Lapas attēli



In seeking to limit the taxpayers' exposure to the GSE's, Congress has enacted three disclaimers of liability. But the phrasing of these disclaimers, far from hindering the GSEs' double game, fits it neatly.

First, the Federal Housing Enterprises Financial Safety and Soundness Act of 1992 declares “neither the Enterprises [that is, Fannie and Freddie) . . nor any securities or obligations issued by the Enterprises . are backed by the full faith and credit of the United States.” 12 U.S.C. $ 4501(4). But this disclaimer merely restates the obvious: That the Government has no formal, legally enforceable liability for the GSEs' securities. It does not disclaim implicit backing, nor does it signal that market participants err in perceiving such backing. It thus avoids the real issue.

Second, a statutory section entitled “Protection of taxpayers against liability” declares that the 1992 Act “may not be construed as obligating the Federal Government, either directly or indirectly, to provide any funds” to Fannie or Freddie "or to honor, reimburse, or otherwise guarantee any obligation or liability” of Fannie or Freddie. $ 4503. This disclaimer also avoids the real issue. No one argues that the 1992 Act created implicit backing where it did not already exist. Market participants had long believed such backing to exist under the GSEs' charters. Congress did not act to correct that perception. 13

Third, each firm's securities must include “appropriate language ... clearly indicating" that the securities are not guaranteed by the United States and do not constitute a debt or obligation of the United States or of any agency or instrumentality thereof” other than the GSE in question. $$ 1455(h)(1), 1719(b), (d)-e). This require ment repeats the fundamental weakness of the first disclaimer: It disclaims formal, legally enforceable liability, even as it fails to disclaim implicit backing. “Indeed, the disclaimer itself hints at a special Federal relationship; completely private firms do not need to disclaim Federal backing because no one believes such backing exists."

No one argues that the Government has any formal, legally enforceable liability for the GSEs' securities. Thus the disclaimers ignore the real issue: Whether the Government, although not legally bound to rescue the GSE's, would nonetheless do so (for example, because it felt a moral obligation for their debts or feared that a GSE default might damage the Nation's financial system). SUBSIDY DENIAL

The GSEs' double game helps the GSE's argue that they get little or no Government subsidy. Yet no one can honestly dispute that Fannie and Freddie receive valuable benefits not available to businesses generally. These benefits include exemption from most State and local taxes and exemption from the registration and reporting requirements of the securities laws. The benefits also include a conditional line of credit at the U.S. Treasury and special rules relating to the GSEs' securities—for example, rules that: Equate those securities with U.S. Treasury securities for some purposes; permit issuance and transfer of those securities over the system used for issuing and transferring U.S. Treasury securities; and fail to limit FDICinsured banks' investments in those securities. This special treatment strongly abets the market perception of implicit Federal backing. The Congressional Budget Office's reports and testimony demonstrate the great value of these special benefits.

Yet Fannie, in particular, insists that it receives no subsidy. Relying on a narrow dictionary definition to the effect that a “subsidy” is “monetary assistance granted by a Government to a person or private commercial enterprise,” Fannie asserts: “Fannie Mae does not receive a penny of public funds. To the contrary, last year our Federal tax liability was $1.6 billion. True subsidies also are tangible. Fannie Mae's Government benefits are not.” 15 Fannie's reasoning—that a subsidy involves only a tangible payment of money by the Government-produces absurd results. If Congress were to exempt Fannie from ever again having to pay any corporate income tax, that would supposedly not be a subsidy because it involves no cash payment to Fannie. Similarly, if a foreign government gave an energy-intensive, capital-intensive export industry unlimited access to free electricity and no-interest loans, that would supposedly not be a subsidy, either. These examples highlight the unreality of Fannie's arguments. CURBING THE DOUBLE GAME

13 The second disclaimer also replicates the weakness of the first disclaimer in declaring that the 1992 Act “may not be construed as implying that any such enterprise „, or any obligations or securities of such an enterprise ... are backed by the full faith and credit of the United States." $ 4503.

