The United States savings and loan and banking mess of the 1980s and early 1990s should have served as an important laboratory from which both U.K. and U.S. officials could have learned regulatory lessons in structuring a regime and
methods both for reducing failures and their consequences and for resolving fi- nancial institution failures in crisis and noncrisis periods. In 1991, Congress enacted the FDIC Improvement Act (FDICIA). This law introduced provisions for regula- tory actions to attempt to turn troubled banks around before insolvency and, if this failed, to legally close and recapitalize, sell, or liquidate a bank before its capital was fully depleted. The underlying theory was that, if successful, losses could be confined to shareholders; depositors and other creditors would remain whole.
The structure of FDICIA reflected the experience that as they approach insolvency troubled banks tend to overstate their revenues, income, and capital; understate their expenditures and loss reserves; and to blame their problems on lack of market liquidity. Thus, FDICIA assumed that both low levels of capital and so-called liq- uidity problems frequently masked underlying insolvency problems that required prompt regulatory actions, including, ultimately, resolution.
The United States also realized that insolvent banks needed to be legally closed more quickly than other firms to reduce both depositor runs and losses and to maintain uninterrupted operation of the payments systems, and that the closure decision should be made by regulators and not by the courts. Thus, Congress enacted a special bank bankruptcy code separate and different from the general corporate bankruptcy code (Bliss and Kaufman 2007). Legal closure powers to cancel a bank’s charter and place it in receivership were given to the bank regula- tors, not the courts. With rare exceptions, losses would be assigned to the insured claimants and the FDIC, standing in the shoes of insured depositors in absolute priority order. In addition, in the late 1980s, the FDIC also received authority to charter a temporary bridge bank to assume the activities of an insolvent large bank if a buyer could not be found sufficiently quickly. The United Kingdom had no such provisions in place in the first half of 2007. Nor were similar options available in the United States to deal with Bear Stearns, Lehman Brothers, or AIG.
Were these institutions solvent at the time the regulators acted to resolve them, or were they victims of short-term liquidity crises and only suffering temporarily depressed asset values? The answer depends on the capital measures and account- ing procedures applied. Clues can be gained from several pieces of evidence which ex post at least suggest the answers. For example, the inability of U.K. regulators to find a buyer for Northern Rock, which did not demand significant government financial compensation, implies that, at least in market value terms, NR was not solvent.
Similarly, IndyMac reported leverage and risk-based capital ratios well above the minimum requirements. Yet, like FSA’s problems with selling NR, the OTS had a difficult time finding a buyer for IndyMac when it tried shopping the insti- tution before resolution. This supports the argument made earlier that as a bank approaches insolvency, it tends to overstate its revenues and understate its ex- penses, thereby overstating both its income and capital. The delay caused by the search enlarged the losses substantially. On the other hand, the successful sale of Countrywide to the Bank of America without financial assistance suggests that it was perceived to be market value solvent or close enough to it to make it attractive to at least one potential buyer.6Bear Stearns was a more difficult case, but given the losses the Federal Reserve was accruing on the assets it guaranteed in the rescue, it is likely that Bear was insolvent as well.
The problems of reliance upon book value accounting and regulatory account- ing standards plague effective U.S. implementation of the PCA and FDICIA closure
352 The Problem of Regulation
rules. Some proponents of the FDICIA PCA procedures argued that the capital tranche triggers for regulatory intervention and closure rules should be based upon fair market values rather than book values, which lag changes in market conditions.7 The fact that the OTS could claim that IndyMac was not only ad- equately capitalized, but well capitalized just a few days before its closure and yet the FDIC could estimate at closure that it would lose between $4 billion to
$8 billion isres ipsa loquitur,and points to the critical need for further evaluation of how FDICIA should be implemented.
Given the lines of U.K. depositors in the street and no feasible alternatives to promptly close or otherwise resolve Northern Rock’s problems, the government was forced to guarantee all deposits and attempt to broker a merger or acquisition of the troubled institution. However, uncertainty about the value of the bank’s assets reduced its market value and made brokering a deal acceptable to most current shareholders, who, unlike in the United States, have a strong voice in the resolution process, doomed to failure. These actions introduce potential serious moral hazard problems by encouraging institutions with government guarantees to take on more risk and time inconsistently problems in that prevention of failure in the short run only promises to increase the magnitude of problems in the future with higher costs of failure. Losses will surely also be visited upon taxpayers.
