9 Systemic Challenges Posed by Greater Reliance on Over-the-Counter Derivatives Markets
- Garry Schinasi
- Published Date:
- December 2005
Rapid growth, development, and widespread use of over-the-counter (OTC) derivatives markets have accompanied the modernization of commercial and investment banking and the globalization of finance. Both modernization and globalization have been driven by recent advances in information and computer technologies, and have contributed significantly and positively to the effectiveness of national and global finance, particularly to the effectiveness of international financial markets. Much has been written about derivatives as financial instruments and about the role of highly leveraged institutions. By contrast, less has been written about the markets for OTC derivatives and the heavy reliance on them by the small group of internationally active financial institutions. This chapter attempts to fill part of this gap.86
Derivatives bestow considerable benefits by allowing financial risks to be more precisely tailored to risk preferences and tolerances; they contribute to more complete financial markets, improve market liquidity, and increase the capacity of the financial system to bear risk and intermediate capital. Derivatives instruments, the structures for trading and risk-managing them, and the infrastructures for ensuring their smooth functioning play a central role in the smooth functioning of the major financial and capital markets. These instruments and markets have been designed and developed by the internationally active financial institutions that derive a large share of their earnings from these activities. These are the same financial institutions that make up the core of the international financial system and have access to financial safety nets.
While derivatives instruments and markets have improved the effectiveness of intermediation and finance generally, and are likely to continue to do so, as crises in the 1990s demonstrated OTC derivatives activities can contribute to the buildup of vulnerabilities and adverse market dynamics in some circumstances. The severity of repeated episodes of turbulence in the 1990s and early 2000s suggested at that time that OTC derivatives activities were capable of producing fragility and of threatening stability, in some cases akin to a modern form of traditional bank runs. A most relevant example occurred in the autumn of 1998 as revealed in the contours of the market dynamics in the aftermath of the near-collapse of Long-Term Capital Management (LTCM). A substantial buildup in derivatives credit exposures and leverage contributed significantly to this turbulence. This substantial leverage—LTCM accumulated $1.2 trillion in notional positions on equity of $5 billion—was possible primarily because of the existence of large, liquid OTC derivatives markets. The virulence of the 1998 turbulence in the mature financial markets took market participants and authorities by surprise, and some—including U.S. Federal Reserve System Chairman Alan Greenspan (Greenspan, 1998) and former Bundesbank President Hans Teitmeyer (Teitmeyer, 1999)—acknowledged at that time that they did not fully understand the rapidly changing structure and dynamics of global financial markets.87
While some progress has been made since then, market structures have continued to change and evolve, leading to a growing recognition that understanding markets is a never-ending challenge and that these dynamic markets require vigilant surveillance and supervision. The issue has remained so important that the systemic implications of, and challenges posed by, continuing structural change in finance was discussed years after the events of 1998 by the President of the Federal Reserve Bank of New York, Tim Geithner (Geithner, 2004). On July 27, 2005, as this book was being completed, the Counterparty Risk Management Group II (CRMPG II), led by Gerald Corrigan—a managing director at the investment bank Goldman Sachs and former president of the Federal Reserve Bank of New York (FRBNY)—called for “urgent” action by the financial industry to strengthen the market infrastructure for confirming and processing complex financial transactions, including credit derivatives.88 Moreover, in late August 2005, the FRBNY invited 14 key derivatives market participants (including Citigroup, Deutsche Bank, Goldman Sachs, JPMorgan, Merrill Lynch, and Barclays Capital) and other regulators (including the U.S. Federal Reserve Board, the U.S. Securities and Exchange Commission, the U.K. Financial Services Authority, and both German and Swiss market regulators) to attend a meeting (on September 14, 2005) to discuss what it called “important” issues in the credit derivatives market.89 The issues to be discussed apparently include problems that surfaced during a sharp increase in trading volumes in credit derivatives markets following sharp credit-rating downgrades of bond issues of General Motors and Ford during the summer of 2005.
While difficult to measure, there can be little doubt that the net private and social benefits of OTC derivatives markets are overwhelmingly positive. If these markets are so overwhelmingly beneficial, why is it necessary to focus attention on the potential downside risks? Three reasons follow:
- OTC derivatives markets are mostly unregulated, except indirectly through the regulation and supervision of financial institutions.
- Disclosure is poor and market transparency is an issue.
- OTC derivatives markets are often a flashpoint for problems early in the process of fragility, yet conducting surveillance of these markets is difficult because of their limited transparency.
In short, because of their importance in global finance, it is important to understand more fully the potential capacity for the OTC derivatives activities of internationally active financial institutions to contribute to international financial volatility and perhaps fragility and thereby threaten international financial stability.
The chapter is organized as follows: Taking the benefits and efficiency-enhancing characteristics of OTC derivatives as a given, the next section of this chapter begins with a brief discussion of modern financial intermediation. It reveals that internationally active financial institutions have become exposed to additional sources of fragility because of their large and dynamic exposures to the counterparty credit risks embodied in their OTC derivatives activities. Before identifying these sources of fragility, potential volatility, and threats to stability, the succeeding section compares OTC with exchange-traded derivatives, including their respective trading environments. This comparison reveals significant differences in the way in which private and collective risks are managed and suggests that OTC activity may be more prone to producing systemic risks. The next section discusses that certain features of OTC instruments, modern financial institutions, and the underlying OTC infrastructures can pose risks to stability, separately and jointly, that in some circumstances create the tendency toward volatility, fragility, and instability in global financial markets. It is easier to identify potential threats to stability in OTC derivatives markets than it is to find remedies, which can only be pragmatically formulated and implemented by private and official practitioners in these markets. Nevertheless, in the concluding section the chapter points to both private efforts (more effective market discipline, risk management, and disclosure) and public efforts (strengthened incentives for market discipline, removal of legal and regulatory uncertainties, and improvement of effectiveness of OTC market surveillance) required if the risks to stability are to be contained in modern OTC derivatives markets.
OTC Derivatives Markets and Modern Banking
Large internationally active financial institutions have transformed the business of finance dramatically since the late 1980s. In doing so, they have improved the ability to manage, price, trade, and intermediate capital worldwide. Many of these benefits come from the development, broadening, and deepening of OTC derivatives and a greater reliance on OTC derivatives activities and markets. Although modern financial institutions still derive most of their earnings from intermediating, pricing, and managing credit risk, they are doing increasingly more of it off balance sheet, and in less transparent and potentially riskier ways. This transformation accelerated during the 1990s and the early 2000s.
A brief examination of the difference between traditional banking and modern banking provides a sense of the importance and extent of this transformation. Traditional banking involves extending loans on borrowed funds (deposits) of different maturities. Each side of this ledger has different financial risks. A simple loan is for a fixed sum, term, and interest rate; in return the bank is promised a known schedule of fixed payments. The risk in lending, of course, is that the borrower may become unable or unwilling to make each fixed payment on schedule. This is credit (or counterparty) risk,90 composed of both the risk of default (missing one or all payments) and the expected loss given default (that less than is promised is paid). Loans are funded by deposits with much shorter maturities than most bank loans, a situation that imparts liquidity risk. The basic business of banking is to manage these two sets of cash flows, each having a different, stochastic structure. As the history of bank runs and failures indicates, managing these cash flows is inherently risky and banking is prone to fragility and, at times, instability. (See Bryant, 1980; Diamond and Dybvig, 1983; and Kindleberger, 1996.)
This tendency toward fragility may have increased with the further development and more widespread use of OTC derivatives markets in the 1990s and early 2000s, including in emerging markets (Mathieson and others, 2004). In modern finance, financial institutions’ off-balance-sheet business involves extensions of credit. For example, a simple swap transaction is a two-way credit instrument in which each counterparty promises to make a schedule of payments over the life of the contract. Each counterparty is both a creditor and debtor and, as in traditional banking, the modern financial institution has to manage the cash inflows (the creditor position) and outflows (the debtor position) associated with the derivatives contract. There are important differences, however. First, the embedded credit risk is considerably more complicated and less predictable than the credit risk in a simple loan, because the credit exposures associated with derivatives vary with time and depend on the prices of underlying assets. Traditional bank lending is largely insulated from market risk because banks carry loans on the balance sheet at book value, which means that they may not recognize and need not respond to market shocks. Nevertheless, market developments can contribute to unrecognized losses that can accumulate over time. By contrast, OTC credit exposures are subject to volatile market risk and are, as a matter of course, marked to market every day. This creates highly variable profit and loss performance, but imparts market discipline and prevents undetected accumulations of losses. Day-to-day shifts in the constellation of asset prices can have a considerable impact on credit risk exposures—both the exposures borne by any particular financial institution and the distribution and concentration of such exposures throughout the international financial system.