14 Ronald C. Moe and Thomas H. Stanton, Government Sponsored Enterprises as Federal Instrumentalities: Reconciling Private Management with Public Accountability, 49 PUB. ADMIN. REV. 321, 323 (1989).

15 Timothy Howard, Fannie Mae's Benefits to Home Buyers: The Business Perspective, 37 FED. RESERVE BANK CHI. ANN. CONF. BANK STRUCTURE & COMPETITION 68, 69 (2001).

I suggest that any GSE legislation:

(1) correct the faulty statutory disclaimers of Federal liability for Fannie and Freddie (discussed above);

(2) correct sloppy language in the Secondary Mortgage Market Enhancement Act of 1984 stating that for some purposes Fannie and Freddie securities “shall be considered to be obligations issued by the United States," 15 U.S.C. $77r-1(a)(1)-(2);

(3) prohibit any GSE from representing that the U.S. Government directly or indirectly backs the GSE (except in discussing formal, legally enforceable obligations of the Government) with the intent to induce anyone to rely on that representation in connection with the purchase or sale of any security; and

(4) prohibit any Government agency or official from characterizing GSE securities as Government securities. Properly Comparing Banks and GSE's

Fannie and Freddie often argue that the Federal Government gives FDIC-insured banks 16 benefits comparable to, or even greater than, those it gives Fannie and Freddie; that concern about subsidies to Fannie and Freddie is accordingly unwarranted and even hypocritical; and that any greater financial success shown by Fannie and Freddie simply reflects their greater efficiency.

Let us start with the issue of efficiency. Fannie and Freddie have lower overhead than banks because they do a different business than banks. Most banks do a predominantly retail business. To deal directly with large numbers of small customers, they have more offices and larger staffs than they otherwise would. By contrast, Fannie and Freddie do a wholesale business, which enables them to have lower overhead.

Now let us turn to the issue of relative subsidy. FDIC insurance has a different set of costs and benefits than the Government's sponsorship of Fannie and Freddie. You might expect FDIC insurance to provide a greater net subsidy:17 After all, FDIC insurance is established by law and carries the Government's full faith and credit. Yet the Government's perceived implicit backing of Fannie and Freddie actually tends to provide a greater net subsidy than FDIČ insurance, for six structural


1. Unlimited Coverage. Federal deposit insurance applies only to deposits and then only up to a $100,000 limit. The FDIC can protect a failed bank's uninsured deposits and nondeposit creditors (such as bondholders) only under very narrow circumstances. By contrast, the Government's perceived implicit backing of GSE's has no limits: It applies to all of a GSE's obligations, with no dollar ceiling.

2. No Receivership Mechanism. When an FDIC-insured bank fails, the FDIC becomes receiver for the bank: It takes control of the bank, gathers the bank's assets, and pays the bank's creditors in a specified order of priority. The bank's depositors must get paid in full before the bank's other creditors can get paid at all. If the bank's liabilities exceed its assets, its shareholders lose their ownership interest, its nondeposit creditors normally incur a partial or total loss, and its uninsured depositors often incur some loss. Similarly, when an ordinary nonfinancial company fails, it is liquidated under Chapter 7 of the Bankruptcy Code. The bankruptcy court appoints a trustee, who takes control of the company, gathers its assets, and pays creditors in a specified order of priority. The lack of any receivership mechanism for Fannie and Freddie reinforces the market perception that the Government would assure full payment of each firm's creditors.


16 For simplicity I use "banks” to refer to all FDIC-insured depository institutions, including thrift institutions.

17 The gross subsidy represents the total value of the special benefits provided by the Federal Government–benefits not available to businesses generally or even financial institutions generally. The net subsidy represents the difference between the gross subsidy and the offsetting costs that the entity must incur as a bank or GSE-costs not imposed on financial institutions generally.

18 I have set forth these arguments more fully in The Structure of Subsidy: Federal Deposit Insurance Versus Federal Sponsorship of Fannie Mae and Freddie Mac, in SERVING TWO MASTERS, YET OUT OF CONTROL: FANNIE MAE AND FREDDIE MAC 56-83 (2001).