CONCLUSION
The combination of poorly designed deposit insurance, poor regulatory supervi- sion, and a poor insolvency resolution regime led to a very visible and disruptive run on Northern Rock in the United Kingdom and larger-than-necessary losses in the failure of IndyMac, Bear Stearns, Countrywide, and AIG in the United States.
The run on Northern Rock has resulted in broad changes in the existing U.K.
deposit insurance contract, the supervisory structure, and the bank bankruptcy code. The recent failure and large losses at U.S. institutions have accelerated a reconsideration of its supervisory and regulatory systems. It has also resulted in a white paper on financial regulatory reform issued by the Treasury Department that contains a long list of suggested reforms.
This paper has described the more important faults in the bank regulatory structure in the United States and the United Kingdom that suggest areas in which reform would be most beneficial in reducing the number of bank failures and costs of those failures to taxpayers. Inefficient or unlucky banks should be permitted to fail, as in any other industry, but at low cost to the economy. But regardless of how well any regulatory legislation is drafted, without the support of the regulators in implementing the required actions effectively, the promised favorable outcomes are unlikely to by fully achieved. It should, however, be noted that these recom- mendations work best in noncrisis periods when large number of banks are not simultaneously endangered. In crisis periods, game books are generally thrown out in favor of quick, intuitively appealing ad hoc actions.
NOTES
1. HM Treasury, Bank of England, and Financial Services Authority (2007).
2. SeeFinancial Week(2007).
3. Data from FDIC Quarterly Reports of Condition and Income.
4. Indeed, the Treasury’s Inspector General’s Material Loss Review on IndyMac, which is required by FDICIA 1991, is highly critical of the OTS and its supervisory performance in the IndyMac case. The IG indicates that the OTS identified problems in IndyMac early on but never followed up to ensure appropriate corrective actions were taken, despite a history on the part of the institution of not addressing problems raised in examinations. In some instances, identified problems weren’t always reported in the Reports of Examination of IndyMac. See Office of Inspector General (2009).
5. See FSA Internal Audit Division (2008).
6. However, recent mortgage losses imply that BofA may have overestimated the true value of Countrywide’s assets.
7. Shadow Financial Regulatory Committee (2000).
REFERENCES
Bank of England. 2007. Financial Stability Report. October.
Bliss, R. R., and G. G. Kaufman. 2007. U.S. corporate and bank insolvency regimes: a com- parison and evaluation.Virginia Law and Business Review2 (1): 143–177.
Financial Week. 2007. Bank-to-thrift shift helps countrywide sneak by. Financial Week, July 30.
FSA Internal Audit Division. 2008. “ The Supervision of Northern Rock: A Lessons Learned Review,” March.
Gorton, G. 2008. The Panic of 2007. inMaintaining Stability in a changing Financial System.
A Symposium sponsored by the Federal Reserve Bank of Kansas City. Federal Reserve Bank of Kansas City.
HM Treasury, Bank of England, and Financial Services Authority. 2007.Liquidity support facility for Northern Rock PLC, 14.
Office of Inspector General, U.S. Department of the Treasury. 2009.Safety and soundness:
material loss review of IndyMac bank. Federal Savings Bank, OIG-09-032, February 26.
U.S. Shadow Financial Regulatory Committee. 2000. Reforming bank capital regulation.
Washington, D.C.: American Enterprise Institute, Policy Statement 160.
ABOUT THE AUTHORS
George G. Kaufmanis the John Smith professor of finance and economics at Loyola University Chicago. He previously taught at the University of Oregon and was a visiting professor at Stanford University and University of California at Berkeley.
Kaufman has also been associated with the Federal Reserve Bank of Chicago as both a full-time employee in the research department and as a consultant. Kaufman has been president of the Western Finance Association, Midwest Finance Association, and North American Economic and Finance Association and a director of the American Finance Association and the Western Economic Association. Kaufman is co-chairman of the Shadow Financial Regulatory Committee and was executive director of the Financial Economists Roundtable. Kaufman has published widely in professional journals and books and his money and banking textbook went through six editions. He is a co-editor of theJournal of Financial Stabilityand is on the editorial boards of a number of professional journals.
