|Financial Services Outlook logo 130|
Although it has a valuable endorsement from an important Democratic senator, Michael Lewis's best-selling book The Big Short raises questions about the validity of the ideas underpinning the financial regulation legislation Congress is now considering.
Key points in this Outlook:
- The Big Short shows that Wall Street firms were as slow as most others in recognizing the dangers of subprime mortgages and mortgage-backed securities (MBS) based on these risky loans.
- Although the financial crisis narrative that drove the new bill posits that Wall Street firms knew the MBS they were selling were likely to default, the Lewis story shows that Wall Street resisted opportunities to bet against these securities, or even to hedge the MBS and collateralized debt obligations in their own portfolios, until the losses became obvious.
- Regulators have consistently failed to recognize or to prevent bubbles, but The Big Short shows that--for those who recognize that a bubble is developing--credit default swaps (CDS) can provide powerful incentives to speculate against bubbles.
- The regulatory legislation now before the House and Senate will make it difficult for participants in the CDS market to speculate against a bubble, making future bubbles more likely.
During the first week of June, the press reported that Senator Dick Durbin (D-Ill.) had interrupted a speech on the Senate floor to recommend a book to his colleagues. That book is Michael Lewis's The Big Short, which documents some of the events on Wall Street leading up to the financial collapse of 2008. It is a bit puzzling that Durbin chose to single out this book, which describes how several neophyte money managers used credit default swaps (CDSs) to make substantial profits by speculating against the value of collateralized debt obligations (CDOs) and mortgage-backed securities (MBS) based on subprime mortgages. Given that Durbin endorsed it, one might suppose that the book supports the narrative that underlies the financial regulatory legislation now before the House and Senate: that CDSs and Wall Street greed were major causes of the financial crisis. But the events that Lewis documents in The Big Short actually contradict the idea that Wall Street greed caused the financial crisis and cast doubt on whether restrictions on the use of CDSs--a major element of the Democratic plan for regulating the financial industry--would be good policy.
Credit Default Swaps. CDSs are not as mysterious as they seem and can most easily be understood as a form of credit insurance. The buyer of a CDS purchases something like insurance coverage (called protection) against a loss he will suffer if a particular company (known as the reference entity) defaults on an obligation. For this protection, the buyer pays an annual premium to the seller of protection who, in turn, promises to reimburse the buyer in a fixed amount (the "notional amount") if the reference entity defaults. This kind of contract is not new; for many years, banks have been writing contracts like this, called standby letters of credit, which often back municipal or other securities. For purposes of this discussion, the difference between CDSs and bank letters of credit is that CDSs are bought and sold in an over-the-counter market, in which the price of the coverage fluctuates with the market's assessment of the risk of default associated with the reference entity, while bank letters of credit are negotiated between the bank and the issuer, without reference to a prevailing market assessment about the issuer's credit. In a typical CDS transaction, the holder of, say, a $10 million IBM bond may want to be protected against IBM's default. To gain this protection, he buys a CDS from a dealer, usually a bank or other financial institution, in which he agrees to pay an annual premium for a predetermined period. The dealer, in turn, will usually attempt to hedge the obligation it has assumed by purchasing an offsetting obligation from another party, perhaps another dealer or a third party hoping to earn the premium by taking on this exposure. Banks and other financial institutions use CDSs as risk-management tools to reduce concentration in their portfolios and to take on risk that adds to their diversification. An important element of the CDS market is that participants can buy CDS coverage on the same IBM bond even if they do not actually own the bond. In effect, this transaction--called a "naked swap"--is roughly equivalent to a short sale in an equity market. They are speculating that IBM will default, or that its credit will weaken so that the short position becomes more valuable. Naked swaps are a major feature of the transactions described in The Big Short.