Most of these structural reasons hold true for the Federal Home Loan Banks, which also enjoy exemption from the Federal income tax. But the Federal Home Loan Banks do face the possibility of receivership, and must pay 10 percent of their net income to an affordable housing program and another 20 percent toward interest payments on debt securities issued to help finance the thrift clean-up. 12 U.S.C. $$ 1430(j), 1433, 1441b(f)(2XC), 1446.

3. No Cross-Guarantees to Protect Taxpayers. Federal deposit insurance involves strong safeguards designed to ensure that banks-rather than the taxpayers—bear any losses incurred in protecting insured depositors. Banks must normally pay premiums large enough to ensure that the FDIC's insurance funds have at least $1.25 in reserves for each $100 of insured deposits. This obligation to pay premiums gives each insurance fund a claim on the capital and earnings of all banks insured by that fund—and in effect creates a network of indirect cross-guarantees among FDIC-insured banks. Thus each member of the Bank Insurance Fund is liable for ensuring that the FDIC can protect insured depositors at every other BIF member bank. As long as the fund can replenish its reserves, its existence precludes any loss to the taxpayers.

No similar cross-guarantees reduce the Government's risk-exposure to Fannie and Freddie. The two GSE's pay no insurance premiums and have no insurance fund. The two GSE's do not even cross-guarantee each other. If one GSE failed, the survivor would have no responsibility to pay the failed GSE's creditors.

4. Special Deals Instead of General Rules. To a much larger degree than banks, Fannie and Freddie reap the benefits of special, company-specific laws and avoid the discipline of generic law. Instead of operating under laws applicable to thousands of businesses, the two GSE's often get to operate under statutes designed for them alone.

5. Protection from Effective Competition Subsidizes GSE Shareholders. Federal and State regulators routinely issue bank charters to qualified applicants. Once chartered, a bank can typically engage in a wide range of activities statewide and even nationwide. No longer does each bank charter require special legislation. No longer do regulators grant charters sparingly so as to limit competition with existing banks. Entry into banking is relatively easy, and banking law affords banks little protection against competition. Thus, if banks receive a net Federal subsidy, they should generally face enough competition to force them to pass the subsidy through to their customers.

Fannie and Freddie, by contrast, enjoy significant protection against competition. Their Government sponsorship reduces their borrowing costs and increases the value of their guarantees to such an extent that no fully private firm can compete against them effectively. And only Congress can charter a competing GSE. By impeding competition with Fannie and Freddie, these constraints on entry increase the potential for the two GSEs' Government benefits to end up in the hands of their shareholders rather than their customers.

6. Free Ride. Banks must normally pay for deposit insurance. They must also comply with an array of restrictions and requirements not applicable to businesses generally. But Fannie and Freddie pay no fee for their Government sponsorship. They make no payments to an insurance fund or affordable housing fund. They need not provide public benefits that impose significant costs on their shareholders. HUD's affordable housing goals are so weak that Fannie and Freddie can meet them without doing more for affordable housing than banks do. I believe that the two GSE's would have a profit motive to do their affordable pusing business in any event, even without a Government subsidy.

Considering the great value of the benefits Fannie and Freddie receive from the Government, they should be doing far more to increase homeownership at the margin (for example, by the lower-middle class, the working poor, and members of certain historically disadvantaged minority groups). Systemic Risk

GSE's are often characterized as “too big to fail”—meaning that if they neared default on their debts, the Government would have to rescue them lest their failure unleash “systemic risk” that would gravely damage the Nation's financial system and economy

Discussions of systemic risk (whether in the GSE or the bank context) often have a tone of inevitability. But systemic risk is not a force of nature like earthquakes, hurricanes, and tornados. It results from human decisions: For example, decisions by market participants and Government officials about how to structure the financial system, what risks to take, and how to respond to problems. If investors expect the Government to protect them from the full pain of downside scenarios, they will tend to take greater risks than they otherwise would have taken. Thus “too big to fail” and “systemic risk” are to a large extent circular: They have their roots in prevailing expectations, and they easily become self-fulfilling prophecies. Insofar as investors expect the Government to rescue troubled GSE's, market discipline on GSE's will weaken, which will tend to increase the risk that the GSE's ultimately will get into financial trouble.

