Berkshire Hathaway
Derivatives Collapse?

Format for Printing

Format for printing

Request Reprints


By niccraven
September 19, 2003

Posts selected for this feature rarely stand alone. They are usually a part of an ongoing thread, and are out of context when presented here. The material should be read in that light. How are these posts selected? Click here to find out and nominate a post yourself!

I recently wrote this essay on the potential for derivatives to lead to a financial collapse, as part of the assessment for a risk management paper that I'm doing towards my honours degree in finance.
I've posted it here as it seems relevant to some of the topics that come up from time to time, although it's long and maybe not exactly news to many posters on this board. Anyway, here it is if anyone's interested, please excuse any errors that I've made. Oh, I've copied the text over from MS word and the footnotes don't seem to have copied... my apologies for this.

On 15 August 2003, 50 million American and Canadian citizens were plunged into darkness in the most severe electricity blackout of American history. While the causes of the blackout are still being investigated at the time of writing, what seems clear is that the magnitude of the blackout was due to a cascade effect where the problems in one area spread to affect the neighboring areas. The last severe blackout to hit the eastern cities of the USA was in 1977, after which citizens were assured that such an event could not happen again1. The blackout demonstrates that in a crisis correlations may appear between events normally thought to be unrelated � the power supply in southern Canada is normally independent of the power supply in the eastern USA, but the crisis grew through a chain reaction to encompass both areas.

This essay investigates the potential for financial derivatives to cause such a cascade effect in global financial markets. Organizations that would normally be thought to be independent of each other may find themselves intrinsically linked through a chain of derivative contracts. In this way a default by a single firm may trigger a series of further defaults that might grow into a financial crisis.

The essay does not attempt to be a comprehensive description of the various types of derivative contracts available, but some types of contract are mentioned in detail in order to illustrate certain points. Some existing knowledge of derivatives is assumed on the part of the reader.

A historical context, measurement issues

Derivatives, as a class of financial securities, are not a new phenomenon: Aristotle in his Politics referred to an option over the use of olive presses. Delayed payment agreements resembling modern futures contracts were common during the 17th century tulip bubble in Holland, and trading in such instruments was exported to London later that century where they became known as 'time bargains'. It is during the last twenty years however that the use of derivatives has exploded, both in the size of the contracts outstanding and in the variety of contracts available. Measured by the notional value of current contracts, the quantity of derivatives on issue has risen from $1 trillion in 1987 to $128 trillion in 2003. The use of derivatives to manage risk has become common practice in the business arena. A 1999 survey of the US-based Association of Finance Professionals found that 63% of respondents' organizations used derivatives traded on an over-the-counter (OTC) basis.

This growth has stemmed from the increasing liberalization of trade world-wide, and the internationalization of economic life that has been experienced in recent history. With the advent of globalization in a post Bretton-Woods world, an increasing number of organizations have felt the need to manage the risk posed to their operations by such factors as variations in currency values and interest rates. Derivatives fill such a gap, while also playing a role in facilitating the greater integration of global capital markets.

The murkiness of the world of derivatives begins with the measurement of their use. The figures referred to as 'notional value' measure the dollar value encompassed by a derivatives contract, such as a forward contract. If company A signs an agreement with bank B to exchange US$1m for NZ$1.7m in six months' time, the notional value of the contract in NZ dollars would be $1.7m. Such an agreement would be subject to counterparty risk. If bank B defaulted within the six months time period then company A would be left without the cover provided by the forward contract. However, the loss to company A would not be the notional value of the contract, NZ$1.7m, instead the magnitude of the loss would be the cost to A of replacing the contract. This may be large or small (or negative) depending on the exchange rate at the time of default, but in any case it is unlikely to be anywhere near the contract's notional value. The total exposure of US banks (which own the vast majority of derivative contracts) to defaults as measured using this cost of replacement methodology was estimated in 2002 to be US$500b. While still large, this figure is less than one percent of the US$56 trillion in notional value of contracts held by US banks at that time.

Derivatives as grease in the economic machine

Like insurance, derivatives play an important role in transferring risk between organizations. A manufacturing company with exposure to foreign currency risk can reduce or eliminate that risk by taking out forward contracts and locking in today an exchange rate that they can use in the future. Alternatively, a company that relies on a commodity as a key raw material may wish to avoid the impact of short or medium-term changes in its cost. It may do so by entering into a forward contract with the commodity in question acting as the underlying asset. Such hedging operations allow companies who wish to avoid risk to trade the risk away in a derivatives contract. This results in greater predictability with regards to future costs and allows them to focus on what is important to their business.