Robert A. Eisenbeishas Ph.D. and master’s degrees in economics from the Uni- versity of Wisconsin-Madison and a B.S. from Brown University. He is presently
354 The Problem of Regulation
chief monetary economist at Cumberland Advisors. He has held positions at the Federal Reserve Board, the FDIC, and most recently at the Federal Reserve Bank of Atlanta, where he was executive vice president and director of research. He has held academic positions as the Wachovia professor of banking and associate dean for research in the Kenan-Flager Business School, UNC-Chapel Hill. His pub- lished research includes work on banking structure and regulation in theJournal of Finance;theJournal of Financial Services Research;theJournal of Money, Credit, and Banking; and theJournal of Regulatory Economics. Current research examines the financial crisis, regulatory reform, and the development of methods for evaluating the forecasting performance of economic forecasters. He was named a fellow by the National Association for Business Economics in 2004.
CHAPTER 44
Why Securities Regulation Failed to Prevent the CDO Meltdown
RICHARD E. MENDALES
Visiting Professor of Law, Penn State Dickinson School of Law, Supreme Court Fellow, 1999–2000
The spark that set off the 2007–2008 explosion in the world financial markets was the failure of the market for collateralized debt obligations (CDOs),1 starting with one sector of that market, the one based on subprime mort- gages. Proposals to deal with the financial crisis often focus on measures drawn from bank regulation, such as capital requirements and leverage ratios. In fact, however, the crisis began when highly rated CDOs that formed significant parts of the capital of financial institutions turned out to be toxic: there was not enough information to fairly value them, although it was clear they were worth signifi- cantly less than their face value. This failure in transparency is a classic securities law problem, but SEC regulations failed to prevent it. Nonetheless, the underlying problems that led to the meltdown can best be addressed by giving the SEC the right tools for the job.
Mortgage lenders have long relied on a secondary mortgage market, enabling them to sell mortgages to investors and relend the proceeds to new mortgagors.
At least as early as the 1880s, investors bought mortgages from lenders, assembled them into pools, and sold participations in the pools as securities. The market for mortgage-backed securities collapsed with the Panic of 1893, but revived during the 1920s, and widespread defaults on these securities following the Crash of 1929 were noted in the Pecora Committee hearings that led to the Securities Act of 1933.
The New Deal recognized the need for a secondary mortgage market to pro- mote lending to homeowners. The National Housing Act of 1934 authorized one or more national associations to buy and sell mortgages, and Congress created the Federal National Mortgage Association (Fannie Mae) in 1938. It was authorized to buy and sell mortgages insured by the Federal Housing Administration. After World War II, veterans’ need for housing led Congress to authorize Fannie Mae to buy and sell mortgages insured by the Veterans Administration as well.
In 1966, the U.S. Treasury, facing costs such as that of the Vietnam War, found itself bumping against the upper limit fixed by Congress for the national debt. One 355 Copyright © 2010 John Wiley & Sons, Inc.
356 The Problem of Regulation
of the ways in which it dealt with this was to reinvent mortgage-backed securities.
Under this program, Fannie Mae pooled mortgages it had purchased and placed them in trusts, which sold participation certificates, giving buyers fractional inter- ests in each pool. Payments of principal and interest on the underlying mortgages flowed through to participation holders. The obligations represented by the par- ticipations were considered those of the mortgagors and not counted toward the federal debt, so the Treasury had not only reinvented mortgage-backed securities, but introduced off-balance-sheet financing.
Two years later, Congress removed more exposure from the federal balance sheet by splitting Fannie Mae in two. Half retained the Fannie Mae name and charter, but became a quasi-independent corporation, and continued to buy qual- ifying mortgages to collateralize securities. The other half became the Govern- ment National Mortgage Association (Ginnie Mae), which remained a government agency, with the function of guarantying securities backed by pools of FHA and VA mortgages.
Congress chartered another corporation, the Federal Home Loan Mortgage Corporation (Freddie Mac), in 1970, to package conventional mortgages (those not backed by the FHA or VA, and thus ineligible for Fannie Mae purchase) into mortgage-backed securities for sale on the securities markets. Although Fannie Mae and Freddie Mac gradually moved to private ownership, their government charters, federal regulation, and lines of credit from the government led to their designation as “government sponsored entities” (GSEs), and their securities were initially exempt from registration with the SEC.