Second, the liquidity dynamics of modern finance are considerably more complex than those of deposit markets. Deposit flows have a degree of regularity associated with the flow of underlying business. By contrast, flows associated with OTC derivatives and liquidity conditions in these markets, and in related markets, can be highly irregular and difficult to predict, even for the most technically advanced dealers with state-of-the-art risk management systems. Overall, the stochastic processes that govern the cash flows associated with OTC derivatives are inherently more difficult to understand and are more unstable during periods of extreme volatility in underlying asset prices.
Thus, in addition to assessing and managing the risk of default and the expected loss given default, the modern financial institution has to assess the potential change in the value of the credit extended and form expectations about the future path of underlying asset prices. This, in turn, requires an understanding of the underlying asset markets and establishes a link between derivatives and underlying asset markets.
The unpredictable nature of OTC derivatives markets would merit little concern if OTC derivatives were an insignificant part of the world of global finance. They are not, and they are increasingly central to global finance. OTC derivatives markets are large, at end-2004 consisting of nearly $248 trillion in notional principal, the reference amount for payments; nearly $9.1 trillion in gross market value; and nearly $2.1 trillion in off-balance-sheet gross credit exposures after controlling gross market values for legally enforceable netting and other risk reducing arrangements (Tables 9.1 and 9.2). The markets are composed of systemically important financial institutions, and together the instruments and markets interconnect the array of global financial markets through a variety of channels.91
|Derivatives Dealers||Rank||Members of Exchanges|
|J. P. Morgan||1||2||X||X||X||X||X|
|Deutsche Bank Securities||2||4||X||X||X||X||X|
|Goldman, Sachs & Co.||5||3||X||X||X||X||X|
|Credit Suisse First Boston||7||8||X||X||X||X||X|
|Bank of America Securities||9||10||X||X||X||X|
|Royal Bank of Scotland||12||11||X||X||X||X||X||X|
|Bear, Stearns & Co.||17||17||X||X||X||X|
|Credit Agricole Indosuez||18||n.a.||X||X||X||X|
|Dresdner Kleinwort Wasserstein||19||n.a.||X||X||X||X|
|Banc One Capital Markets||20||n.a.||X||X||X||X||X|
|Notional Amounts||Gross Market Values|
|With other reporting dealers||6,842||7,954||8,660||10,796||11,664||285||284||395||247||485|
|With other financial institutions||7,597||8,948||9,450||10,113||11,640||377||427||535||352||665|
|With nonfinancial customers||4,009||5,168||6,365||6,088||6,271||220||286||370||267||412|
|By remaining maturity1|
|Up to one year||14,522||17,543||18,840||21,252||23,115||…||…||…||…||…|
|One to five years||2,719||3,128||3,901||3,912||4,386||…||…||…||…||…|
|Over five years||1,208||1,399||1,734||1,834||2,073||…||…||…||…||…|
|By major currency2|
|With other reporting dealers||46,722||53,622||63,579||72,550||82,190||1,848||2,266||1,872||1,606||2,146|
|With other financial institutions||43,607||53,133||57,564||70,219||86,256||1,845||2,482||1,768||1,707||2,655|
|With nonfinancial customers||11,328||15,044||20,847||21,857||18,894||573||710||687||638||505|
|By remaining maturity3|
|Up to one year||36,938||44,927||46,474||57,157||62,185||…||…||…||…||…|
|One to five years||40,137||46,646||58,914||66,093||76,444||…||…||…||…||…|
|Over five years||24,583||30,226||36,603||41,376||48,711||…||…||…||…||…|
|By major currency|
|Gross credit exposure5||n.a.||n.a.||n.a.||n.a.||n.a.||1,511||1,750||1,969||1,478||2,076|
Counting both currency sides of each foreign exchange transaction means that the currency breakdown sums to twice the aggregate.
Single-currency contracts only.
Adjustments for double counting are estimated.
Gross market values after taking into account legally enforceable bilateral netting agreements.
Counting both currency sides of each foreign exchange transaction means that the currency breakdown sums to twice the aggregate.
Single-currency contracts only.
Adjustments for double counting are estimated.
Gross market values after taking into account legally enforceable bilateral netting agreements.
Since the late 1980s, the major internationally active financial institutions significantly increased the share of their earnings coming from derivatives activities, including from trading fees and proprietary trading profits. These institutions manage portfolios of derivatives involving tens of thousands of positions and daily aggregate global turnover was roughly $2.4 trillion at end-June 2004. The market is an informal network of bilateral counterparty relationships and dynamic, time-varying credit exposures whose size and distribution are intimately tied to important asset markets. Because each derivatives portfolio is composed of positions in a wide variety of markets, the network of credit exposures is inherently complex and difficult to manage. During periods in which financial market conditions stay within historical norms, credit exposures exhibit a predictable level of volatility and risk management systems can, within a tolerable range of uncertainty, assess the riskiness of exposures. Risk management systems guide the rebalancing of the large OTC derivatives portfolios. In normal periods risk management systems can enhance the efficient allocation of risks among firms, but—especially in times of financial stress—they can be a source of trading and price variability that feeds back into the stochastic nature of the cash flows.
Expansions and contractions in the level of OTC derivatives activities are a normal part of modern finance and typically occur in a nondisruptive manner even when volatility or isolated turbulence occurs in one underlying market. The potential for excessively rapid contractions and instability seems to emerge when credit exposures in OTC activities rise to levels that create hypersensitivity to sudden unanticipated changes in market conditions (such as interest rate spreads) and when new information becomes available, as in the emerging markets during the Russian default in July 1998 and later in the mature markets with the collapse of LTCM in September 1998. The creditor and debtor relationships implicit in OTC derivatives transactions between the internationally active financial institutions and their counterparts can create situations in which the possibility of isolated defaults can threaten the access to liquidity of key market participants—similar to a traditional bank run. This can significantly alter perceptions of market conditions, particularly perceptions of the riskiness and potential size of OTC derivatives credit exposures. The rapid unwinding of positions as all counterparties run for liquidity is characterized by creditors demanding payment, selling collateral, and putting on hedges, while debtors draw down capital and liquidate other assets. Until OTC derivatives exposures contract to a sustainable level, markets can remain distressed, giving rise to systemic problems. This is what happened in 1998: after it became known that Russia had defaulted, for example, investors and dealers were concerned that their counterparties were heavily exposed to Russian paper. The induced changes in market conditions quickly created a run for liquidity in seemingly unrelated markets such as the derivatives exchange in Brazil.
Greater asset-price volatility related to the rebalancing of portfolios may be a reasonable price to pay for the efficiency gains from global finance. However, in the 1990s OTC derivatives activities sometimes exhibited an unusual volatility, and added to the historical experience of the extremes that volatility can reach. For example, in the 1990s there were repeated periods of volatility and stress in different asset markets (exchange rate mechanism crises in Europe; bond market turbulence in 1994 and 1996; the Mexican, Asian, and Russian crises; LTCM; Brazil) as market participants searched for higher rates of return in the world’s major bond, equity, foreign exchange, and derivatives markets. Some of these episodes suggest that the structure of market dynamics has been adversely affected by financial innovations and become more unpredictable, if not unstable. (See the IMF’s International Capital Markets reports from 1994 to 2001.)
Examples of extreme market volatility include movements in the yen-dollar rate in both 1995 and 1998. In both cases the yen-dollar exchange rate exhibited extreme price dynamics—beyond what changes in fundamentals would suggest was appropriate—in what was, and is, one of the deepest and most liquid markets. The extreme nature of the price dynamics resulted in part from hedging positions involving the use of OTC derivatives contracts called knockout options (see Box 9.2). Knockout options are designed to insure against relatively small changes in an underlying asset price. Yet, once a certain threshold level of the yen-dollar rate was reached, the bunching of these OTC options drove the yen-dollar rate to extraordinary levels in a very short period—an event that the OTC options were not designed to insure against.