Collateralized Debt Obligations. CDOs are also not complicated to understand, but they can be difficult to value. In a CDO, pools of various kinds of debts--auto loans, boat loans, retail-credit loans of various kinds, and mortgages--are aggregated and used to provide a stream of cash payments to the holders of securities backed by the pool. The CDOs that are the subject of The Big Short were mostly composed of subprime mortgages. The securities that are backed by the pool of loans have different priorities--called tranches--for receiving the cash that is paid into the pool. The group with the highest priority usually gets all the cash before any of the lower tranches are paid anything. Because of this priority, credit-rating agencies generally rate the top tranches AAA. Lower tranches, which are in effect subordinated, are rated AA, A, BBB, and so on. Because they have prior rights to the cash received by the pool, the AAA and other higher tranches take less risk than the others and are paid a lower return. The lower tranches, correspondingly, take more risk and receive a higher return. The CDOs that are the subject of The Big Short largely consisted of subprime mortgages, but with an additional twist; many of them were composed of the lower tranches--the BBB tranches--of MBS assembled previously by their Wall Street sponsors. The fact that cash receipts first have to reach the BBB tranches in the prior MBS makes the valuation of the CDO very complex.
Most of The Big Short focuses on several newly minted money managers who thought early in 2005 that in the relatively near future large losses would occur in the CDOs based on subprime mortgages and that they could profit from these defaults by purchasing CDS protection. This would be similar to believing that IBM will suffer losses and selling IBM stock short. The opportunity for profit existed because most people on Wall Street at the time seemed to believe a number of things that turned out to be wrong: (1) that the rating agencies' models accurately reflected the risks of the MBS and CDOs, (2) that housing prices never fall on a national basis (and thus that there was little correlation between housing prices in different regions), and (3) that if the housing bubble began to deflate the losses would not exceed historic levels. Because of these assumptions, the premiums required to purchase CDSs on the AAA and other top tranches of CDOs were extremely low--only twenty basis points (0.2 percent, or one-fifth of a penny) for each dollar of coverage on the AAA tranches of a CDO. Thus, in an extreme example, where the cost of a CDS is twenty basis points, an annual premium of $100,000 will enable the buyer of the CDS to obtain protection against the failure of the AAA tranch of a CDO with a value (before any default) of $50 million--a payoff of five hundred to one if a default occurs.
Persuading Wall Street That AAA CDOs Would Fail Was a Tough Sell
We now know that all of the assumptions of most investors and Wall Street experts turned out to be wrong, but the interesting thing about the Lewis book is how difficult it was to get Wall Street traders and investors to abandon their initial beliefs about subprime loans and CDOs. The money managers Lewis follows throughout the book--the people who saw the possibilities of failure even in the topmost tranches of CDOs--tried to persuade others to make the same bets or at least to understand why the traders and others continued to dismiss the possibility that a major mortgage meltdown could happen. In quoting the money managers with whom he talks, Lewis notes their amazement that none of the people they spoke to could even present rational counterarguments to the proposition that CDO defaults were inevitable: "Nobody we talked to," said one of the money managers, "had any credible reason to think this [the failure of subprime CDOs] was going to become a big problem. . . . No one was really thinking about it." As Lewis notes, "The catastrophe was foreseeable, yet only a handful noticed."
Here we come, then, to the first lesson of The Big Short: the inconsistency between events recounted in the book and the narrative about the financial crisis that powered the regulatory legislation that Congress is considering. Among many in Congress, it is widely held that the financial crisis was caused by reckless Wall Street players selling low-quality financial products like CDOs, despite knowing that these securities were bound to default and cause losses to their customers. As Senator Carl Levin (D-Mich.) described the results of his committee's investigation: "[W]e learned that . . . Goldman Sachs repeatedly put its own interests and profits ahead of its clients . . . and when the system finally collapsed under the weight of those toxic mortgages, Goldman profited from the collapse." This and statements like it promoted the idea that Wall Street financial institutions were at least reckless, and probably worse, in selling CDOs and MBS based on subprime mortgages.