If a GSE's troubles coincide with a broader financial crisis, Government officials will face additional pressures to rescue the GSE. For if during the crisis those officials seriously upset established expectations, they may create contagious uncertainty about the Government's willingness to meet other expectations. A crisis is thus a particularly inopportune time for attempting to reeducate market participants about the scope of the Government's undertakings. So if the Government tacitly accepts "too big to fail” expectations during good times, it may find itself constrained during a crisis to rescue a GSE against its better judgment.

But the circularity of systemic risk also has a positive side: If the Government acts in a timely way, it can correct “too big to fail” expectations. Congress did just that in the FDIC Improvement Act of 1991 (FDICIA) by curtailing the practice of treating FDIC-insured banks as “too big to fail.” 19 FDICIA's “least-cost resolution" rule allows the FDIC to protect a failed bank's uninsured depositors and nondeposit creditors only if doing so is the “least costly to the deposit insurance fund of all possible methods” for meeting the FDIC's obligation to insured depositors. 12 U.S.C. § 1823(c)(4). The rule has a narrow systemic-risk exception, which has never been used.20 Before FDICIA, the FDIC was spending extra money from the deposit insurance fund to protect uninsured depositors at banks as small as $500 million in total assets. But less than 1 year later, when an $8.8 billion bank group in a swing State failed just a few days before the 1992 Presidential election, the FDIC did not protect uninsured depositors and financial markets took that action in stride. By giving clear and timely notice of the new policy, Congress had succeeded in changing market participants' expectations. Proper and timely Government action can thus reduce the potential for systemic risk.21

Effective safety and soundness regulation of the GSE's can further reduce that risk. Yet we should beware of relying excessively on regulation. Regulation did not prevent the U.S. banking system from collapsing in 1933. Regulation did not prevent the 1980's thrift debacle, with its $125 billion cost to the taxpayers. Regulation did not prevent the bank failures of the 1980's and early 1990's, which depleted the FDIC's Bank Insurance Fund. Nor, more recently, did OFHEO detect Fannie and Freddie's accounting problems, even though it had examiners scrutinizing each GSE year-round. The failings of regulation underscore the need to maintain market discipline on the GSE's and other large financial institutions. Miscellaneous ENDING THE GOVERNMENT'S ROLE IN APPOINTING GSE DIRECTORS

Under current law, Federal officials appoint some members of each housing GSE's board of directors. For both Fannie and Freddie, the President appoints 5 of each GSE's 18 directors. 12 U.S.C. $$ 1452(a)(2)(A), 1723(b). The Federal Housing Finance Board appoints 6 of each Federal Home Loan Bank's 14 directors. § 1427(a).

The Administration rightly proposes to end governmental appointment of GSE directors (and, as an initial step in that direction, has indicated that it will no longer appoint directors of Freddie). Government-appointed directors face a conflict of interest. They presumably have some duty to serve the public. Yet under corporate law they presumably also have fiduciary duties to their corporation's shareholders. These duties conflict whenever the shareholders' interests run counter to some broader public interest: For example, when the shareholders' interest in maximizing profits conflicts with the public interest in protecting the taxpayers or promoting affordable housing. In 1988, Freddie's directors concluded that their fiduciary duties compelled them to side with the shareholders against the taxpayers.22 In any event, government appointments to GSEs' boards of directors have served more as political plums than as vehicles for upholding the public interest.

19 In context of a failed FDIC-insured bank, “too big to fail” treatment involves spending extra money from the deposit insurance fund to protect deposits above the $100,000 limit on deposit insurance coverage. It may also involve extra spending to protect nondeposit creditors.

20 The systemic-risk exception becomes an option only if recommended to the Secretary of the Treasury by two-thirds majorities of both the Federal Reserve Board and the FDIC's Board of Directors. The Secretary can make the exception only if the Secretary determines, “in consultation with the President,” that least-cost resolution of a given institution "would have serious adverse effects on economic conditions or financial stability.” The Secretary must document the determination. The General Accounting Office must review and report on the exception, including the potential for it to diminish market discipline and encourage unsound risk-taking. To recoup the additional cost of deviating from least-cost resolution, the FDIC must levy a special assessment on insured depository institutions. $ 1823(c)(4)G). Congress designed these rules to promote accountability and make the process sufficiently unpleasant that systemic-risk exceptions would be made rarely (if at all) and never lightly.

