HUD is not authorized to assess Fannie Mae or Freddie Mac the cost of maintaining a capable regulatory staff that is compensated at competitive levels. HUD is likely to be overwhelmed by responsibility for two GSEs with combined lending amounting to well over half a trillion dollars.

The Financial Institutions Reform, Recovery and Enforcement Act of 1989 (FIRREA) created a new federal agency, the Federal Housing Finance Board, to oversee the safety, soundness, and activities of the Federal Home Loan Bank System. The FHFB today has only a few examiners to inspect the entire FHLBS and its $166 billion of loan assets. Finally, Sallie Mae has no financial regulator at all.

Current GSE Capital Requirements

The federal government has not set consistent capital standards for GSEs. Fannie Mae and Freddie Mac are not subject to effective capital requirements and this is reflected in the low capitalization of the two GSEs. At year-end 1990 Fannie Mae had only $3.9 billion of shareholder capital to support $433 billion of assets and guaranteed securities; Freddie Mac had only $2.1 billion of shareholder equity to support $357 billion of assets and guarantees. This is substantially below the capital requirements established for commercial  banks by the FDIC, Comptroller of the Currency, and the Federal Reserve Board, or for thrifts by the Office of Thrift Supervision.

The Farm Credit Administration is responsible for setting capital standards for Farm Credit System institutions. FCA has promulgated regulations requiring that most FCS institutions meet minimum risk-based capital requirements. Those requirements involve a minimum ratio of permanent capital to risk-weighted assets of seven percent, to be achieved by 1993. The Farm Credit Administration has not set any capital standards for Farmer Mac, the newest government sponsored GSE, and lacks express authority to do so by regulation.

The Federal Home Loan Bank System is required by law to maintain a capital reserve, and has maintained substantial capital, amounting to over seven percent of assets at year-end 1990. Sallie Mae has no capital requirements at all, but on its own has maintained capital that at year-end 1990 amounted to 2.8 percent of its assets. 

IV. IMPROVING SUPERVISION OF GSES: REPORTS AND RECOMMENDATIONS

After years of neglect, the federal government has now begun to recognize the immense taxpayer contingent liability for GSEs. In 1990 and 1991, the U.S. Treasury Department published a Report of the Secretary of the Treasury on Government Sponsored Enterprises. The U.S. General Accounting Office (GAO) also published reports on GSEs in 1990 and 1991.⁴ In 1991 the Congressional Budget Office (CBO) published a report, and the Administrative Conference of the United States also made recommendations. ⁵

The reports and recommendations recognize significant shortcomings in today's federal supervision of safety and soundness of most GSEs. They recommend that the government set meaningful capital standards and supervise safety and soundness effectively. Application of such capital standards would require increases in capitalization of Fannie Mae and Freddie Mac (and Farmer Mac, when it begins operations).

The reports do differ in their judgments about the appropriate agency to supervise safety and soundness of GSEs. The U.S. General Accounting Office strongly favors a single regulator structured in a manner similar to the Federal Deposit Insurance Corporation (FDIC) to supervise safety and soundness of all GSEs; oversight of programmatic aspects of GSE activities would be left with program departments and agencies. By contrast, the Treasury Department in 1991 proposed that an "arm's-length bureau" be established at HUD for Fannie Mae and Freddie Mac, that Treasury supervise safety and soundness of Sallie Mae, and that the Federal Housing Finance Board and Farm Credit Administration continue to supervise the FHLBS and FCS (including Farmer Mac), respectively.

The Congressional Budget Office, while not making an express recommendation, found the GAO approach superior to that of the Treasury, and favored creation of a central regulator, especially compared with leaving oversight of Fannie Mae and Freddie Mac at HUD.⁶ The Administrative Conference of the United States did not express an opinion as to where the regulatory authority should be located.

Given the conspicuous financial and management failures in the administration of HUD's own programs,⁷ it is curious that Treasury would select that department to supervise the two largest financial institutions in the United States, Fannie Mae and Freddie Mac. In contrast to a centralized regulator with considerable institutional capacity and independence, as was recommended by the GAO, HUD is particularly vulnerable to political pressure applied through the housing subcommittees of the House and Senate Banking Committees. The subcommittees have traditionally been very congenial to Fannie Mae and Freddie Mac and to the housing interest groups supportive of the two GSEs. Moreover, the Treasury proposal left considerable authority with the HUD Secretary to affect the activities of the new safety and soundness bureau. Under the Treasury proposal, the Secretary would have authority to approve regulations and the new bureau's budget, for example.

