No one should be surprised that the Office of Federal Housing Enterprise Oversight (OFHEO) found substantial risk-taking as well as accounting fraud at Fannie Mae. The combination of government backing and the opportunity for private profit invites both. Fannie’s supporters argue that stronger regulation is enough to protect the taxpayers and the economy, but this ignores the unique characteristics of government sponsored enterprises (GSEs) as well as the inability of OFHEO itself to detect in advance the poor controls and false accounting at both Fannie Mae and Freddie Mac. Accordingly, the only sure way to eliminate the risk is to reduce the kinds of risky activities in which government-backed management can engage.
Many in Washington expressed surprise at the revelations in the May 2006 special OFHEO report on Fannie Mae. The report disclosed that the company had not only manipulated its financial statements, but also had weak internal controls, inadequate accounting and computer systems, and had taken substantial business risks in order to improve its profitability and pay unearned bonuses to its management. But surprise was unwarranted; it is axiomatic that profit-making companies, when offered a subsidy in the form of government backing, will exploit the opportunity to its limit. Indeed, it can be argued that a board of directors and management of a shareholder-owned company have a fiduciary obligation to do so. Nor should it have been a surprise that OFHEO has now found Fannie Mae to be an entirely different company from the one on which it reported so favorably during the very years covered by the report. Regulation is an inherently weak method of controlling risk-taking, especially when government backing--such as that enjoyed by the GSEs--encourages both risk-taking and weak controls.
The Absence of Market Discipline--and Its Consequences
One of the advantages of government support is that a government-backed management is permitted to take risks that are forbidden to others. This is true because government backing eliminates the restrictions--known as market discipline--imposed by wary creditors. Ordinary financial institutions such as banks borrow their raw material--money--and lend it to others. If creditors perceive that the risk of lending to a bank or other institution is rising, they will demand a higher rate of interest for any loan, increasing the institution’s borrowing costs. Eventually, if the risk is not controlled, the institution will not be able to borrow at all, and thus will be prevented from expanding its operations. In this way, market discipline supplements bank regulation by signaling to regulators that there is something amiss, and limiting excessive risk. When large U.S. banks were deemed financially weak during the debt crisis involving Mexico, Brazil, and Argentina in the 1980s, the rates they had to pay for money rose substantially, and in some cases they were unable to access the credit markets at all without offering substantial collateral. Investors perceived a risk that these banks would default. Although bank deposits up to $100,000 are nominally insured by the Federal Deposit Insurance Corporation (FDIC), deposits are only a small proportion of the obligations of large banks; their other debt, consisting of commercial paper and longer term notes, is not seen as backed by the federal government--so market discipline continues to work to some degree, even for large banks.
But Fannie and Freddie are different from banks in one important way: despite the fact that their securities explicitly state that they are not backed by the federal government, their government charter and mission--plus the government’s past behavior--have persuaded investors that neither company will be allowed to default. Thus, in a very practical sense, all their debt obligations--not just some limited amount corresponding to a bank’s deposits--are seen by U.S. and foreign investors as nearly risk-free, and therefore are not subject to market discipline. In effect, they are given a free pass to take risk. The name for this phenomenon--in which government backing reduces market discipline--is moral hazard, and the GSEs represent moral hazard on steroids.
If there was ever any doubt about this, it has been erased since 2003, when first Freddie and then Fannie were forced to restate their financial reports because of false accounting and gross manipulation of their financial results. In the succeeding years, neither company has been able to file financial reports with the Securities and Exchange Commission or submit them to their shareholders or lenders. For ordinary corporations and banks, this would long ago have precipitated a plunge into bankruptcy--but not Fannie and Freddie. They have continued to operate normally, despite the fact that no one knows their true financial condition, and this has revealed a remarkable phenomenon. Because of their perceived government backing, the GSEs pay interest at a rate slightly higher than that of the U.S. Treasury. The difference between what they pay and what the Treasury pays is known as the “spread,” and the width of the spread signals the degree of risk that the market sees in holding GSE debt. The accompanying chart shows that after OFHEO reported in October 2004 that Fannie had falsified its financial statements, the spread of Fannie’s debt over the Treasury’s actually narrowed. In other words, Fannie’s debt strengthened, rather than weakened, despite the financial scandals that then surrounded the company. And that was true not only of Fannie’s senior debt; it was also true of its subordinated debt, which was originally created to provide an equity-like signal of the company’s financial condition. But it now appears that Fannie’s subordinated debt is also considered government-backed, and is therefore not subject to significant risk of default. In other words, investors have not lost any confidence in eventually getting paid by Fannie and Freddie, even though they can have no idea whether the two companies can repay their debts.