21 By engineering a bailout of Long Term Capital Management in 1998, the Federal Reserve Bank of New York squandered some of FDICIA's hard-won gains. Yet the dramatic change in how the FDIC dealt with failed banks during the early 1990's shows that progress can be made in curtailing too big to fail treatment.

Ending such appointments-so that GSE shareholders would elect all GSE directors—would remove the conflict of interest. By strengthening shareholders' control over each GSE, ending such appointments could also help shareholders hold management more accountable and thus promote better corporate governance. COMPLYING WITH SECURITIES LAWS

The GSEs' statutory exemption from the registration and reporting requirements of the Federal securities laws is an anachronism and deserves to be repealed. The exemption sends the wrong signal: That GSE's are so “special,” so close to the Government, that investors in their securities have no need for the protections afforded by those requirements. OPPORTUNITIES FOR IMMEDIATE ADMINISTRATIVE ACTION

Regulators can and should act now to improve the regulation of Fannie and Freddie.

First, to help avoid unhealthy concentrations of credit risk among FDIC-insured banks, the Federal banking agencies should prescribe guidelines for banks' credit exposure to individual GSE's and to GSE's generally.2

Second, the Securities and Exchange Commission should prohibit mutual funds whose portfolios consist largely of GŠE securities from mislabeling themselves as "Government” or “U.S. Government" funds.24

Third, the Federal Reserve Board should proceed with its proposal to curtail socalled “daylight overdrafts” by GSE's.

Fourth, HUD should tighten its scrutiny of the GSEs’ mission, using its authority to review activity-expansion, prescribe affordable-housing goals, and interpret relevant statutes. Conclusion

The GSE's argue as though the present is always the wrong time to enact any reform that they do not want, such as reform that benefits taxpayers or homebuyers rather than the GSEs' managers and shareholders. In the GSÉs' view, either (1) there is no acute problem warranting reform, or (2) reform now would crimp housing markets and risk destabilizing the financial system. But now is the right time to act—to correct the deficiencies in GSE regulation before a crisis hits or another scandal breaks.


FEBRUARY 10, 2004

The 215,000 members of the National Association of Home Builders (NAHB) appreciate the opportunity to present their views to the Senate Committee on Banking, Housing, and Urban Affairs on the regulatory framework for Fannie Mae, Freddie Mac, and the Federal Home Loan Bank (FHLBank) System, including safe

22 The three members of the old Federal Home Loan Bank Board-appointed by the President and confirmed by the Senate-served ex officio as Freddie's board of directors. Freddie's preferred share price had more than doubled in response to a proposal to allow anyone to own Freddie shares. (By severing Freddie's historic link to the thrift industry, the proposal would free Freddie to increase its profits by amassing a large portfolio of mortgage-backed securities.) Senate Banking Committee Chairman William

Proxmire developed a plan to grant the relief desired upon payment of a fee to reduce the taxpayers' bill for the thrift clean-up. But Freddie's management convinced Freddie's directors that their fiduciary duties compelled them to oppose the Proxmire plan.

23 Regulators could, for example, prescribe rules or guidelines under Section 305(b)(1)(A)(ii) of FDICIA, which requires risk-based capital standards to “take adequate account of . centration of credit risk.” Regulators could also issue more specific examination standards.

24 The SEC prohibits a mutual fund from having “name suggesting that the Fund . [is] guaranteed by the United States Government.” 17 CFR § 270.35d-1. But many mutual funds that invest predominantly in GSE securities nonetheless call themselves “U.S. Government Securities Funds.” To take what is probably an extreme example, the Pacific Capital U.S. Government Securities Cash Assets Trust had 98.8 percent of its assets in GSE securities as of September 30, 2003; it evidently held no U.S. Government securities at all. See http:// (semi-annual report), at 9.


« iepriekšējāTurpināt »