Interest rate risk may be reduced by similar measures, under a type of contract known as an interest rate swap. A company A with floating rate debt may enter into an agreement with another party whereby the two parties effectively exchange interest regimes with each other. Company A effectively assumes a fixed-rate obligation, and transfers the uncertainty of a floating interest rate to the counterparty of the contract.

In general, the transfer of risk allows a greater degree of specialization in the economic roles of different organizations. US banks have managed to expand the quantity of loans they are able to make by entering into derivative contracts based on the credit-worthiness of their borrowers. Essentially, the banks have bought credit protection and in the process have laid off some of the credit risk that they face onto more willing parties (mainly insurance companies). The banks can then do what they do best, making loans, while the credit protection contracts allow them to make more efficient use of their capital. The greater quantity of loans being made has important positive ramifications for the economy as a whole, by channeling more capital to a greater number of profitable uses.

US Federal Reserve chairman Alan Greenspan has said that the smoothing role played by derivatives and the increased flexibility that they provide to organizations has raised living standards, accelerated globalization and helped the United States economy avoid recession. Commenting in March 2003 on the performance of the US economy amid the stock market decline of the past several years, the fall in capital investment and the September 11 terrorist attacks, Mr. Greenspan told a Banque de France symposium "There can be little doubt that globalization and global finance have been a major contribution to the flattening of the business cycle. Despite all of that, the American economy has seemingly been able to struggle along without collapsing into a much deeper recession, which certainly would have been the case in the period 20 or 30 years ago."

Derivatives as tools for speculation

It is axiomatic that derivative contracts themselves do not eliminate risk, they merely transfer it. Certainly a single agency may act as counterparty to many different agreements, and thus be able to offset the risk of one agreement against another. This would result in a situation akin to an insurance company that holds a portfolio of risks whose effects can be better predicted in aggregate than individually. In such a case, while the quantum of risk that exists has not been reduced, derivative contracts have facilitated its transfer to parties that are more able to deal with it. However, it is not necessary that the counterparty to a hedging contract always play such a role. Derivatives have for a long time been used as a tool for speculation with concrete examples dating back to the 17th century, as mentioned above. By trading using a derivative agreement such as a futures contract, a speculator can draw profits (or losses) from underlying assets worth the notional value of the contract. Also, entering into such a contract requires relatively little collateral relative to the notional value involved. In this way derivatives make it easy to leverage small amounts of capital into very large investments. A case in point is the hedge fund Long-Term Capital Management, which with about $5b in capital made use of debt and derivative agreements to control investments worth $1.25 trillion. The fund's ignominious collapse followed soon after.

It may be argued that hedge funds such as LTCM are legitimate players in the role of transferring risk. Hedge funds and speculators are willing to bear risk that their business counterparties may not be � thus there is an economic benefit to speculative transactions in transferring risk away from firms. However there are numerous types of derivative contracts whose hedging properties are dubious and whose only conceivable purpose is to facilitate speculation. Chancellor (2000) asks, "What conceivable risk exposure ... is a 'LIBOR-cubed swap' � a security that multiplies by three times changes in the London Interbank Offered Rate, the rate of interest in the wholesale money market � designed to hedge? And to what bona fide purpose is a 'Texas hedge,' a combination of two related derivative positions whose risk is additive rather than offsetting?" Without making a judgment on the use of derivatives for speculative purposes, it is worthwhile to be aware of this role that they play.

Risk of derivative contracts increasingly difficult to evaluate

While most of the derivative contracts in force are 'plain swap' agreements, there are a dramatically increasing amount of more complex contracts. An example are so-called 'synthetic' agreements, which combine different properties in such a way as to be completely different from their underlying asset. One type of synthetic security might be collateralized mortgage obligations (CMOs). CMOs are created during the process of securitizing mortgages. The cashflows from a group of mortgages are split into principal and interest components, with different securities being issued which derive their payoffs from different components. The interest components may be further split into different tranches with higher-ranking tranches having a priority claim on the interest received. Thus lower ranking tranches carry a greater amount of default risk. The overall result is that a relatively simple security, a securitized mortgage, has been turned into numerous component securities which each bear a complex relationship with a wide variety of economic variables. For example, the holder of a security whose value is dependent on the principal component of a group of mortgages would benefit during a period of declining interest rates, when large-scale mortgage refinancing leads to the principal on loans being repaid earlier than expected.