Such episodes of rapid and severe dynamics can also pose risks to systemic stability. In particular, the turbulence surrounding the near-collapse of LTCM in the autumn of 1998 posed the risk of systemic consequences for the international financial system, and seemed to have created consequences for real economic activity (see Box 9.1). This risk was real enough that major central banks reduced interest rates to restore risk-taking to a level supportive of more normal levels of financial intermediation and continued economic growth. LTCM’s trading books were so complicated and its positions so large that the world’s top derivatives traders and risk managers from three major derivatives houses could not determine how to unwind LTCM’s derivatives books rapidly in an orderly fashion without retaining LTCM staff to assist in liquidating the large and complex portfolio of positions.
Derivatives have also featured in market dynamics in emerging-market countries, where markets for pricing and trading derivatives are significantly less well developed. While there is little doubt that emerging-market countries should continue to allow the development of derivatives market activities—in part to improve financial efficiency through price-discovery processes—allowing them to develop too rapidly can lead to unwanted adverse consequences. In particular, while the possibility to hedge the risks associated with emerging-market investments with derivative instruments clearly encourages a higher level of capital flows than would otherwise occur, derivative instruments also create the potential for severe market dynamics when circumstances change and foreign investors want to rebalance portfolios to reduce their exposures. Thus, a balance must be struck between allowing these activities to occur in local domestic markets in emerging-market countries and not allowing them to develop so rapidly that they outdistance the capacity of the country to deal with rapid outflows associated with the unwinding of derivatives positions (Mathieson and others, 2004). (See Box 9.3.)
Both private market participants and those responsible for banking supervision and official market surveillance are learning to adapt to the fast pace of innovation and structural change. This already-challenging learning process is even more difficult because OTC derivatives activities may have changed the nature of systemic risk in ways that are not yet fully understood (Greenspan, 1998; Tietmeyer, 1999; and Geithner, 2004a and 2004b). The heavy reliance on OTC derivatives appears to have created the possibility of systemic financial events that fall outside the more formal clearinghouse structures and official real time gross-payment settlement systems designed to contain and prevent such problems. Heavy reliance on new and even more innovative financial techniques, and the possibility that they may create volatile and extreme dynamics, raises the concern that OTC derivatives could yet produce even greater turbulence with consequences for real economic activity—perhaps reaching the proportions of real economic losses typically associated with financial panics and banking crises.
Box 9.1.Long-Term Capital Management and Turbulence in Global Financial Markets
The turbulent dynamics in global capital markets in late 1998 were preceded by a steady buildup of positions and prices in the mature equity and bond markets during the years and months preceding the Russian crisis in mid-August 1998 and the near collapse of the hedge fund Long-Term Capital Management (LTCM) in September. The bullish conditions in the major financial markets continued through the early summer of 1998, amid earlier warning signs that many advanced country equity markets, not just those in the United States, were reaching record and perhaps unsustainable levels. As early as mid-1997, differences in the cost of borrowing between high- and low-risk borrowers began to narrow to the point where several advanced country central banks sounded warnings that credit spreads were reaching relatively low levels and that lending standards had been relaxed in some countries beyond a reasonable level. A complex network of derivatives counterparty exposures, encompassing a very high degree of leverage, had accumulated in the major markets through late summer 1998. The credit exposures and high degree of leverage both reflected the relatively low margin requirements on OTC derivatives transactions and the increasingly accepted practice of very low, or zero, “haircuts” on repurchase transactions.
Although the weakening of credit standards and complacency with overall risk management had benefited a large number of market participants, including a variety of highly leveraged institutions, LTCM’s reputation for having the best technicians as well as its high profitability during its relatively brief history earned it a particularly highly valued counterparty status. Many of the major internationally active financial institutions actively courted LTCM, seeking to be LTCM’s creditor, trader, and counterparty. By August 1998, and with less than $5 billion of equity capital, LTCM had assembled a trading book that involved nearly 60,000 trades, including on-balance-sheet positions totaling $125 billion and off-balance-sheet positions that included nearly $1 trillion of notional OTC derivatives positions and more than $500 billion more of notional exchange-traded derivatives positions. These large and highly leveraged trading positions spanned most of the major fixed income, securities, and foreign exchange markets, and involved as counterparties many of the financial institutions at the core of global financial markets.
Sentiment weakened generally throughout the summer of 1998 and deteriorated sharply in August when the devaluation and unilateral debt restructuring by Russia sparked a period of turmoil in mature markets that was virtually without precedent in the absence of a major inflationary or economic shock. The crisis in Russia sparked a broadly based reassessment and repricing of risk and large-scale deleveraging and portfolio rebalancing that cut across a range of global financial markets. In September and early October, indications of heightened concern about liquidity and counterparty risk emerged in some of the world’s deepest financial markets.
A key development was the news of difficulties in, and ultimately the near-failure of, LTCM, an important market maker and provider of liquidity in securities markets. LTCM’s size and high leverage made it particularly exposed to the adverse shift in market sentiment following the Russian event. On July 31, 1998, LTCM had $4.1 billion in capital, down from just under $5 billion at the start of the year. During August alone, LTCM lost an additional $1.8 billion, and LTCM approached investors for an injection of capital.
In early September 1998, the possible default or bankruptcy of LTCM was a major concern in financial markets. Market reverberations intensified as major market participants scrambled to shed risk with LTCM and other counterparties, including in the commercial paper market, and to increase the liquidity of their positions. LTCM’s previous preferred creditor status evaporated, its credit lines were withdrawn, and margin calls on the fund accelerated. The major concerns were the consequences—for asset prices and for the health of LTCM’s main counterparties—of having to unwind LTCM’s very large positions as well as how much longer LTCM would be able to meet mounting daily margin calls. As a result, LTCM’s main counterparties, including Bear Stearns, LTCM’s prime brokerage firm, demanded additional collateral. On September 21 Bear Stearns required LTCM to put up additional collateral to cover potential settlement exposures. Default by as early as September 23 was perceived as a real possibility for LTCM in the absence of an injection of capital.
In response to these developments and the rapid deleveraging, market volatility increased sharply, and there were some significant departures from normal pricing relationships among different asset classes. In the U.S. Treasury market, for example, the spread between the yields of “on-the-run” and “off-the-run” treasuries widened from less than 10 basis points to about 15 basis points in the wake of the Russian debt restructuring, and to a peak of over 35 basis points in mid-October, suggesting that investors were placing an unusually large premium on the liquidity of the on-the-run issue. Spreads between yields in the eurodollar market and on U.S. Treasury bills for similar maturities also widened to historically high levels, as did spreads between commercial paper and Treasury bills and those between the fixed leg of fixed-for-floating interest rate swaps and government bond yields, pointing to heightened concerns about counterparty risk. Interest rate swap spreads widened in currencies including the U.S. dollar, deutsche mark, and pound sterling. In the U.K. money markets, the spread of sterling interbank rates over general collateral repurchase rates rose sharply during the fourth quarter, partly owing to concerns about liquidity and counterparty risk (and also reflecting a desire for end-of-year liquidity).
As securities prices fell, market participants with leveraged securities positions sold those and other securities to meet margin calls, adding to the decline in prices. The decline in prices and rise in market volatility also led arbitrageurs and market makers in the securities markets to cut positions and inventories and withdraw from market making, reducing liquidity in securities markets and exacerbating the decline in prices. In this environment, considerable uncertainty about how much an unwinding of positions by LTCM and similar institutions might contribute to selling pressure fed concerns that the cycle of price declines and deleveraging might accelerate.
In response to these developments, central banks in major advanced economies cut official interest rates. In the United States, an initial cut on September 29 failed to significantly calm markets; spreads continued to widen, equity markets fell further, and volatility continued to increase. Against this background, the Federal Reserve followed up on October 15 with a cut in both the federal funds target rate and the discount rate, a key policy action that stemmed and ultimately helped reverse the deteriorating trend in market sentiment. The easing—coming so soon after the first rate cut and outside a regular Federal Open Market Committee meeting (the first such move since April 1994)—sent a clear signal that the U.S. monetary authorities were prepared to move aggressively if needed to ensure the normal functioning of financial markets.