But the real story, as Lewis tells it, is virtually the opposite. Rather than being greedy and aware of the poor quality of the CDOs they were selling, the traders and their managers on Wall Street consistently resistedthe idea that it would be profitable to bet against these instruments, even though the returns from such a bet could be astoundingly high. In other words, far from being greedy and dishonest, these traders would be more easily characterized as dumb or deluded. Not only were they unwilling to bet against the subprime-mortgage market through CDSs when the profit possibilities were enormous, but they also continued to hold onto the very assets that others were telling them were almost certain to default when the housing bubble came to an end. As Lewis describes it, until actual losses were staring them in the face in late 2007 and early 2008, they even resisted the idea that they should hedge against the subprime-mortgage risks in their own portfolios. Only when the scale of defaults on the CDOs became obvious did the major institutions, in an attempt to reduce their losses, suddenly begin to buy the CDS protection that they had previously spurned
The Wall Street crowd is not the only group that acted this way. Lewis recounts the difficulties the money managers faced for having speculated against the subprime market by investing in what ultimately became highly profitable CDSs. Many of these positions had been purchased in 2005, when the housing market was still strong and CDSs were very cheap, but the subprime losses did not begin to show up until late 2007. In the meantime, the investors in these funds grew restless; they saw money being paid out in premiums but did not see any gains. They began to clamor for their investments back, which would have required the sale of the CDSs well before the subprime mortgages lost value and the CDS speculations became valuable.
This is fully consistent with testimony given before the Financial Crisis Inquiry Commission, of which I am a member. The chief executive officers of the major banks, senior officials of Citigroup who were responsible for the bank's mortgage business and risk management, the top officers of Fannie Mae, the ratings specialists at Moody's, and even Warren Buffett all admitted that they could not imagine losses on subprime mortgages on the scale reached in 2007 and 2008. Even the most informed financial regulators seemed unaware of the potential losses right in front of them. As late as May 17, 2007, Federal Reserve chairman Ben Bernanke reinforced the beliefs of all those who thought the possibility of a major meltdown in the mortgage market was overblown. "Given the fundamental factors in place that should support the demand for housing," he said, "we believe the effect of the troubles in the subprime sector or the broader housing market will be limited, and we do not expect significant spillovers from the subprime market to the rest of the economy or to the financial system."
Although it may be unsatisfying to blame the financial crisis on something as mundane as human pigheadedness, that should be the conclusion of any careful reader of The Big Short. The book certainly does not validate the views of those in Congress and elsewhere who place the blame on "Wall Street greed."
Asset Bubbles and Public Policy
Nevertheless, there is a major public policy issue implicit in The Big Short, and it is probably not one that Durbin or other supporters of the new legislation intended to endorse. Bubbles have always been a feature of economies and financial systems, including the huge housing bubble that drove and was driven by the origination and sale of subprime mortgages between 1997 and 2007. As Carmen M. Reinhart and Kenneth S. Rogoff outline in their recent, widely acclaimed work, This Time Is Different, financial and other bubbles are caused by wishful thinking and delusion, features of human nature. When prices for assets are rising, they tend to feed on themselves. Soon, the true value of an asset--what can be earned from its possession--is subordinated to the fact that it can be sold for more in the future. In the stock market, this is called "the greater fool theory": optimists believe there will always be a greater fool than themselves to buy the stock they bought in the hope that, against all reason, its price would continue to rise.
As the data in Reinhart and Rogoff's book imply, there are very few defenses against this market behavior. They cite hundreds of events over the past eight hundred years involving financial assets of all kinds. Nonfinancial assets--tulip bulbs in Holland in the 1600s are the classic case--are also subject to bubbles and crashes. In the classic work Manias, Panics, and Crashes, Charles Kindleberger and Robert Aliber attempt to find and describe the common elements in these events, including how and why they begin, develop, and ultimately crash. In reality, there are only two viable defenses against bubbles: effective government regulation and short selling. Effective government regulation is easy to conceptualize but difficult to implement. In theory, a government agency could decide that a bubble is developing and use its regulatory authority to reduce or deflate it. In practice, it is not easy to recognize bubbles, as Bernanke's 2007 comment on the subprime bubble demonstrates; even if recognized, regulators may be reluctant to act against them.
The current financial crisis is a case in point. The subprime and other weak mortgages that were at the heart of the crash in housing and mortgage values enabled large numbers of people to buy homes who previously had not had access to housing credit. The infusion of government funds into this market through Fannie Mae, Freddie Mac, and other government housing programs drove a ten-year housing bubble from 1997 to 2007 that grew to unprecedented size. Even if banking regulators recognized that a bubble was developing--a point that is still unclear--they would have been powerless to stop it. Attempts to do so would have met with universal disapproval in a Congress that, along with the Bush administration, was delighted with the growth in homeownership that resulted from these programs. (For the thirty years between 1964 and 1994, homeownership in the United States remained at about 64 percent, but by the time the bubble burst in 2007, the homeownership rate had hit a record high of over 69 percent). Whether it was the same blindness to reality that afflicted Wall Street, or timidity in the face of congressional opposition, U.S. regulators did not act to arrest the largest housing bubble in U.S. history, and the outcome was a worldwide financial crisis.