The trend towards greater complexity is partly being driven by investment banks that are focusing on "adding value and customizing products to the specific needs of the client". Eric Bertrand, managing director of fixed income derivatives at J.P. Morgan is on record as saying that 'structured derivatives' (customized derivatives that are not traded on an exchange) are growing at 30 to 40 percent per annum.

The increasing complexity of derivatives makes them difficult to value, and therefore prone to mispricing. This pitfall has been recognized by sophisticated investors such as Warren Buffett and George Soros. Soros in 1994 told the House Banking Committee that "there are so many of them [derivatives] and some are so esoteric that the risk involved may not be properly understood even by the most sophisticated investor, and I'm supposed to be one. Some of these instruments appear to be specifically designed to enable institutional investors to take gambles which they would not otherwise be willing to take."

Some firms that traded in credit protection derivatives (CDS) were hurt financially by apparent confusion over what constituted a 'credit event' for the purposes of the contract. Under a CDS agreement the buyer of protection might have the right to sell bonds (the underlying asset) to the counterparty at par in the event of a credit event occurring. Ambiguity over the definition of a credit event came home to roost when Xerox restructured its balance sheet in 2001. Its bank debt was increased to solve a liquidity problem, but Xerox bonds fell in value as they were ranked junior to the increased bank debt. Holders of credit protection contracts on Xerox bonds declared that a credit event had occurred and demanded that their counterparties purchase their holdings of Xerox bonds at par. The sellers of credit protection had not anticipated this definition of a credit event and had as a result mispriced the contracts that they sold. Since then CDS have been sold specifically with or without a provision for a 'restructuring event' such as occurred with Xerox. The difference in price between the two types of securities ranges from 15 to 20 percent. The bottom line is that the sellers of the contracts failed to understand completely the implications of contracts they were selling.

Illiquidity and its implications

Due to the specialized nature of derivative contracts, trading in them can be illiquid depending on the type of contract. Liquidity is growing in more 'traditional' derivatives such as plain swaps, but is lacking particularly in the more complex, OTC-based contracts. The lack of liquidity adds to derivatives' valuation difficulties due to the lack of a reliable market price. Rather than use a 'mark-to-market' valuation methodology, organizations resort to 'mark-to-model' where they essentially try to value their contracts themselves. By definition a valuation methodology such as mark-to-model is only as good as the model that is used. Due to the complex nature of some derivative contracts, one would expect significant valuation errors to occur.

The lack of liquidity also poses a serious risk in the event of a major adverse market movement. In such a situation what liquidity there is could well dry up, leading to large-scale losses, as derivative-trading firms are unable to unwind their positions. It has been argued that once the possibility of such an event is taken into account, some types of derivatives may be fundamentally mispriced. Indeed, such an event has already occurred, namely the LTCM collapse in 1998. Faced with a sudden fall in capital and with contracts outstanding having a notional value of $1.25 trillion, the hedge fund found itself in dire straights. Due to the extended influence of LTCM through its derivative agreements, Wall Street's major banks were forced to step in and bail out the hedge fund rather than let the global financial markets descend into chaos. It was estimated at the time that had the banks not stepped in and instead left LTCM to try and unwind its contracts by itself, it would have led to losses of $14 trillion, a huge sum by any measure. The situation was one of a crisis of confidence, as liquidity in the markets in which LTCM traded dried up when rumors of its troubles spread. This would have magnified the size of losses had LTCM's contracts had to be unwound. In the event, the bailout package eventually stabilized the situation, however parts of the fixed-income markets remained paralyzed for weeks. Such was the impact on liquidity when disaster was averted; one can only imagine what the situation may have been otherwise.