Calm began to return to money and credit markets in mid-October. Money market spreads declined quickly to pre-crisis levels, while credit spreads declined more slowly and remained somewhat above pre-crisis levels, probably reflecting the deleveraging. The Federal Reserve cut both the federal funds target and the discount rate at the Federal Open Market Committee meeting on November 17, noting that although financial market conditions had settled down materially since mid-October, unusual strains remained. Short-term spreads subsequently declined. The calming effect of the rate cuts suggested that the turbulence stemmed primarily from a sudden and sharp increase in pressures on liquidity (broadly defined), including securities market liquidity, triggered by a reassessment of risk.Note: This box draws on the analysis in IMF (1998b; 1999).
In sum, the internationally active financial institutions have increasingly nurtured the ability to profit from OTC derivatives activities and now benefit significantly from them. As a result, OTC derivatives activities play a central role in modern financial intermediation, raising the issue of whether this new, more modern form of banking fragility—that associated with modern finance and OTC derivatives markets—could give rise to systemic problems that potentially could affect national financial markets and, more generally, the international financial system. The remainder of this chapter discusses facets of these markets relevant for assessing their potential to create fragility or threaten financial stability.
Exchange versus OTC Derivatives Markets
Key differences between exchange-traded derivatives and OTC derivatives, including the different trading and risk-management environments, indicate why OTC derivatives activities are both efficiency enhancing and prone to problems. Compared to exchange-traded derivatives, OTC derivatives markets have the following features:
- Management of counterparty risk (credit risk) is decentralized and located within individual institutions.
- There are no formal centralized limits on individual positions, leverage, or margining.
- There are no formal rules for risk and burden sharing.
- There are no formal rules or mechanisms for ensuring market stability and integrity, and for safeguarding the collective interests of market participants.
Broad Similarities, but Important Differences, in Contract Structure
Derivatives offer significant benefits because they facilitate the unbundling and transformation of financial risks such as interest rate risk and currency risk (see the discussion of knockout options in Box 9.2). Individual components of risk can be isolated, individually priced, repackaged, and, if desired, traded. In this way, derivatives allow market participants to tailor more precisely the risk characteristics of financial instruments to their risk preferences and tolerances. By contributing to more complete financial markets, derivatives can improve market liquidity and increase the capacity of the financial system to bear risk and intermediate capital.
Box 9.2.The Role of OTC Currency Options in the Dollar-Yen Market
OTC derivatives activities can exacerbate disturbances in underlying markets—even some of the largest markets, such as foreign exchange markets. Such a disturbance occurred in the dollar-yen market in March 1995 and again in October 1998; once the yen had appreciated beyond a certain level, the cancellation of OTC knockout options and the unwinding of associated hedging positions fueled the momentum toward further appreciation.1 During these periods of heightened exchange rate volatility, OTC derivatives activities also significantly influenced exchange-traded option markets because standard exchange-traded options were used by derivatives dealers as hedging vehicles for OTC currency options.
In 1995, the yen appreciated against the dollar from 101 yen in early January to 80 yen in mid-April, strengthening by 7 percent in four trading sessions between March 2 and March 7. A combination of macroeconomic factors was widely cited as having contributed to the initial exchange rate move. The speed of the move also suggests that technical factors (such as the cancellation of knockout options) and short-term trading conditions (such as the unwinding of yen-carry trades, also involving OTC derivatives) reinforced the trend. In early 1995, relatively large volumes of down-and-out dollar put options were purchased by Japanese exporters to partially hedge the yen value of dollar receivables against a moderate yen appreciation.
In September and October 1998, the yen again appreciated sharply against the dollar from 135 yen to 120 yen per dollar. Of particular interest are the developments during October 6–9, 1998, when the yen strengthened by 15 percent against the dollar. Talk of an additional fiscal stimulus package in Japan and a reassessment of the relative monetary policy stances in Japan and the United States may have sparked the initial rally in the yen and corresponding weakening in the dollar. The initial spate of dollar selling, in turn, was viewed as having created the sentiment that the dollar’s longstanding strengthening against the yen had run its course. However, as in March 1995, in addition to reversals of yen-carry trades, knockout options were widely viewed as having provided additional momentum that boosted demand for yen and contributed to the dollar selling.
Knockout options (a type of OTC barrier option) differ from standard options in that they are canceled if the exchange rate reaches certain knockout levels; they therefore leave the investor unhedged against large exchange rate movements. Nonetheless, knockout options are widely used because they are less expensive than standard options. In 1995 and 1998, knockout options, particularly down-and-out put options on the dollar, amplified exchange rate dynamics through three separate channels:
- As the yen appreciated toward knockout levels, sellers of knockout options had an incentive to try to push the yen up through the knockout levels to eliminate their obligations.
- Japanese exporters who bought knockout options to protect against a moderate depreciation of the dollar sold dollars into a declining market when the knockout options were canceled to prevent further losses on their dollar receivables.
- Dynamic hedging strategies employed by sellers of knockout options required the sudden sale of dollars after the knockout levels had been reached.
Ironically, OTC knockout options that protect only against moderate exchange rate fluctuations can sometimes increase the likelihood of large exchange rate movements—the very event they do not protect against.
Although knockout options represented a relatively small share of total outstanding currency options (between 2 percent and 12 percent), they had a profound effect on the market for standard exchange-traded options. It is easy to see why: knockout options are sometimes hedged by a portfolio of standard options. Dealers who employed this hedging technique needed to buy a huge amount of standard options at the same time as other market participants were trying to contain losses from canceled down-and-out puts. As a consequence, prices of exchange-traded put options (implied volatilities) doubled in March 1995 and almost doubled in October 1998.1See International Monetary Fund (1996, 1998b, Box 3.1) and Malz (1995).
Both exchange-traded and OTC contracts offer these benefits in broadly similar ways. However, exchange-traded contracts have rigid structures compared with OTC derivatives contracts. For example, the Chicago Board of Trade’s treasury bond futures contract dictates
- how many treasury bonds must be delivered on each futures contract;
- the types of treasury bonds acceptable for delivery;
- the way prices are quoted;
- the minimum trade-to-trade price change;
- the months in which contracts may expire;
- how treasury bonds may be delivered from the seller of the contract to the buyer.
Box 9.3.The Role of Derivatives in Crises in Emerging Markets
Derivatives played an important role in the financial crises experienced by several emerging markets. This discussion focuses mainly on two issues: (1) the types of financial derivatives used by market participants before the onset of a crisis and how the use of these instruments affected the stability of the domestic financial system; and (2) the impact of the unwinding of derivative positions on the crisis dynamics. While the Mexican and Asian crises highlighted the role of structured notes and swaps in magnifying balance sheet mismatches and the associated volatility in foreign exchange markets, the Russian and Argentine crises demonstrated the importance of counterparty risk and spillovers through credit markets. (The analysis of the role of derivatives in emerging-market crises is seriously hampered by data availability, because transactions in OTC derivatives are not reported systematically. Thus, in many cases, anecdotal evidence and reported losses on derivatives positions by major investment banks of the industrial countries are the main sources of information.)
The Mexican Crisis, 1994
In the early 1990s, the recently privatized Mexican banks aggressively built up their on- and off-balance-sheet positions, which increased their credit and market risk exposures well beyond prudent limits. In particular, they used various derivatives to achieve leveraged returns. One of the popular instruments that allowed local banks to leverage their holdings of exchange-rate-linked treasury bills (the tesobonos) was a tesobono swap (Garber, 1998). In a tesobono swap, a Mexican bank received the tesobono yield and paid U.S. dollar LIBOR plus a specified number of basis points to an offshore counterparty, which in turn hedged its swap position by purchasing tesobonos in the spot market. The only transactions that were recorded in the balance of payments were an outflow of bank deposits related to the payment of collateral by the Mexican bank, and a U.S. dollar inflow related to the purchase of tesobonos by the foreign investor. Thus, traditional balance of payments accounting provided a misguided representation of capital flows and associated risks—that is, although it appeared that the foreign investor had a long position in government bonds, it was, in fact, the local bank that bore the tesobono risk, while the foreign investor was effectively providing a short-term dollar loan. Tesobono swaps were not the only instruments that allowed local banks to establish leveraged positions financed by short-term U.S. dollar loans from their offshore counterparties; structured notes and equity swaps were also used.1
At the onset of the crisis, the tesobono yields jumped from 8 percent to 24 percent and the U.S. dollar value of the collateral fell, triggering margin calls on Mexican banks. Quoting market sources, Garber (1998) suggested that the total of margin calls on tesobono and total return swaps was about $4 billion (compared to $6.1 billion in foreign exchange reserves of the Banco de Mexico at year end 1994). The continued pressure on the exchange rate forced the authorities to float the peso on December 21, 1994.