Perhaps mindful of political pressure that it itself had endured, the Treasury highlighted the problem of regulatory capture in its 1991 report:

The principal GSEs are few in number; they have highly qualified staffs; they have
strong support for their programs from special interest groups; and they have
significant resources with which to influence political outcomes. A weak fmancial
regulator would find GSE political power overwhelming and even the most powerful
and respected government agencies would fmd regulating such entities a challenge.⁸

Congress is now beginning to translate the Treasury, GAO, and CBO reports and the Administrative Conference recommendations into legislation. It is useful here to present a conceptual framework for improving federal supervision that draws upon elements of each of the reports and upon the experience of bank and thrift supervisors.

The Need for Federal Supervision

There are two reasons why active federal financial supervision of GSEs is necessary.First, monitoring by outside parties may be inadequate to detect excessive risk-taking.⁹ The government itself must have systems in place to assure effective surveillance of risk-taking. This requires skilled examiners with timely access to necessary information that may not be readily available to outside analysts. Second, whether or not excessive risk-taking is detected beforehand, if losses occur the government may need to act quickly to protect its financial
position. 

As a general rule, regulatory involvement in the affairs of a fmancial institution should be roughly proportional to the extent that the activities of that institution threaten financial loss to taxpayers. When the institution is well run and well capitalized, as indicated by reports and examinations, regulatory involvement should be minimal; as perceived safety and soundness problems arise, regulatory involvement should increase. If an institution begins to falter or fail, the government must be empowered to act promptly to reorganize the institution or even to wind up its affairs.

Disclosures and Reports

Officials of a number of GSEs use sophisticated financial models, including scenario analyses, that could lend themselves to regulatory reporting and disclosures.

One approach to regulating at least some of the GSEs that serve secondary market functions (Fannie Mae, Freddie Mac and Sallie Mae are promising candidates) would be to require them to report the application of these models, or of models devised by the regulator, to GSE activities {and to proposed new activities) as a means of assessing possible exposure to credit risk, interest rate risk, and other risks. The regulator would assure that appropriate models were utilized and that they were revised or replaced as necessary.

A Treasury Department proposal to use nationally recognized credit rating agencies such as Standard & Poor's would provide an additional valuable source of information on each GSE. The rating agencies use a combination of quantitative analysis and qualitative assessment of management which could provide useful information about the safety and soundness of each GSE and about changes over time.

If such a sophisticated approach were used, two issues would require active attention by the regulators: First, models are inherently imperfect; adjustments must be made as markets and financial knowledge change. The regulator must have the technical capacity and legal authority to generate and evaluate appropriate models and to deter institutions from providing inaccurate, incomplete, or otherwise useless information. Second, especially if an institution gets into trouble, self-reporting can become quite problematic. The regulator must be prepared to examine a GSE thoroughly and take other prompt action if there are signs that self-reporting is not effective. Finally, for GSEs such as the Farm Credit System and Federal Home Loan Bank System, bank-type disclosure and reporting requirements and bank-type examinations are appropriate.

Examination

To the extent that a GSE joins in a process of providing accurate and sophisticated risk assessment models, traditional bank-type examination procedures can become correspondingly more sophisticated and less cumbersome. The examiners may be able to concentrate on testing whether models and management information systems accurately reflect the risks and kinds of exposure that have been reported to the regulator. One criticism of examination of the Continentallllinois National Bank before its failure in 1984 was that federal examiners concentrated on loans identified by Continental as being of poor quality rather than undertaking a statistical sample of Continental's entire business. The federal government will require highly qualified staff familiar with the markets served by each GSE and with the activities of comparable financial institutions (such as mortgage insurance companies, for example, with respect to Fannie Mae, Freddie Mac and perhaps Farmer Mac).