Another practice that would be inherently unstable should liquidity disappear is the process of 'dynamic hedging' or 'delta hedging', used by the major derivative-dealing banks. The process is necessary because the banks must hedge against changes in the value of the trillions of dollars of derivative securities on their books. This involves buying the underlying assets when they rise and selling them when they fall. The effectiveness of the strategy depends on market liquidity. George Soros has warned that during a market panic, the widespread sales of securities by firms practicing delta hedging could lead to a severe "financial dislocation".18

Linkages and counterparty risk

Derivatives differ from conventional securities in that the parties to a derivatives contract become bound to each other. In a transaction involving a conventional security, one party sells and transfers title of the security to the other party. There are no ongoing obligations. This is not the case for derivative contracts. Inherent in the nature of derivatives is some future obligation on the part of one party to the other. Should the party with the obligation fold, their counterparty may be out of pocket. This concept is known as counterparty risk and has been briefly discussed earlier. Counterparty risk is something that organizations that trade in derivatives must consider, however it is difficult to quantify. The problem would not be so bad were it not for the widespread use of derivatives binding many companies to each other in many different ways. A 1994 Economist article states: "The more that derivatives proliferate, the more companies will be linked together by these elongated promises and the harder it will be to arrive at a proper assessment of the risks they bear."19 Essentially, the solvency of one company that deals in derivatives becomes linked to the solvency of their counterparties, which in turn are linked to the solvency of their counterparties, and so on. This makes counterparty risk basically unanalyzable, in the opinion of William Nemerever, co-manager of a $26 billion US-based fixed income group � "It's an unanalyzable risk because you don't have access to the multifaceted risk that your counterparties are assuming."

The counterparty risk posed by derivatives becomes more serious when consideration is taken of the concentration of risk among relatively few organizations. Just seven US banks own 96% of the derivatives in the banking system. This concentration may increase in the future, as counterparty risk is perceived as being of greater importance and organizations become more careful about which firms they are willing to use as counterparties. With the organizations among which derivative use is concentrated trading extensively with each other in turn, their fortunes are inextricably linked. In addition, these organizations may share similar worldviews leading to systemic mispricing of contracts and a potentially volatile market should those views prove wrong.

A combination of extensive linkages between organizations due to counterparty risk, mispricing of derivative contracts and concentration of risk among few parties could cause a chain reaction of defaults within the financial industry. Such a scenario would start with one party defaulting on its obligations, causing one or more of their weaker counterparties to default. The effects would begin to spread and liquidity would begin to dry up, magnifying paper losses and restricting the efforts of firms to extricate themselves from the looming crisis. As losses mount, the number of firms that appear insolvent would increase exponentially, and eventually a major counterparty to many of them could collapse � this may be one of the big seven US banks mentioned earlier. By this stage the other major banks would be hurting enough to have difficulty structuring a rescue package � disaster may ensue. Diversification across different counterparties may not be an effective strategy to avoid being caught up in such a crisis, due to the linkages inherent between different firms. According to Warren Buffett, "A participant may see himself as prudent, believing his large credit exposures to be diversified and therefore not dangerous. Under certain circumstances, though, an exogenous event that causes the receivable from Company A to go bad will also affect those from Companies B through Z. History teaches us that a crisis often causes problems to correlate in a manner undreamed of in more tranquil times."

Transparency issues

Part of the derivatives issue is the lack of transparency. Uncertain accounting treatment of derivative transactions provides the potential to hide undesirable numbers. Mark-to-model based valuation in particular allows considerable leeway to come up with different results simply by changing the model. Warren Buffett talks about mark-to-model degenerating into 'mark-to-myth'.23 His business partner Charlie Munger has said that in aggregate derivative portfolios are likely to be overvalued � because overvaluation errors are likely to be more common than undervaluation errors, whether by simple over-optimistic human nature or deliberate fraud.

The actions of companies in resisting new accounting standards in this area is worrying, such as pressure earlier this year from French and Germans banks (who, it was worried, had excessive exposure to credit derivatives) against the European Commission and its plan to adopt an international accounting standard that would require banks to value their derivatives using the mark-to-market method. Due to the complex nature of derivatives, it is by no means certain what the correct accounting treatment should be. What is certain, however, is that current methods of disclosure do not adequately convey the necessary information to evaluate the risks posed by derivative contracts within an organization. Warren Buffett has said that when he finishes reading the footnotes detailing the derivative portfolios of major banks, "the only thing we understand is that we don't understand how much risk the institution is running." He is not alone in his views. According to the chairman of the International Accounting Standards Board, Sir David Tweedie, "the derivatives we have out there can destroy companies... someone [could be] making a massive loss on derivatives at the moment and you would not even know".