The Asian Crises, 1997–98
As in the Mexican crisis, unhedged currency and interest rate exposures were key determinants of the severity and scope of the Asian crises (IMF, 1998a). Banks and nonfinancial corporations in Asia left their exposures unhedged because (1) domestic interest rates were higher than foreign interest rates, (2) the pegged exchange rates were generally perceived as stable, and (3) domestic hedging products were underdeveloped, while offshore hedges were expensive. Because of these factors, foreign banks were eager to lend to East Asian banks that tried to capture carry profits on the interest rate differentials. However, local prudential regulations, such as restrictions on the net open foreign exchange exposures and risk-to-capital ratios, limited the amount of profitable arbitrage trade. Therefore, Asian financial institutions turned to derivatives “to avoid prudential regulations by taking their carry positions off balance sheet” (Dodd, 2001, p. 10).
According to market sources, the majority of losses reported by both U.S. and European banks on their Asian lending were listed as due to swaps contracts, presumably including both total return swaps and currency swaps (Kregel, 1998). In a total return swap, one counterparty pays the other the cash flows (both capital appreciation and interest payments computed on a mark-to-market basis) generated by some underlying asset (equity, bond, or loan) in exchange for dollar LIBOR plus a specified number of basis points.
Thus, the flows between Asian financial institutions and foreign counterparties were similar to those in the Mexican tesobono swap. As in the case of tesobono swaps, offshore counterparties were buying the underlying assets to hedge their swaps positions, while local banks were left with short U.S. dollar positions. When the exchange rate peg collapsed and domestic interest rates rose, both counterparties had incentives to either unwind the swaps or hedge their foreign exchange exposures, which exacerbated the sell-off in Asian assets and currencies.2
Russia’s Default and Devaluation, 1998
Although the poor state of Russia’s fiscal accounts was well known by mid-1998, the announcement of a 90-day moratorium on external debt payments on August 17, 1998, caught most market participants by surprise. At the time of the default, the estimates of the outstanding notionals of the U.S. dollar–ruble nondeliverable forward (NDF) contracts ranged from $10 billion to $100 billion and the total foreign exposure to the domestic bond market (GKOs and OFZs) was around $20 billion. According to market sources, the U.S. dollar–ruble foreign exchange forwards with Russian firms as counterparties were the largest source of credit losses by major swap dealers during 1997–98, exceeding the losses made on their Asian lending. The events in Russia highlighted the presence of convertibility risk even when local currency positions in emerging markets were hedged, and raised the issue of the NDF valuation when an official rate was not available. In addition, Russia’s default sent shock waves through the credit derivatives markets, with the cost of protection increasing in all sectors, including the investment-grade segment. Ambiguous and often misleading definitions of reference obligations, credit events, and settlement mechanics made it very difficult for protection buyers to enforce the contracts. According to dealers, many credit default swap contracts were initially triggered under “failure to pay” clauses, but the attempts to enforce the contracts under such clauses were often frustrated by other credit events that appeared more significant and therefore had to carry more weight under contractual law. To address the legal issues highlighted during the Russian crisis, the International Swaps and Derivatives Association (ISDA) issued new credit derivative documentation guidelines in 1999.3
Argentina’s Default and Devaluation, 2001
In contrast with the Russian crisis, the Argentine default and devaluation in December 2001 were widely anticipated and occurred at a time when the credit derivatives market was relatively more mature. The protracted recession and gradual deterioration of the sovereign’s credit quality gave market participants sufficient time to exit the bond and credit protection markets and also allowed the main sellers of credit protection on Argentine sovereign bonds (broker-dealers) to hedge their books in the repurchase market. According to market sources, liquidity in the Argentine credit default swap (CDS) market dried up in August–September 2001, following a bout of volatility in July. The announcement of the moratorium on all debt payments on December 23, 2001, was unanimously accepted as a “repudiation/moratorium” credit event consistent with the ISDA definitions. Reportedly, some disputes occurred as to which bonds could be considered as “deliverable,” but they were resolved fairly quickly. According to market sources, 95 percent of all CDSs were settled by mid-February 2002 and there were no reported failures to deliver, with the total sum of contingent payments from the protection sellers to the protection buyers estimated at $7 billion (Ranciere, 2002).1Equity swaps are a subset of total return swaps discussed in the Asian Crisis section of this box.2Other structured instruments were also used in the run-up to the Asian crisis. For example, one of the well known instruments was called a PERL—principal exchange rate linked note. A PERL was a dollar-denominated instrument that generated cash flows linked to a long position in an emerging market currency. If the exchange rate remained stable, the return on the PERL was significantly higher than the return on the similarly rated dollar paper, but in the event of major depreciation, the return could become negative (Dodd, 2001).3The most recent (1999) ISDA guidelines include the following types of credit events: “failure to pay,” “obligation acceleration,” “obligation default,” “repudiation/moratorium,” and “restructuring.”Note: The author is grateful to Don Mathieson and Anna Ilyna, who allowed the use and adaptation of this material from Chapter IV of the December 2002 Global Financial Stability Report (IMF, 2002).
Another key difference is that exchange-traded contracts are regulated, often by both a regulatory authority and an exchange’s self-regulatory organization. In the United States, the Securities and Exchange Commission (SEC) regulates exchange-traded derivatives that are legally “securities” (for example, certain options); the Commodity Futures Trading Commission (CFTC) regulates those that are legally “commodities” (for example, financial futures). Regulations promote investor protection because exchange members act as agents for customers; promote market integrity against the potential for manipulation when supplies of underlying goods, securities, or commodities are limited; and promote efficient price discovery, an important function of exchange-traded derivatives (United States, President’s Working Group on Financial Markets, 1999b).
According to market participants, in the exchange environment, regulatory authorities evaluate proposed new contracts in a time-consuming and costly process. By contrast, OTC derivatives contracts can involve any underlying index, maturity, and payoff structure. OTC contracts can fill the gaps where exchange-traded contracts do not exist, including exotic currencies and indexes, customized structures, and maturities that are tailored to other financial transactions. Nonetheless, some OTC derivatives instruments have become “commoditized,” as market conventions and standards have developed over time for payment frequencies, maturities, and underlying indexes. About two-thirds of the gross market value of OTC derivatives is accounted for by simple forwards and swaps, many of which could be traded on an exchange except for minor differences in maturity dates, notional amounts, and underlying indexes (Figure 9.1). In addition, OTC derivatives instruments are lightly and indirectly regulated, often because they fall into regulatory gaps. In the United States, for example, swaps contracts are classified neither as securities nor as commodities, so are regulated neither by the SEC nor the CFTC. Many justifications for regulating exchange-traded derivatives contracts are not relevant for OTC derivatives. As was recognized by U.S. courts (Procter and Gamble v. Bankers Trust), OTC derivatives are principal-to-principal agreements between sophisticated counterparties, and investor protection is not regarded as an important issue. In addition, the risk of manipulation in OTC derivatives markets is minimal, because contracts do not serve a price-discovery role as do exchange-traded derivatives (Greenspan, 2000).
Figure 9.1.Structure of OTC Derivatives Markets, End-December 2004
Source: Bank for International Settlements.
OTC and exchange markets are viewed by market participants as existing in parallel, and OTC contracts are hedged by using standard, exchange-traded derivatives. The major participants who benefit most from OTC derivatives markets envision that exchange-traded derivatives will remain an important part of their risk-management toolbox, and that organized exchange markets will continue to exist alongside OTC markets.
Organization of Markets
Apart from contract flexibility, the most salient differences between OTC and exchange-traded derivatives lie in the organization of trading and the corresponding frameworks for promoting market stability. Trading, clearing and settlement, risk management, and contingency management (handling a clearing-member default, for example) are highly formalized and centralized in exchange markets, but are informal, bilateral, and comparatively decentralized in OTC markets.
Organized exchange markets: centralized, formal, regulated, rule-driven
Organized exchange trading has several standard features:
- membership requirements;
- rules governing conduct (including risk management);
- centralized trading, clearing, and settlement;
- rules that mutualize risk, including loss-sharing in case of defaults.