Although regulators cannot be relied upon to act against bubbles, at least one group--short sellers--has a real financial incentive to do so. Short sellers are those pessimistic folks who see opportunities in others' wishful thinking. In the equity markets, they borrow shares and sell them, hoping to reap a profit when share prices decline; if that happens, they buy the shares at the lower price, return the borrowed shares, and pocket the difference. They are active daily in the stock market but always under fire from those--especially corporate managements--who are both sensitive to criticism about the true value of their firm's shares and beneficiaries of rising share values. Whatever one thinks of short selling, there can be no question that it adds to selling pressure and thus tends to suppress the optimism that leads to the development of bubbles. In response to any concern about the growth of bubbles, short selling should be encouraged in all markets, not hindered.
The Big Short and the Importance of Short Selling
Herein lies another lesson of The Big Short. The book focuses on a group of short sellers in a debt market. Lewis does not point this out, but he is describing a new phenomenon. Until CDSs became available, it was not possible to sell short--to speculate against the prices of debt securities--in a debt market. In general, debt markets are too thin (that is, they do not have enough buyers and sellers) and buying and selling a debt security is too costly to permit active short selling. The advent of CDSs, which permit the risk of a debt exposure to be transferred apart from the debt security itself, changed those limitations. In the CDS market, a participant can easily sell short by contacting a market maker or dealer and negotiating a price for protection against the default of any publicly available security. This is possible--as it is in the equity markets through short selling--even if the buyer does not own the security he is speculating against and is not actually exposed to the risk. The money managers in The Big Short were, in effect, selling short through naked swaps by purchasing protection against defaults on various tranches of CDOs.
A CDS-based naked swap in a debt market has the same function and effect as a short sale of a stock in an equity market. In each case, it adds to the selling pressure on a security and--most importantly--supplies new information that aids in what is called price discovery. The most important function of any market is price discovery. Without markets, we would have no way of placing a value on a tangible or intangible asset. To decide, for example, that a barrel of oil is worth $50 or $80 today requires knowledge about its use, its transportation costs, its quality, its likely future availability, and dozens of other elements. These are all composed into a single price by supply and demand at a given moment in time. The more efficiently markets do this--that is, the more opinions about value that are included in a given offer to buy or to sell--the better the price discovery. In effect, greater liquidity reduces the spread between bid and asked prices. As the Squam Lake Working Group on Financial Regulation--a group of leading academic scholars in finance--stated in a recent report, "Buying and selling credit default swaps without the underlying bond is like buying and selling equity or index options without the underlying security. The advantages of these activities are well understood. Eliminating this form of speculation would make CDS markets less liquid, increasing the cost of trading and making CDS rate quotes a less reliable source of information about the prospects of named borrowers."
In the CDS market, as in any market, the presence of as many buyers and sellers as possible improves the price of the credit protection that is purchased. For example, the fact that someone wants to buy protection against the default of IBM means that the dealer must go into the market and find a counterparty willing to take that exposure. The price (the premium) at which the CDS contract is written communicates information about the confidence that the market has in IBM. If many buyers suddenly appear in the market looking to buy CDS protection on IBM, supply and demand alone--apart from other factors, including news about IBM--will cause the premium to rise. The opposite is also true; the fact that many market participants are offering to sell protection on IBM will cause the premium--known as the spread--to fall.
The ability to speculate against the value of subprime MBS and CDOs in the CDS market played a major role in The Big Short. The main individuals whom Lewis follows in the book made enormous profits by purchasing CDSs--in the form of naked swaps--as protection against the losses they thought would occur when the bubble deflated. Lewis describes how easy this was. In some cases, the cost of protection was considerably less than 1 percent, implying that market makers thought the likelihood of losses on these securities was substantially less than one in one hundred. Of course, this seems ridiculous now, but that is the nature of bubbles. When a bubble is developing, the vast majority of people believe that--as Reinhart and Rogoff put it--this time is different. There are always reasons prices, although inflated, do not seem out of line. Some participants in a bubble know it is going to deflate at some point, but they believe they can get out before this occurs. Others may believe the bubble is based on fundamentals and thus is not a bubble at all.