This privileged position was not lost on the management of Fannie Mae. The OFHEO report recounts in detail the sorry state of Fannie’s risk controls, internal controls, staff competence, accounting systems, computer systems, compliance with accounting requirements, and corporate governance. A central element of the report is that Fannie and its management consistently misled investors, Congress, and the public about both its financial position and the risks it was taking. Thus, OFHEO noted in the special report: “Fannie was not ‘one of the lowest risk financial institutions in the world’ but was exposed to significant interest rate risk and quite large operational and reputational risks.” Indeed, Fannie’s weakness in basic corporate and accounting controls was so serious that the company has thus far been required to spend approximately $800 million on repairs of these systems, and both Fannie and Freddie (which has been at the repair process for three years) are said by the new director of OFHEO still to be years from adequacy.
The fact that both Fannie and Freddie have had similar internal control and accounting problems makes it reasonable to assume that there is a similar cause, and that this cause can be found in their unique position as GSEs--privately-owned companies that are not subject to market discipline. They were not required to take the steps that all other companies must take to maintain their good credit standing, and so they did not take these steps. As long as the GSEs exist, they will always exhibit this tendency. Why indeed should any management spend money on all of the systems and controls that market discipline demands if they do not need these elements to persuade the market to lend to them?
Why Regulation Cannot Cut It
This presents a policy problem that is not just different in degree, but different in kind from the usual difficulties of supervising financial institutions. Solutions that might work for other kinds of financial firms and even banks will not work for the GSEs. It is clear that the capital markets cannot be induced to believe that the government will allow either Fannie or Freddie to fail. If investors will not lose confidence in the GSEs when neither of them can keep or publish accurate accounts, it is obvious that the capital markets will never believe the Treasury’s traditional vow that Fannie and Freddie will not be bailed out if they encounter financial difficulty. Thus, without some kind of extraordinary action, the inducement to risk-taking by the managements of both companies will remain intact.
When the problems at the GSEs first came to light, however, the first instinct in Congress was not extraordinary at all: it was the reflexive effort to increase the degree and restrictiveness of regulation without questioning whether it is really a solution to the problem at hand. Thus, in the wake of the problems at Fannie and Freddie, the call in Congress was to create a new regulator with “world class” powers equivalent to those of a bank regulator.
The first problem with this idea was that, as Alan Greenspan pointed out, tighter regulation only increases the tendency of the markets to believe that Fannie and Freddie are wards of the government: “World-class regulation, by itself,” said Greenspan, “may not be sufficient and indeed . . . may even worsen the situation if market participants infer from such regulation that the government is all the more likely to back GSE debt.” In addition, as noted above, bank regulators are assisted, at least to some degree, by the existence of market discipline, and the new GSE regulator will not have this vital support. So if it were the intention of Congress to give the new regulator as much authority as a bank regulator wields, that would not have been enough. Given the complacent attitude of the capital markets, the new regulator would need more authority than a bank regulator in order to deal effectively with risk-taking by Fannie and Freddie. But the bill ultimately adopted by the House of Representatives in 2005 was deficient in this respect, as in many others. It did not even provide the new GSE regulator with all of the authority of bank regulators.
Ultimately, however, there is this question: even if the new regulator were given extraordinary powers, beyond those of a bank regulator, would they be enough to protect the taxpayers or the economy against the incentives and opportunities for risk-taking and weak controls that government backing makes available to the GSEs’ managements? The answer here appears to be no. Although there has been a great deal of talk about the weakness of OFHEO’s authority, in the major respects necessary to prevent the risk-taking, financial manipulation, and false accounting at Fannie and Freddie, OFHEO has always possessed the powers necessary to do the job. Under its authority today, without any enhancement, it has the same examination authority as a bank regulator, complete access to the books, records, staff and management of both GSEs, the power to request and review documents and internal memoranda, and subpoena power to back up its requests. Indeed, the special OFHEO report itself, citing an earlier agency statement, notes that “OFHEO possesses supervisory responsibilities and powers ‘essentially similar to those of Federal bank regulatory agencies.’”
As much as Congress might wish it were not so, the problem was not and is not authority, but knowing what questions to ask. True, OFHEO has only limited cease-and-desist authority, and does not have the ability to put either of the GSEs into receivership, but neither of these customary bank powers would have been invoked by OFHEO against either Fannie or Freddie, simply because the agency lacked the necessary knowledge about what Fannie and Freddie were actually doing.