These features are designed to ensure market integrity, promote efficient price discovery, and safeguard the resources of the clearinghouse. A clearinghouse may be part of the exchange, or a separate legal entity. Exchange members normally commit capital or have an ownership interest in the clearinghouse.
To maintain market stability and financial integrity, exchanges impose soundness, disclosure, transparency, and prudential requirements on members. Typically, there are minimum capital requirements, rules governing protection of customer funds, reporting requirements, and compliance with other rules and regulations. Exchanges closely monitor trading activity to identify large customer positions or concentrations of positions. They also promote transparency by reporting positions, turnover, and price data, and determining settlement prices, usually on a daily basis. Following the collapse of Barings, some clearinghouses share information and assess members’ net exposures across markets (Steinherr, 1998). (See p. 147 for more detail on Barings.)
The clearinghouse manages credit risk and is the central legal counterparty to every transaction; it has a matched market-risk position, but has current credit exposures. Credit risk arises because a change in the price of the underlying asset could cause one counterparty to owe a considerable amount on its position, particularly if the contract is highly leveraged. If an exchange member defaults, the clearinghouse normally has the right to liquidate the member’s positions; take the member’s security deposit, margin, and performance bonds; attach certain other member assets; and invoke any guarantee from the member’s parent company. If the defaulting member’s resources cannot cover the obligation, the exchange can normally turn to the resources of other clearing members by invoking loss-sharing rules. In the event of member default, most clearinghouses transfer the member’s client positions to another member; a few close out the client positions and liquidate the margin. Exchanges also have backup credit lines (Kroszner, 1999). Overall, clearinghouse defaults have been exceedingly rare.
Most important, exchanges formalize risk-management and loss-sharing rules designed to protect the exchange’s capital and the capital of its members. Members are usually, but not always, required to keep speculative positions within strictly defined limits, mark to market at least daily, and post initial and variation margin to limit the exchange’s net credit exposure to the member. Members are subject to surprise inspections and surveys of their financial condition, compliance with exchange rules, and risk-management abilities. Likewise, certain rules protect the exchange and its members from trading activities of nonmembers, which must trade through members. For example, on some exchanges, members of the exchange need not be members of the clearinghouse, but trades must be cleared through clearinghouse members. Exchanges also dictate minimum margin requirements for member exposures to clients (often higher than the requirements for members), as well as client position limits. In addition, clearing members handling clients’ accounts may face more stringent capital requirements compared with those only trading on their own account.
OTC markets: decentralized, informal, lightly supervised and regulated, market-discipline driven
By contrast, OTC derivatives markets lack a formal structure. Because of the lack of membership criteria, counterparties prefer to deal only with highly rated and well-capitalized intermediaries to minimize counterparty risk. OTC derivatives markets are similar to interbank and interdealer markets. They consist of an informal network of bilateral relationships; there is no physical central trading place. Instead, the OTC derivatives markets exist on the collective trading floors of the major financial institutions. No central mechanism limits individual or aggregate risk-taking, leverage, and credit extension, and risk management is completely decentralized. Market participants individually perform risk management, particularly management of the credit risk in the bilateral, principal-to-principal agreements, which is especially challenging because exposures vary with the price of the underlying security and can rise sharply.
The operational aspects of OTC derivatives markets are also decentralized. OTC markets have no centralized trading, clearing, or settlement mechanism. Transparency is generally limited as well. Except for semiannual central bank surveys, market participants do not report outstanding positions or prices for aggregation or dissemination. Information about market concentration and who owns which risks is generally unavailable; at best, a trading desk might know that some institutions are building up positions. This lack of transparency enabled LTCM to build up outsized positions during 1997 and 1998 (IMF, 1999).
OTC instruments and trading are essentially unregulated, although they are affected indirectly by national legal systems, regulations, banking supervision, and market surveillance. None of the major financial centers has an “OTC derivatives regulator” similar to a banking or a securities regulator.92 Market participants create instruments to minimize regulatory burdens (including capital requirements), and in many jurisdictions, supervisory and regulatory frameworks geared toward traditional banking and securities activities impinge only indirectly on OTC derivatives markets. Institutional coverage is not comprehensive either, because while banks are regulated and heavily supervised, hedge funds and certain securities affiliates are not regulated. Because financial activities evolve more rapidly than official oversight, the gap between regulator and regulated seems to have widened. Official surveillance of these markets also is limited. Overall, the supervision of financial institutions (including of brokers and dealers) and market surveillance play a critical but limited role in ensuring the smooth functioning of OTC derivatives markets, primarily by seeking to ensure the overall soundness of the institutions that make up those markets.
Regulations are also highly fragmented, both nationally and internationally. In the United States, for example, there are at least three groups of regulators—securities, commodity futures, and banking—impinging, however slightly, on OTC derivatives activities. In addition, while the major market-making institutions flexibly book trades around the globe, supervision and regulation are nationally oriented. Over time, efforts have been made to adapt the current framework, including through the 1995 amendment to the Basel Accord on Capital Adequacy.93 Authorities acknowledge that significant gaps in coverage remain and new gaps will likely emerge between market practices and official frameworks.
Despite its limited role, the current regulatory framework has had a visible impact on the market. Existing regulation and concerns about possible regulation have influenced the choice of jurisdictions where trading takes place; the type of legal structure (including unregulated subsidiaries) used to handle dealer activities; the structure of trading, clearing, and settlement (including the degree of centralization and automation); and contract design. These choices reflect efforts to minimize or eliminate the impact of regulations (capital requirements, for example) and also reflect the effects of regulatory uncertainty, including whether regulators might construe types of OTC derivatives as falling under their purview and hence being subject to, for example, more burdensome disclosure and capital requirements.
This light regulation and supervision exists alongside a set of private mechanisms that facilitate smoothly functioning OTC derivatives markets. Market discipline, provided by shareholders and creditors, promotes market stability by rewarding financial institutions based on their performance and creditworthiness. Recent research finds market discipline to be strong only during periods of banking sector stress and volatile financial markets (Covitz, Hancock, and Kwast, 2000).
Market discipline is present when a firm’s private sector financial stakeholders (shareholders, creditors, and counterparties) are at risk of financial loss from the firm’s decisions and can take actions to “discipline” the firm and to influence its behavior. Market discipline may operate through share price movements, by constraints to the supply of credit, or through the willingness to do business through counterparty relationships. Market discipline in financial markets therefore rests on two key elements: investors’ ability to accurately assess a firm’s financial condition (monitoring) and the responsiveness of the firm’s management to investor feedback (influence). (See United States, Board of Governors of the Federal Reserve System, 1999.) Institutions mark their trading books to market daily so that unprofitable decisions and poor risk management can be reflected immediately in measured performance (profits and losses). This informs senior management and, through disclosure, financial stakeholders. These mechanisms have some influence, as demonstrated during the turbulence in 1998 when those institutions that appeared to manage well enjoyed the most buoyant stock prices, and creditors of institutions perceived to be less creditworthy refused to roll over credit lines or bond issues, and sold their credit instruments in the secondary market. The subsequent reductions in proprietary trading activity seem to have been largely motivated by financial stakeholders’ desires for less risky earnings.
In OTC derivatives markets, special obstacles to effective market discipline (both monitoring and influence) tend to be related to information disclosure—one of the fundamental preconditions for effective market discipline. For example, the off-balance-sheet character of derivatives makes it difficult for outside financial stakeholders to evaluate the financial health of an institution and its contingent liabilities. Data on individual exposures is proprietary, and disclosure could diminish potential profits. In addition, competitive pressures and the desire to generate order flows can lead creditors to extend credit without insisting on adequate counterparty disclosure, as occurred, for example, with LTCM. Therefore, more emphasis may have to be placed on counterparty monitoring, because the application of broader market discipline for complex institutions active in the OTC derivatives markets may have significant limitations.
As a supplement to these mechanisms, a number of industry groups are involved in initiatives designed to support well-functioning OTC derivatives markets, notably the International Swaps and Derivatives Association (ISDA), the Counterparty Risk Management Policy Group II (reconstituted in February 2005 after completing its initial work in 1999), the Group of 30, and the Derivatives Policy Group.94 Efforts include dissemination of best practices in risk management, standardization of documentation, identification of gaps in risk-management practices, and flaws in the operational infrastructure, assessments of legal and other operational risks, efforts to foster interindustry and public-private dialogues on key issues, and initiatives to voluntarily disclose information to regulatory authorities. The activities of these groups reflect the fact that market participants see it as in their best interests to encourage an orderly, effective, and efficient market—and to discourage regulation.