Another issue concerns management quality. Management risk is both the most important risk associated with financial institutions, and is also the risk least possible to quantify. Again, the federal government will require examiners of considerable sophistication - preferably including some who have had experience assessing the management of large money center banks - to form qualitative judgments about the management of each GSE. The rating companies are experienced in assessing management quality and, again, the Treasury proposal to require annual ratings of each GSE could supply useful supplementary information.

Enforcement Powers

Discussing the regulators' enforcement powers, it is useful to distinguish institutions with low-risk profiles from others. As a general rule, a regulator's involvement with an institution must increase as the institution's capital declines below prescribed levels or as other risk factors increase. For the most safe and sound GSEs, exchanging safety and soundness information with the regulator without acrimony in an atmosphere of arm's length respect, enforcement powers are unlikely to be called upon.

As an institution becomes increasingly risky, either intentionally or as a result of adverse market circumstances, or as capital fell below prescribed levels, the relationship of the GSE and the safety and soundness regulator would begin to look more like the relationship of banks to their regulators. Bank-type examinations would begin to replace primary reliance on sophisticated computer models and other informal communications; regulators might require the institution to establish business plans intended to restore required capital levels. 

At some point, if the regulator were to face systematic non-responsiveness to its legitimate requests or detects significant unsafe or unsound practices, the regulator must be empowered to prescribe financial requirements and enforce them with a cease and desist order. Enforcement powers of federal bank regulators have been tested over the years, both politically and in litigation. These powers provide a range of sanctions to be applied in the event that a financial institution engages in unsafe or unsound practices or violates other legal standards that are being increasingly applied m litigated cases. Abuses of regulatory discretion are reviewable in court.

As has been learned from experiences with banks and thrift institutions, enforcement powers must be strong enough to permit prompt action while protecting the rights of managers and owners of affected institutions and providing ample and prompt opportunity to obtain redress from regulatory abuse. Moreover, use of the statutory standards already applicable to banks, thrifts, and farm credit institutions will provide the affected parties with some sense of their rights and authority in the event that enforcement powers are contemplated or applied.

Note that so long as a GSE maintained a low risk profile and adequate capital, use of banking-type enforcement powers would not permit the regulator to second guess or micro-manage a GSE's business plans. Instead, the regulator would concentrate upon designated activities or conditions considered to constitute unsafe or unsound practices and seek to prohibit those explicitly. The regulator should leave business decisions to the regulated institution, except as they constitute a direct fmancial risk to taxpayers.

Appointment of a Conservator or Receiver

A major lesson from the thrift industry fiasco is that the likelihood of imprudent behavior increases substantially as the proportion of shareholder-contributed capital decreases. The federal government must be able to intervene before the net worth of the federally backed institution drops to zero. At that point, the problems that have brought a financial institution's net worth to zero are likely to continue and create a substantial negative net worth. Once again, the standards applicable to bank regulators, and part of the analogous provisions available to the Farm Credit Administration, provide a useful benchmark for regulatory intervention. As with respect to other bank regulatory powers, the statute and case law on conservators and receiverships specifies the limits of regulatory discretion, and affords affected managers and owners the opportunity for redress in court.

In practice, of course, Congress is likely to become involved in any regulatory decision to appoint a conservator or receiver for a faltering or failing GSE, if only to assure that the public purposes of that institution continue to be carried out. Congressional involvement can range from providing fmancial assistance that restores the ability of an institution to function to assuring that other institutions serve the public purposes of a failed institution. Congress did the former with respect to the financial failure of Farm Credit System in the 1980s and the latter in its monitoring of the liquidation of the Federal Land Bank of Jackson, Mississippi in 1988. 

v. CONCLUSION: IMPLEMENTING EFFECTIVE REFORMS

There are important lessons to be learned from the failure of hundreds of thrift  institutions- at a cost to taxpayers of hundreds of billions of dollars. First, the government's guarantee (for thrifts through federal deposit insurance) creates powerful distortions in the usual market incentives. Second, effective federal supervision and meaningful capital standards help compensate for the market discipline that is undercut by the federal guarantee.

It is now time to apply these simple but expensive lessons to GSEs. GSEs represent over a trillion dollars of contingent taxpayer liability, backed by an implicit federal guarantee that is similar in essential respects to federal deposit insurance.