Lack of knowledge--what is called actionable intelligence in warfare--is the Achilles’ heel of regulation, and even if OFHEO had had extraordinary powers to intervene in the decision-making processes of Fannie and Freddie, it would not have been able to exercise that authority without the knowledge that excessive risk-taking or financial fraud was going on. The billions of dollars lost by Fannie Mae because of its excessive risk-taking and use of false accounting to hide its losses occurred while OFHEO was performing all the usual functions of a regulator--reviewing books and records, reading minutes of meetings, analyzing financial results, and interviewing employees. In its special report, OFHEO declared with appropriate rectitude that the company misled investors, Congress, and the public generally by falsely claiming that it was one of the best-run and least risky companies in the world. But if this was indeed false, why was OFHEO--exercising the usual authorities of bank supervisors--unable to discover any of these things in prior examinations? Thus, in a director’s statement accompanying the agency’s regular Report to Congress in 2002, Director Armando Falcon noted that in its supervision of the GSEs during the preceding year, OFHEO had found nothing amiss: “I am also pleased to report that OFHEO has found both Fannie Mae and Freddie Mac to be well-capitalized and operating in a safe and sound manner.” The similar message accompanying the 2003 report, in light of later events, rises to parody: “The Enterprises have remained safe and sound through another year of exceptional growth in the housing sector of our economy. In a year when more and more Americans have become homeowners, the public can take comfort in knowing that OFHEO is on the job, doing its part to ensure the strength and vitality of the nation’s housing finance system.” The sad fact is that OFHEO was “on the job,” but its assignment was beyond the capacity of any financial regulator or supervisor.
OFHEO’s failure, unfortunately, is not something that can be cured by legislation; it is the result of the same deficiency shared by all regulators, and the reason why regulation should not be treated as a panacea. Risk-taking and fraud originate in the decisions of top management, in meetings to which no regulator and few ordinary employees are ever privy. Were it not for the unusual events, outlined below, which put the OFHEO watchdog onto the scent, it is virtually certain that both Fannie and Freddie would still be manipulating their accounts and misleading investors, the public, Congress, and OFHEO itself.
It all goes back to Enron and WorldCom, and the fact that the auditor of both companies, Arthur Andersen, was in very bad odor after their collapse in 2002. It happened that Andersen was also the auditor for Freddie Mac, and in what might be called an act of caution, Freddie’s board fired Andersen and engaged new auditors. In taking over the Freddie account, the new auditors discovered some troubling aspects of the company’s accounting, which, in turn, moved the board to hire an independent counsel to look into the matter. The independent counsel found not only accounting problems, but also an apparent management cover-up, which precipitated the dismissal of the company’s top three officers.
This event apparently came as a complete surprise to OFHEO, which had only days before released the Report to Congress cited above, in which it said that “the public can take comfort in knowing that OFHEO is on the job.” The agency clearly had not been aware of the seriousness of the accounting problems at Freddie or perhaps even the existence of the independent counsel’s investigation of the company. In its 2003 report, OFHEO dutifully reported only what it had apparently been told by Freddie’s management--that there was a technical question of accounting interpretation, and that Freddie’s management was “taking the appropriate action.” Humiliated by the public disclosure of its cluelessness--the Wall Street Journal reported that “no one seemed more surprised [by the dismissal of Freddie’s top officers] than the company’s regulator”--OFHEO became far more aggressive toward both GSEs, eventually requesting and receiving from Congress a supplemental appropriation of $7.5 million to do a “forensic audit” of Fannie. The audit found evidence of false accounting. And the rest, as they say, is history.
The clear lesson here is that regulation has inherent limitations--and its first and most important limitation is what regulators and supervisors can actually know about the financial institutions they oversee. For this reason, regulation is highly overrated as a means of controlling risk. Risk is created at the top of an organization by management policies, in meetings that the regulators do not and never will be able to attend. The regulators, instead, are down at the grass roots, seeing the numbers but without understanding their meaning or the policies behind them. Like all financial supervisors, OFHEO was unable to obtain the essential knowledge about management’s intent that was inevitably denied to it. Accordingly, placing confidence entirely in regulation as a protection for the taxpayers would be, in the end, an imprudent action for Congress, and even more so where the regulator’s efforts are not supplemented by market discipline.