Regulation on trading

Regulation of derivatives trading has been proposed several times in recent years. In 1998, following the LTCM debacle, and a bill regarding trading in energy derivatives was recently defeated in the US. Opponents of regulation argue that the oversight provided by bank inspectors with respect to banks' derivative portfolios has become more rigorous, and that banks have developed sophisticated 'value at risk (VAR)' processes for monitoring the risk that they bear. However VAR, based on historical volatility, has its pitfalls and it is difficult to capture 'mega-catastrophe' risk in such a model. Benoit Mandelbrot, famous for fractal geometry, describes risk modeling as "a theory of sea waves that forbids their swells to exceed six feet." Currently there are no real regulatory limits on organizations' activities in derivatives. It is possible that some amount of regulation would help matters, but what is certain is that whoever was to be given the task of regulation would have a hard time keeping up with the growing complexity in the derivative contracts being used.

At a micro level, derivatives are a useful economic tool. They allow risk to be transferred from organizations that do not wish to hold it to organizations who are willing to hold it for a price, or who have an opposite risk to offset it against. So far the headlines such as LTCM and Enron are not representative of derivatives as a whole � most derivative transactions during the 1990s were problem free. However, the risk remains that illiquidity, mispricing and linkages inherent between counterparties could combine to transform a single reasonably sized default into a major dislocation for global financial markets. The probability of this happening is impossible to quantify, but it is not zero. What the probability is depends on the risk management processes used by those organizations that trade in derivatives, and how well they understand the contracts they transact in, as well as those of their counterparties. This is a lot for an organization to consider, and errors are almost certain to occur from time to time. A certain British bank chairman once said, "Derivatives need to be well controlled and understood, but we believe we do that here." The bank, Barings, collapsed shortly afterwards.

Anonymous (2003, Mar 8) Greenspan, Unlike Buffett, Sees Derivatives as Positive Influence New York Times, New York, pp.C14

Anonymous (1994, May 14) Derivatives: The beauty in the beast The Economist London, pp.21

Buffett, W. (2003) 2002 Letter to shareholders of Berkshire Hathaway

Chancellor, E. (2000) Devil take the hindmost: A history of financial speculation New York: Farrar, Straus and Giroux

Coy, P. (2003, Mar 31) Are derivatives dangerous? Business Week, New York, pp. 90

Davenport, T. (2003, Jul 1) Assessing risk: Peeling apart data on derivatives American Banker New York, pp.1

Fitch, T. (1994, Aug) What financial derivatives bring to pension portfolios Corporate Cashflow 15(9) Atlanta, pp.16

Foroohar, R. (2003, May 12) The Alarmist of Omaha; Are 'financial weapons of mass destruction' for real? Newsweek New York, pp.50

Gascoigne, C. (2003, Apr 30) Investing in derivatives London, pp.1

Guerrera, F., Parker, A. and Pretzlik, C. (2003, Mar 10) As criticism of derivatives grows, regulators push for tougher accounting rules Financial Times London, pp.19

Haubrich, J. (1995, Sep 1) Derivative mechanics: The CMO Federal Reserve Bank of Cleveland, Economic Commentary Cleveland, pp.1

Kettel, B. (1999, Feb) Derivatives: Valuable tool or wild beast? Balance Sheet 7(2), Bradford, pp. 14

Knox, L. (2003, Jan) Selecting the safest route Institutional Investor 37(1) New York, pp.55

Leander, T. (1999, Mar) The perils that lurk in risk models Global Finance 13(3), New York, pp.10

Leone, M. (2003, Jun 16) DOA: Derivatives Oversight Agency Boston, pp.1

McCarthy, E. (2000, May) Derivatives revisited Journal of Accountancy 189(5) New York, pp. 35

Reuters (2003, Aug 17) Normality returns after North American Blackout The New Zealand Herald

Rizzi, J. (2003, Jul) Risk implications of credit derivative instruments Commercial Lending Review 18(4) New York, pp.15

Become a Complete Fool
Join the best community on the web! Becoming a full member of the Fool Community is easy, takes just a minute, and is very inexpensive.