The political lessons of the thrift debacle are also instructive. It is not easy to implement effective reforms before financial risk actually materializes in the form of substantial losses. Even after many thrift institutions had lost substantial net worth, they retained considerable political influence and lobbied effectively to forestall reforms. In 1986, for example, the thrift industry's powerful lobbying arm, the U.S. League of Savings Institutions, managed to block recapitalization of the failed Federal Savings and Loan Insurance Corporation. Recapitalization would have meant that the federal government could have used the new funds to close insolvent thrifts; without the new funds, the thrifts were permitted to stay open and continue activities that greatly compounded taxpayer losses.

The Farm Credit System provides another example. Only when the Farm Credit System failed in 1985 did the Congress enact a substantial overhaul of the supervisory mechanism, transforming the Farm Credit Administration from a captured regulator into an arm's length agency with enforcement powers similar to those of federal bank regulators.

Today GSEs are incredibly powerful politically.¹⁰ A financial institution with tens or hundreds of billions of dollars of lending activity is able to obtain the services of the finest lobbyists available. Constituents who are served by these institutions also constitute a powerful force on their behalf. Finally, commercial competitors of GSEs are rarely to be seen in the political process. This is because such competitors - for example, thrift institutions that compete with Fannie Mae or Freddie Mac as mortgage lenders - also tend to be customers of the same GSEs. It is a rare financial institution that is willing to risk incurring the displeasure of a GSE with which it wants to conduct any significant amount of business. 

Ultimately, the question becomes a simple one: whether the clear dictates of public policy, to assure that GSEs are well-capitalized and effectively supervised, will outweigh the short-term political and financial interests of some GSEs and their shareholders. The policy debate has been joined, and it remains to be seen how much federal policy makers have learned from the catastrophic failure of hundreds of thrift institutions. The question is whether policy makers are willing to take the needed action now, or whether they prefer to wait until financial risk turns into actual losses before they institute the full range of necessary financial reforms.

 

APPENDIX

The Implicit Federal Guarantee of GSE Obligations

GSEs benefit from implicit federal government backing for their corporate guarantees and obligations that is different in form from federal deposit insurance but similar in its economic consequences. This federal credit support permits GSEs to obtain virtually unlimited funds at very low cost, close to the rates at which the U.S. Treasury itself borrows money. The federal government provides this credit support through an ingenious device. Even though GSEs are privately owned and managed, federal law gives their obligations the financial attributes of Treasury obligations; similarly, the law grants guaranteed securities the attributes of federally guaranteed securities.

Like U.S. Treasury obligations, GSE obligations, with some variations, have the following characteristics:

• They may be issued only upon approval of the Secretary of the Treasury, who also
  often approves terms such as interest rate and maturities;

• They are exempt from regulation by the SEe except to the extent that U.S.
  government securities are regulated;

• They are lawful investments for federally supervised institutions, including banks,
  thrift institutions, and credit unions and have favorable government-type status in the
  portfolios of these institutions. (For example, shorter-term GSE obligations may be
  used to meet liquidity requirements of thrift institutions, and national banks may
  invest and deal in GSE obligations without limit);

• They are lawful investments for federal fiduciary, trust, and public funds;

• They are issuable and payable through the facilities of the Federal Reserve Banks;

• They are eligible collateral for Federal Reserve advances and discounts and are
  eligible to be bought and sold in Federal Reserve open market operations; and

• For most GSEs, they are exempt from full state and local taxation.

The one big difference between GSE securities and U.S. Treasury obligations is the nature of the issuer; GSEs are privately owned and are not part of the federal government. To strengthen the perception of governmental backing, the GSEs have additional characteristics:

• They are federally chartered instrumentalities of the United States;

• They have a line of credit with the U.S. Treasury; and

• They are controlled by boards that usually include some governmentally appointed
  directors. 

Taken together, these attributes amount to an implicit federal guarantee that GSEs will  not be allowed to default on their obligations. The federal government makes a strong statement to investors by conferring on GSE securities the same preferred investment status as U.S. Treasury obligations. The exemption from the usual SEC registration laws removes investor protections considered important for the usual corporate securities. The exemption from investment restrictions on banks and thrift institutions is otherwise limited to federally backed securities. Investors perceive that the government would not permit these exemptions from basic investor protection unless GSE securities were extremely safe.