Reducing Risk by Eliminating Risky Activities
So if more and better regulation alone is not able to prevent a loss to the taxpayers and systemic risk to the economy, what will? The only sensible course is to reduce the risks that Fannie and Freddie are able to take. Fortunately, this is easy to do. The principal risk of the GSEs is what is known as interest rate risk, and comes from the fact that they have borrowed over a trillion dollars in order to buy and hold a portfolio of mortgages of roughly the same size. If interest rates rise or fall sharply, they can suffer huge losses. The bill passed almost a year ago by the Senate Banking Committee recognizes this fact and attempts to reduce the GSEs’ retained portfolio to a minimum.
The key and most significant element of the Senate bill would do just this. The reason is simple: carrying a portfolio of mortgages creates substantial interest rate risk--the same risk that destroyed the savings and loans a generation ago. Because of their government backing, the GSEs have the opportunity to make hefty profits from borrowing at a low rate of interest--only slightly more than then the Treasury pays--and buying mortgages and MBS that pay considerably more. For this reason, over the last twenty-five years, the GSEs’ portfolios of mortgages and MBS have increased from $60 billion to almost $1.5 trillion. But the business of borrowing at low rates and buying higher yielding mortgages has its perils--principally interest rate risk.
A peculiar characteristic of interest rate risk is that it exists whether interest rates rise or fall. Two examples will show why this is true. Assume that a GSE borrows $1 million at 4 percent to buy a portfolio of mortgages that yields 5 percent. As long as it holds this portfolio, the GSE pockets the difference between what it has to pay to its lenders (4 percent) and what it is receiving from the mortgages it is holding (5 percent). (This is only a simplified example; in reality, the GSEs’ profits from holding mortgages and MBS are considerably higher. In a paper delivered at an AEI conference in 2004, Professor Dwight Jaffee of Berkeley estimated the average 2003 spread as 172 basis points for Fannie and 186 for Freddie.) However, if interest rates fall to, say, 3 percent, many homeowners will refinance their 5 percent mortgages. The GSE will, of course, get the cash from this refinancing, but the mortgages it will be able to buy with this cash will then only pay 3 percent. Its profitable situation is now reversed; it is paying 4 percent to hold mortgages that are yielding 3 percent, and losing money on every one. That is exactly what OFHEO found had happened at Fannie: “Fannie Mae consistently took a significant amount of interest rate risk and, when interest rates fell in recent years, incurred billions of dollars in economic loss.”
But the GSEs can also lose if interest rates rise. Using the same hypothetical set of facts, what happens if interest rates rise to 6 percent? In that case, homeowners will hold onto their 5 percent mortgages, since refinancing will not pay, and the GSE has to worry about what it will do when its loans mature. If the loans are short-term, and about half of all GSE borrowings are contracted for a year or less, the GSE will be compelled to refinance its initial borrowing at 6 percent, and will again suffer a loss as it pays 6 percent to carry a mortgage portfolio that continues to pay only 5 percent. Thus, if its borrowings are short-term, a GSE will suffer a financial loss if interest rates rise, just as it will suffer a loss if interest rates decline while its liabilities are long-term.
To be sure, interest rate risk can be hedged--the risk can be shared with or transferred to others through various kinds of derivatives such as interest rate swaps. But swaps and other hedges are expensive, and to achieve higher levels of profitability, a GSE will not completely hedge its risk. In its report, OFHEO complained that in trying to hit the profit numbers that would have assured their bonuses, Fannie’s management did not take the prudent steps necessary to hedge its interest rate risks. The report states: “During the period covered by this report, Fannie Mae’s strategy was to match between 50 and 60 percent of the [interest rate risk] of its mortgage assets.” Even this understates the degree of risk that Fannie was assuming. In his 2005 paper, Professor Jaffee notes that the GSEs hedge only the small, foreseeable interest rate risks, and that their hedging strategy “transfers the risk of unexpected, large, and future rate changes onto the U.S. Treasury based on the implicit guarantee . . . imposing on U.S. taxpayers the large and distant risks that would eventually require a U.S. Treasury bailout.” He concludes: “The firms are able to operate in this manner only because the purchasers of their agency debt and their MBS show little concern for the firms’ riskiness, protected as they are by the implicit Treasury guarantee.” Moral hazard, again, flowing from the absence of market discipline.