Over the next decade, the GSEs were unable supply the demands of the mort- gage market. This was partly due to retrenchment in federal spending on housing, and partly to the failure of some mortgages to qualify for purchase by the GSEs, such as jumbo mortgages, whose size exceeded FHA guidelines. As a result, be- ginning in 1977, investment banks began to pool mortgages (and other sources of periodic payments such as car loans), and transfer them to entities designated as
“special purpose vehicles.” These issued “private label” securities, which passed payments from underlying mortgage pools through to security holders. Unlike GSE securities, these were registered with the SEC. The process is calledsecuritiza- tion,or more generally,structured finance.
The Secondary Mortgage Market Enhancement Act of 1984 (SMMEA) aided securitization by removing securities law impediments to private label securities;
for example, it removed them from regulation under state securities laws. With the act’s definition of qualifying securities, however, a snake slithered under the rock:
to be a mortgage-related security for purposes of SMMEA, a security had to be rated at one of the two highest levels provided by at least one major credit rating agency, which the SEC called Nationally Recognized Statistical Rating Organiza- tions (NRSROs).
Rating agencies had rated the creditworthiness of corporations and their secu- rities for close to a century, and were originally paid for their ratings by subscribing investors. In the 1970s, however, they changed their business model to payment by the issuers whose securities they were rating. This obvious conflict of interest was aggravated by an absence of competition: there were only three agencies accredited by the SEC, with Moody’s and Standard & Poor’s getting most of the work, and Fitch bringing up the rear. Ironically, the SEC was the first government body to give
these unregulated ratings official cachet, by using them as criteria for securities that broker-dealers could use to satisfy their capital requirements, beginning in 1975.
During the next 30 years, the SEC did nothing substantial to regulate the rating process, and blocked potentially corrective competition by failing to accredit new agencies.
In the issuance of asset-backed securities, the rating system became a substitute for careful examination of the mortgages in pools being securitized. Rating agencies did not examine individual mortgages in a pool, but instead rated securities based on applying proprietary statistical models to collective characteristics reported on the loan tape, an unverified summary of the assets in a pool. The agencies were largely unregulated until Congress passed the Credit Rating Agency Reform Act of 2006 (CRARA) in the wake of the agencies’ failure to issue timely downgrades of failing corporations such as Enron. CRARA provided only minimal regulation.
It was chiefly intended to force the SEC to accredit more agencies, although in fact it failed to produce a meaningful increase in competition. Though it barred some abusive practices such as threatening to give an adverse rating unless an issuer hired an agency, it was counterproductive in more important ways. It barred the SEC from substantively regulating the rating process, giving the agencies a privi- leged position not shared by other participants in the process of issuing securities, which are generally required by the Securities Act of 1933 to employ due dili- gence to ensure that information they provide concerning newly issued securities is accurate.
By the time CRARA was enacted, flaws based on defective ratings were em- bedded in billions of dollars’ worth of CDOs and their derivatives, particularly those issued after the turn of the century. Mortgage-backed securities received top ratings even though none of the mortgages that backed them had strong repayment ability.
The alchemy by which securities based on pools of low-grade mortgages re- ceived top ratings took several forms. First, there were credit enhancements, guar- antying payments of the securities, such as standby letters of credit issued by banks and insurance policies from monoline insurers such as Ambac Financial Group.
These guarantees were no stronger than the institutions that issued them. Mono- line insurers in particular had no alternate insurance business to support them if enough mortgage-backed securities defaulted. When the crisis broke, Ambac lost its own AAA rating.
Pools of low-grade collateral were also turned into AAA-rated securities by dividing the securities into tranches, through which holders of a first tranche held first priority on payments due from mortgagors, followed by holders of a second tranche, and so on. The rating agencies created models under which, because holders of higher-ranking tranches held first priority rights to payments from all successive tranches—what was called thewaterfall effect—they would supposedly be assured of payment even if some mortgages in a pool defaulted, and first- priority tranches received top ratings, even though their cash flow came entirely from high-risk mortgages.
Matters became more complex when investment banks pooled hundreds of tranches of asset-backed securities into pools that provided cash flow for other securities, known as collateralized debt obligations, or CDOs. CDOs include not only interests in pools of mortgages and securities based on them, but also