This means that investors look primarily to the implicit federal backing as a guarantee of a GSE's creditworthiness, rather than looking toward its balance sheet. Thus, while borrowing costs did rise somewhat for the Farm Credit System, even after the GSE recorded $4.6 billion of losses in 1985 and 1986, Farm Credit System obligations remained eligible investments for AAA-rated debt.

Over time, the market can become even more confident about the government's likelihood of backing obligations of GSEs. As the value of outstanding GSE obligations increases, so does the inability of the federal government to intimate that it would not stand behind this debt.

When the Continental illinois Bank failed in 1984, for example, the federal government stood not only behind FDIC-insured deposits, but also protected uninsured bank creditors, and even creditors of its parent holding company. The Comptroller of the Currency told Congress of fears of a domestic and possibly an international fmancial crisis if all Continental creditors were not protected. The Comptroller of the Currency concluded that because of their size alone, the federal government cannot permit the failure of any of the nation's largest money center banks, many of which are significantly smaller than GSEs. Continental illinois was the nation's seventh largest bank. It was a $41 billion institution --  about a tenth of the size of Fannie Mae today.

Because the federal backing for GSE obligations is implicit rather than express, the market tends to require a slightly higher return on GSE obligations than Treasury debt. Treasury obligations compete with GSE obligations in the same federal agency credit market and the higher yielding GSE securities in turn increase the cost of Treasury borrowing. As the GSEs continue to grow in size, the impact on Treasury borrowing costs can be expected to increase accordingly.

In summary then, the federal backing of GSE obligations may be implicit, but it is very real. As in the case of the Farm Credit System, taxpayers have literally billions of dollars at stake if a GSE fails to meet its obligations. 

ENDNOTES

1. Banks, thrifts, and GSEs are mercantilist corporations; unlike the usual state-chartered corporations, their activities are specified by federal law. Changes in their lending powers involve decisions of Congress and the Executive Branch that, to a greater or lesser extent, involve political rather than purely public or commercial decisions. This dependence on political considerations can make banks, thrifts, and GSEs vulnerable to financial loss if economic circumstances change but Congress declines to permit corresponding changes in their enabling legislation. See Hernando deSoto, The Other Path, Chapter 7, "The  Parallel With Mercantilism," (New York: Harper & Row), 1989. 

2. See e.g., Standard & Poor's Corporation, "AAA Investments' Implied U.S. Support," Credit Week, pp. 19-20, (November 28, 1988). 

3. Fischer Black, Merton H. Miller, and Richard A. Posner, "An Approach to the Regulation of Bank-Holding Companies," Journal of Business, Vol. 51, No.3 (1978), p. 382. 

4. The 1991 GAO Report is titled Government Sponsored Enterprises: A Framework/or Limiting the Government's Exposure to Risks. GAO/GGD-91-90 (May 1991). and is available by request to the GAO.

5. The CBO Report is titled Controlling the Risks of Government Sponsored Enterprises (Apri11991), and is available by request to the CBO. The recommendations of the Administrative Conference of the United States, titled "Improving the  Supervision of the Safety and Soundness of Government-Sponsored Enterprises," have been published in the Code of Federal Regulations, 1 CFR §305.91-6.                                                                                                                                        .

6. Testimony of Robert D. Reischauer, Director. Congressional Budget Office, Before the Subcommittee on Government Information and Regulation. Senate Committee on Governmental Affairs. July 18. 1991.

7. See e.g., Ronald C. Moe, "The HUD Scandal and The Case for an Office of Federal Management," Public Administration Review, Vol. 51 (July/August 1991), pp. 298-307.

8. 1991 Treasury Report, p. 8.

9. See, e.g., Richard Randall, "Can The Market Evaluate Asset Quality Exposure in Banks?" New England Economic Review, July/August 1989, pp. 3-24; and Katerina Simons and Stephen Cross, "Do Capital Markets Predict Problems In Large Commercial Banks?" New England Economic Review, May/June 1991, pp. 51-56.

10. Carol Matlack, "Getting Their Way," National Journal, October 27, 1990, pp. 2584-2588.