There is no reason why Fannie and Freddie should be permitted to buy and hold large portfolios of mortgages. No less an authority than Alan Greenspan has testified to Congress that the portfolios seem to have no purpose other than to increase the earnings of the two companies. As such, it is a classic example of privatizing profits while socializing risk and loss. Fannie and Freddie, in addition to their shareholders and managements, will profit from holding these portfolios, but if the risks they take cause them to suffer serious losses, the taxpayers will pick up the tab.
It is important to recognize that eliminating or severely restricting the size of their portfolios will not reduce the role of the GSEs in providing whatever assistance they are thought to give to the housing market. Indeed, most of what the GSEs do for the housing market today is done through what is known as securitization. In this process, Fannie and Freddie purchase mortgages from banks and other lenders and sell interests in these mortgages to investors. In this case, investors and not the GSEs take the interest rate risk. Because this process involves the purchase of mortgages by the GSEs, it adds the same amount of liquidity to the mortgage market as the GSEs’ purchasing and holding mortgages in their portfolios--but with substantially less risk to them, to the taxpayers, and to the economy.
So the Senate Banking Committee has it right. By providing in its reform legislation for limiting the size of the GSEs’ portfolios, it has adopted the one sure way to eliminate the threat that risk-taking by the GSEs poses to the taxpayers and the economy.
Peter J. Wallison is a resident fellow at AEI.
AEI staff assistant Daniel Geary and AEI editorial assistant Nicole Passan worked with Mr. Wallison to edit and produce this Financial Services Outlook.
1. Office of Federal Housing Enterprise Oversight (OHFEO), Report of the Special Examination of Fannie Mae (Washington, D.C., May 2006), available at www.ofheo.gov/media/pdf/FNMSPECIALEXAM.PDF.
2. Ibid., 47.
3. James B. Lockhart III, Senate Committee on Banking, Housing, and Urban Affairs, The OFHEO Report of the Special Examination of Fannie Mae, 109th Cong., 2nd sess., June 15, 2006.
4. Alan Greenspan, Senate Committee on Banking, Housing, and Urban Affairs, Proposals for Improving the Regulation of the Housing Government Sponsored Enterprises, 108th Cong., 2nd sess., February 24, 2004, available at www.federalreserve.gov/boarddocs/testimony/2004/20040224/default.htm.
5. Peter J. Wallison and Thomas H. Stanton, H.R. 1461: A GSE ‘Reform’ That Is Worse than Current Law, (Washington, D.C.: AEI, June 17, 2005), available at www.aei.org/publication22705/.
6. The Federal Housing Enterprise Financial Safety and Soundness Act of 1992, U.S. Code 12 (1992), chapter 46, available at http://uscode.house.gov/download/pls/12C46.txt.
7. OFHEO, Report of the Special Examination of Fannie Mae, 21.
8. Office of Federal Housing Enterprise Oversight, “Director’s Message,” Report to Congress, June 15, 2002, available at www.ofheo.gov/Media/Archive/reports/ar61402.pdf.
9. Office of Federal Housing Enterprise Oversight, “Director’s Message,” Report to Congress, June 15, 2003, available at www.ofheo.gov/media/pdf/WEBsiteOFHEOREPtoCongress03.pdf.
10. Federal Housing Enterprise Regulatory Reform Act of 2005, 109th Cong., 1st sess., S. 190, July 22, 2005, available at http://banking.senate.gov/_files/CommitteePrint.pdf.
11. Robert A. Eisenbeis, W. Scott Frame, and Larry D. Wall, An Analysis of the Systemic Risks Posed by Fannie Mae and Freddie Mac and an Evaluation of the Policy Options for Reducing Those Risks (working paper 2006-2, Federal Reserve Bank of Atlanta, April 2006), 55, available at www.frbatlanta.org/filelegacydocs/wp0602.pdf.
12. Dwight Jaffee, The Effects of Limiting the Retained Mortgage Portfolios of Fannie Mae and Freddie Mac (speech, AEI, Washington, D.C., April 26, 2005), 5, available at www.aei.org/event1041.
13. OFHEO, Report of the Special Examination of Fannie Mae, 4.
14. Ibid., 46.
15. Dwight Jaffee, 12.
17. “We have been unable to find any purpose for the huge balance sheets of the GSEs, other than profit creation through the exploitation of the market-granted subsidy.” Alan Greenspan, Senate Committee on Banking, Housing, and Urban Affairs, Reform of the Government-Sponsored Enterprises, 109th Cong., 1st sess., April 6, 2005, available at www.federalreserve.gov/BoardDocs/Testimony/2005/20050406/default.htm.