Corporate governance monitoring by financial stakeholders and other private initiatives impose discipline on OTC derivatives activities and increase incentives to reflect the degree of counterparty risk in pricing, margins, or collateral. Monitoring also creates benchmarks against which participants, end users, and regulators can measure progress in dealing with the issues raised in public and private forums. Paradoxically, some of the same factors that complicate market discipline (such as the opacity of OTC derivatives) are also the very factors that make market discipline desirable from the standpoint of financial regulators.
Sources of Volatility and Potential Fragility in OTC Derivatives Activities and Markets
As noted in the previous sections, some of the features of OTC derivatives contracts and markets that provide benefits and enhance efficiency either separately or jointly embody risks to financial market stability. OTC derivatives activities are governed almost exclusively by decentralized private infrastructures (including risk management and control systems, private netting arrangements, and closeout procedures) and market-disciplining mechanisms. By comparison, the more formal centralized rules of exchanges protect the stability and financial integrity of the exchange. In addition, the major financial intermediaries in OTC derivatives markets have access to financial safety nets. Because this can affect their behavior, they are required to adhere to prudential regulations and standards in the form of minimum risk-adjusted capital requirements and accounting and disclosure standards that inform financial stakeholders and, to some extent, support market discipline. The financial industry also has its own standards and best practices, which are promulgated by various industry groups.
Private, decentralized mechanisms have so far safeguarded the soundness of the internationally active financial institutions, in part because many of them have been well capitalized. However, these mechanisms have not adequately protected market stability, and markets and countries only remotely related to derivatives activities experienced instability because of spillovers and contagion. For example, while no major institution failed during the mature market turbulence of 1998 surrounding the near-collapse of LTCM, private, decentralized market-disciplining mechanisms did not prevent the buildup and concentration of counterparty risk exposures within the internationally active financial institutions.
The features of OTC derivatives markets that can give rise to instability in institutions, markets, and the international financial system include
- the dynamic nature of gross credit exposures;
- information asymmetries;
- the effects of OTC derivatives activities on available aggregate credit;
- the high concentration of OTC derivatives activities in the major institutions;
- the central role of OTC derivatives markets in the global financial system.
The first underlying source of market instability is the dynamic nature of gross credit exposures, which are sensitive to changes in information about counterparties and asset prices. This feature played an important role in most of the crises in the 1990s. A disruption that sharply raises credit exposures has the capacity to cause sudden and extreme liquidity demands (to meet margin calls, for example). Just as traditional banks were not always prepared for sudden, abnormally large liquidity demands and withdrawals of deposits during bank runs, today’s derivatives market participants may not be prepared for sudden and abnormally large demands for cash that can and do arise in periods of market stress.
A second, well known and related source of market instability is information asymmetries, as in traditional banking (Diamond, 1984). Not having sufficient information on borrowers complicates the assessment of counterparty risks. This problem is exaggerated for the credit exposures associated with OTC instruments because of the price-dependent, time-varying nature of these credit exposures. A counterparty’s risk profile can change quickly in OTC derivatives markets. As a result, information asymmetries in OTC derivatives markets can be more destabilizing than in traditional banking markets because they can quickly lead intermediaries and market makers to radically scale back exposures, risk-taking, and the amount of capital committed to intermediary and market-making functions.
Third, OTC derivatives activities contribute to the aggregate amount of credit available for financing, and to market liquidity in underlying asset markets. The capacity for the internationally active institutions to expand and contract off-balance-sheet credit depends on the amount of capital they jointly devote to intermediation and market making in derivatives markets. This capital can support more or less activity depending on several factors, including the risk tolerances (amount of leveraging) of the intermediaries and market makers; the underlying cost of internal capital or external financing; and financial-sector policies (for example, capital requirements). A determinant of the cost of capital for OTC derivatives activities is the risk-free interest rate (such as on 10-year U.S. treasury bonds), which is also used for pricing contracts. When underlying financing conditions become favorable, the OTC-intermediation activities can become more profitable and more cheaply funded and the level of activity can expand relative to the base of equity capital in the financial system. This tendency for expansion (and, when conditions change, contraction) can become self-generating, and it can, and has, occasionally become hypersensitive to changes in market conditions.
Fourth, as noted, aggregate OTC derivatives activities are sizable and the trading activity ($2.4 trillion daily turnover) and counterparty exposures are highly concentrated in the internationally active financial institutions. This makes those institutions and the global markets susceptible to a range of shocks and dynamics that impinge on one or more major counterparties. The reason for this concentration is clear. Profitability requires large-scale investments in information technologies (such as sophisticated risk-management systems) and also requires a broad client base and the ability to deal in a wide variety of related cash products. Only the largest organizations with global reach and international networks of clients and distribution channels can effectively compete as the central players in OTC markets. As a result, intermediation and market making are performed by global institutions, which hold and manage the attendant risks through hedging and trading, among other mechanisms. The major intermediaries have access to financial safety nets, which may impart an element of subsidy in the pricing of credit and other risks, possibly contributing to an overextension of credit. This concentration makes OTC derivatives markets and the institutions trading in them potentially vulnerable to sudden changes in market prices for underlying assets (interest rates and exchange rates, for example) and in the general market appetite for risk.
Fifth, OTC derivatives activities closely link institutions, markets, and financial centers, making them possible vehicles for spillovers and contagion. About half of OTC derivatives trading in the largest segments takes place across national borders. Links arise from the contracts themselves (currency swaps mobilize liquidity across the major international financial centers) and through the international institutions that make up these markets. In addition, hedging, pricing, and arbitrage activities link OTC derivatives markets to the major cash and exchange-traded derivatives markets: for example, hedging and arbitrage activities link the market for interest rate swaps and the markets for bonds, interest rate and bond futures, and interest rate options. The interconnections and the opportunities for arbitrage the interconnections provide add to the efficiency and complexity of the international financial system. At the same time, links also mean that disruptions in OTC activities necessarily result in spillovers and contagion to these other markets.
To summarize, certain features of OTC derivatives and how they are traded and managed make OTC derivatives markets subject to instability if the wrong combination of circumstances arises. This instability occurs, in part, because OTC derivatives markets are centered around the internationally active financial institutions that each are counterparty to tens of thousands of bilateral, price-dependent, dynamic, credit exposures embodied in OTC derivatives contracts. OTC derivatives contracts bind institutions together in an opaque network of credit exposures, the size and characteristics of which can change rapidly and, moreover, are arguably not fully understood with a high degree of accuracy even by market participants themselves. These institutions allocate specific amounts of capital to support their perceived current and potential future credit exposures in their OTC derivatives business. However, risk assessments and management of these exposures are seriously complicated by a lack of solid information and analyses about the riskiness of both their own positions and those of their counterparties. As a result, this market is characterized by informational imperfections about current and potential future credit exposures and marketwide financial conditions.
The potential for instability arises when information shocks, especially counterparty credit events and sharp movements in asset prices that underlie derivative contracts, cause significant changes in perceptions of current and potential future credit exposures. Changes in perceptions, in turn, can cause dramatic movements in derivatives positions of the major participants. When asset prices adjust rapidly, the size and configuration of counterparty exposures can become unsustainably large and provoke a rapid unwinding of positions. The experience of the late 1990s and early 2000s strongly suggests that the ebb and flow of credit exposures among the large internationally active financial institutions can be severely affected by events that cannot be easily predicted and that can lead to potentially disruptive systemic consequences.
Weaknesses in the Infrastructure
Certain aspects of the infrastructure for OTC derivatives activities can also lead to a breakdown in the effectiveness of market discipline and ultimately produce unsustainable market conditions and affect market dynamics, including producing or exacerbating underlying instabilities:
- inadequate counterparty risk management;
- limited understanding of market dynamics and liquidity risk;
- legal and regulatory uncertainty.
Each of these areas can be improved through the efforts, separately or jointly, of financial institutions, supervisors, and those responsible for market surveillance.