How to Defeat Bubbles
Given these human tendencies, the following policy question remains: how can we best protect against the growth of bubbles? Bubbles are harmful and destructive. When they deflate, they cause arbitrary losses and disrupt the economy, although few do the enormous harm of the mortgage meltdown of 2007 and 2008. Clearly, regulation has not worked to prevent bubbles, and there is good reason to believe it never will. The Fed had the authority to raise margin requirements on stocks during the dot-com bubble in the late 1990s and the authority to clamp down on subprime lending during the 2000s, but it did not act in either case. Perhaps the reasons for inaction were conceptual (like so many other market observers, the Fed simply did not recognize that a bubble was developing) or political (the Fed did not want to rile Congress by clamping down on markets in which constituents were prospering), but either way, actions that could have prevented those bubbles were not taken. The regulatory legislation now before Congress would set up a systemic risk council made up of the principal financial regulators and headed by the secretary of the Treasury. The purpose of the council would be to identify the bubbles as they are developing and to take the necessary action to deflate them or reduce their rate of growth. This might or might not work, but based on the track record of regulators in the past, this new plan for regulation does not seem promising.
The Big Short, however, points the way to a more effective system for combating bubbles. Future credit bubbles can be kept in check--or at least reduced in size before they collapse and cause major losses--if individuals and institutions are able to go short in the credit markets through the use of CDSs. For this to work, however, naked swaps would have to be permitted, even encouraged. Ordinary selling by those who already hold a particular debt security or are seeking only to hedge would not produce enough liquidity in a debt market to affect the price of protection substantially. Bid and ask spreads would remain wide, and the market would lack price discovery and would not give efficient signals about the credit quality of particular companies. Because neither party must actually own the debt security in a naked swap transaction, there is no necessary limit on the amount of liquidity and trading activity that could occur. Some traders will be speculating that the market's estimate of a company's risk is too great (they will sell CDS protection), while others will be speculating that it is not great enough (they will buy CDS protection). Naked swaps--and naked swaps alone--can make the CDS market efficient, liquid, and capable of producing credible market-based assessments of credit quality. By using naked swaps, the money managers in The Big Short demonstrated that there is financial opportunity in speculating against bubbles in debt markets, just as there is in selling short in equity markets. Unfortunately, too few individuals were willing to bet against the bubble in the years from 2005 to 2007, and when they finally began to recognize the losses that were occurring, they were too late to prevent the meltdown that caused the financial crisis.
Largely because of a limited understanding of the functions of markets, naked CDSs are under attack in both Europe and the United States. Many proposing to ban or limit naked CDSs claim that this activity is simply gambling, even though the benefits of short selling in equity markets--an analogous activity--are well understood. Beyond improvements in price discovery, naked swaps--if given the latitude to flourish--can offer significant protection against the development of future bubbles. As Professor René Stulz observes, short sales and naked CDSs are "critical to create conditions that make it more difficult for bubbles to emerge. . . . Investors help [bring] prices back into line with fundamentals through their speculation. So rather than forbid or restrain short-sales and naked CDS positions we should instead make them easier and less expensive. Derivatives that allow individuals to hedge risks that matter to them in their daily life have to be encouraged, . . . [as should] derivatives that allow investors to take bets against what they think is a bubble."
Fortunately, the legislation now in the House and Senate does not contain overt restrictions on naked CDSs. However, the legislation would place significant burdens on those who might function as buyers or sellers of protection in the CDS market and, thus, on those who might be willing to sell short in the midst of a bubble by purchasing protection against losses when the bubble deflates. For example, the legislation defines a "major swap participant" as one who maintains a "substantial position" in swaps, creates "substantial counterparty exposure," or is "highly leveraged." Major swap participants are ultimately going to be defined by regulations of the Commodity Futures Trading Commission (CFTC) and the Securities and Exchange Commission (SEC). Major swap participants are required to have particular capital levels and provide margin (collateral) for trades. What these terms will eventually mean will also be determined by the SEC and CFTC. Given the suspicions in Congress about CDS activity, it is likely that the regulations will impose heavy costs on the market participants, such as hedge funds, that are most likely to act as short sellers. This would encourage, not prevent, the growth of bubbles in the future. As Stulz notes, "In terms of market regulations, the focus should be on helping markets do their job rather than limiting their scope."
Although regulations that severely restrict CDS activity would be consistent with the views of the senators and representatives who designed the regulatory legislation now under consideration, such regulations ignore the policy lesson implicit in The Big Short--that to reduce or prevent bubbles, we need more, not fewer, market participants willing to take positions contrary to the prevailing views of the majority. Before voting to impose these restrictions, it would be a good idea for Senators Durbin and Levin and their House and Senate colleagues to read The Big Short.
Peter J. Wallison (firstname.lastname@example.org) is the Arthur F. Burns Fellow in Financial Policy Studies at AEI.
1. Michael Lewis, The Big Short: Inside the Doomsday Machine (New York: W. W. Norton, 2010).
2. Although they can be understood as insurance, CDSs are not true insurance and are not regulated as such. Insurance is an actuarial system in which protection is written on a large pool of insured parties. It is not known in advance which of the insured parties will suffer a loss, and when a loss occurs the insured party is reimbursed only for the amount of the loss. A CDS is a contract that provides protection against a specific entity's default on its credit obligations; in the event of a default, the protected party receives a stated value independent of its actual loss.
3. For a more complete description of CDSs and related issues, see Peter J. Wallison, "Everything You Wanted to Know about Credit Default Swaps--but Weren't Told," AEI Financial Services Outlook (December 2008), available at www.aei.org/outlook/29158; and Peter J. Wallison, "Unnecessary Intervention: The Administration's Effort to Regulate Credit Default Swaps," AEI Financial Services Outlook (August 2009), available at www.aei.org/outlook/100065.
4. Michael Lewis, The Big Short: Inside the Doomsday Machine, 147.
5. Ibid., 105.
6. Senator Carl Levin, "Statement of Sen. Carl Levin, D-Mich., on Senate Passage of the Restoring American Financial Stability Act of 2010," news release, May 20, 2010, available at http://levin.senate.gov/newsroom/release.cfm?id=325147 (accessed June 28, 2010).
7. Ben S. Bernanke (speech, Federal Reserve Bank of Chicago's Forty-third Annual Conference on Bank Structure and Competition, Chicago, Illinois, May 17, 2007).
8. Carmen M. Reinhart and Kenneth S. Rogoff, This Time Is Different: Eight Centuries of Financial Folly (Princeton, NJ: Princeton University Press, 2009).
9. Charles Kindleberger and Robert Aliber, Manias, Panics, and Crashes: A History of Financial Crises, 5th ed. (Hoboken, NJ: Wiley, 2005).
10. Squam Lake Working Group on Financial Regulation, "Credit Default Swaps, Clearinghouses, and Exchanges" (working paper, Council on Foreign Relations, Center for Geoeconomic Studies, Washington, DC, July 2009), available at www.cfr.org/publication/19756/credit_default_swaps_clearinghouses_and_exchanges.html (accessed June 22, 2010) .
11. Ben Moshinsky, "Emergency Bans on Naked CDS Trades Considered by EU," Bloomberg.com, June 14, 2010, available at www.bloomberg.com/news/2010-06-14/naked-credit-default-swap-trades-should-be-banned-eu-lawmaker-report-says.html (accessed June 22, 2010).
12. Lynn A. Stout, "Regulate OTC Derivatives by Deregulating Them," Regulation (Fall 2009): 30, available at www.cato.org/pubs/regulation/regv32n3/v32n3-1.pdf (accessed June 22, 2010).
13. René Stulz, "What Should We Do about Bubbles?" Harvard Business Review blog Finance: The Way Forward, June 8, 2010, available at http://blogs.hbr.org/finance-the-way-forward/2010/06/what-should-we-do-about-asset.html (accessed June 28, 2010), emphasis added.