Inadequate Counterparty Risk Management
One of the most important lessons of the near collapse of LTCM and the market turbulence that followed is that counterparty credit risk management by the major global financial institutions was inadequate.95 Counterparty risk is now widely understood to be of primary importance for managing risk in OTC derivatives markets, and it is still actively under discussion.96 While there have been some changes, only limited progress has been made in improving the management of credit risks associated with OTC derivatives. Progress has been particularly slow in developing techniques for managing the interactions of credit, market, and liquidity risks, the latter being particularly important and difficult to deal with. Even less well understood are the interactions with operational risk and legal risk.
Several factors explain this limited progress. First, while counterparty disclosure has improved, it has not done so significantly since 1998. The leading providers of intermediation and market-making services in OTC derivatives markets still have serious concerns about the dearth of information supplied by clients, including hedge funds.97 Accordingly, regulators and supervisors have similar concerns, in part because it affects the ability of the major OTC dealers to provide effective and timely market discipline, especially when competition increases. Second, the conceptual and measurement challenges involved in understanding counterparty risk and other risks are unlikely to be resolved soon. Even sophisticated institutions acknowledge that significant additional progress is necessary.
Consider what happened during the LTCM collapse, which exposed many sources of potential weaknesses that are still relevant (in 2005) and will remain so in the future.98 Widespread problems with assumed counterparty risk assessments and pricing produced turbulence in OTC derivatives markets, in part because incentives for prudent risk-taking proved to be insufficient to prevent the buildup and concentration of counterparty risk exposures in the autumn of 1998. After the turbulence, however, some of these same incentives worked better, including the discipline from losses in shareholder value and the associated lower bonuses for managers, and the discipline imposed by senior management in determining the risk culture, in setting risk tolerances, and in implementing risk management and control systems. Thus, the LTCM affair appears to have taught some valuable lessons.
If information to assess creditworthiness is insufficient, collateral is generally a reasonable counterparty risk mitigation technique. However, the assets held as collateral are subject to market risk and their value can decline precipitously when the protection they offer is most needed, namely, during periods of turbulence when the probability of counterparty default can rise significantly. This risk may not have been adequately accounted for in the management of OTC derivatives trading books. In the aftermath of 1998, institutions acknowledged the inadequacies in collateral management and uncertainties about legal claims on collateral. Both contributed to market turbulence in the 1990s by encouraging financial institutions to liquidate collateral into declining markets. In addition, in the runup to the turbulence of the autumn of 1998, counterparties tended to demand low or no haircuts on collateral, because of competitive pressures and the relatively low cost of funding at that time. These measures could have offered protection against declines in collateral values and helped to reduce pressures to liquidate collateral into declining markets.
Limited Understanding of Market Dynamics and Liquidity Risks
As noted in Chapter 8, market participants and officials acknowledge their limited understanding of market dynamics in OTC derivatives markets and the implications of those dynamics for related markets. Views diverge on whether OTC markets absorb financial shocks or whether they amplify shocks and contribute to volatility. Some believe derivatives markets dissipate shocks by facilitating hedging, while others see these markets as a channel of contagion. Market participants also disagree about how OTC derivatives markets affect the distribution and mix of credit, market, liquidity, operational, and legal risks. One view is that they redistribute risks to those most willing to hold them. Another is that they transform risks in ways that are inherently more difficult to manage because, while reducing market risk, they create credit, operational, and legal risk. Views on relationships between liquidity in derivatives, secondary, and money markets vary considerably. Finally, there is widespread uncertainty about how monetary conditions influence prices and liquidity in OTC derivatives markets.
Market participants acknowledge their previous failures to realize the importance of liquidity risk in OTC derivatives, and that the capacity to manage it is still in an embryonic stage. One common mistake in 1998 was that risk management systems assumed markets would remain liquid and price changes would follow historical norms. Risk managers also failed to engage in stress testing to examine the implications of severe liquidity problems. Few firms were, for the purposes of risk management, marking credit exposures to estimated liquidation values instead of to current market values. Even these few firms seemed to rely on stress tests that did not fully capture the dynamics revealed in 1998. These challenges are still under active consideration as of mid-2005.
Marking positions to liquidation values is likely to become standard practice at sophisticated financial institutions. However, liquidation values may not be uniquely determined because asset prices behave in nonlinear ways at stress points. Thus, even sophisticated institutions will make modeling errors. Less sophisticated firms may rely on margining requirements and haircuts. These, too, have their limitations in times of stress. Reliance on margin calls to limit counterparty credit risk, normally an effective risk management tool, can contribute to liquidity pressures in apparently unrelated markets and can raise the likelihood of default by financial institutions that would be solvent under normal market conditions. Likewise, overreliance on value-at-risk and mark-to-market accounting and other rules that encourage frequent portfolio rebalancing can induce large-scale selling of positions (Schinasi and Smith, 2000).
To address the challenges posed by liquidity risks and market dynamics, sophisticated institutions are focusing on the total risk they face rather than on the individual risks (market, credit, liquidity, operational, and legal). Particularly challenging is the link between liquidity and counterparty risk, which may depend on the underlying trading, risk-mitigation, and legal infrastructure. Liquidity risk can become closely linked to credit risk, because a loss of liquidity can depress market prices and increase the credit exposure on OTC derivatives. Conversely, heightened concerns about counterparty credit risk can precipitate a loss of liquidity by causing market participants to pull back from markets. International financial institutions recognize the need to incorporate such links into risk management systems, and the formidable challenges of measuring and modeling them. The market turbulence in September 1998 may have been the first event that revealed the importance of these links. Seven years later, because of the difficulty of understanding and modeling these links, institutions may still lack sufficient experience to reliably incorporate these links into their stress tests. Future improvements in the management of total risk should contribute to the smooth functioning of OTC derivatives markets.
Legal and Regulatory Uncertainties
Another important source of weakness in the financial infrastructure is legal and regulatory uncertainty. This type of uncertainty encompasses the possibility that private arrangements to mitigate risks (such as definitions of default and legality of closeout and netting arrangements) may turn out to be ineffective. To the extent that risk mitigation fails to work as designed, misperceptions, mispricing, and misallocation of financial risk can result. Legal and regulatory uncertainties can also be important sources of liquidity risk, because they can contribute to adverse market dynamics.
Cumbersome closeout procedures and uncertain enforcement of security interests in collateral can be impractical and ineffective in protecting firms against default. Such concerns contributed to the rapid liquidation of collateral in the autumn of 1998, and of credit-default swaps during the U.S. recession and corporate governance irregularities during 2001 (see discussion in Chapter 10). However, closeout procedures for some contracts are as legally uncertain in 2005 as they were then. The uncertainty arises because of important differences in bankruptcy laws among countries. Specifically, a number of countries do not allow the termination of contracts upon the initiation of insolvency proceedings, thus giving the trustee the opportunity to continue those contracts favorable to the estate (known as cherry picking). Moreover, even among countries that allow for the termination of contracts, some do not allow for the automatic set-off of contractual claims, which is necessary for netting and closeout. An increasing number of national bankruptcy laws are allowing for stays on the enforcement of security interests in collateral. In these cases, the law will often provide that the interests of secured creditors must be protected during the stay (for example, by compensating for the depreciation of the value of the collateral). One uncertainty arising in many countries is whether such protection will be provided and, if so, whether it will be adequate.
When market participants cannot close out positions or reclaim collateral as specified in private contracts, collateral does not give the expected protection against credit risk. Once a counterparty realizes that protection is absent, credit risk can quickly cross a threshold and be perceived as a default event. With this kind of uncertainty, firms holding collateral with creditor-stay exemptions (which allow counterparties to close out exempt OTC derivatives transactions outside of bankruptcy procedures) have the incentive to exercise their legal right to sell collateral. Closeout valuations require three to five market quotes per contract, and a derivatives desk may have thousands of contracts with a single counterparty. A dealer attempting to close out the number of swaps with LTCM in 1998 might have had to collect 16,000 market quotes from other dealers at a time of market stress when every other major desk was attempting to do the same. Alternative valuation procedures, including good-faith estimates, internal valuations, or replacement value, would be an improvement; this possibility is still under discussion nearly seven years after the near-collapse of LTCM.
Widely used netting agreements (such as the ISDA master agreement) have limitations in mitigating risk. Netting arrangements can reduce the credit exposures on a large number of transactions between two counterparties to a single net figure. Thus, netting arrangements are a risk-mitigating technique with significant potential to reduce large gross credit exposures. If netting cannot be relied upon as legally enforceable, the hint of default can trigger the unwinding of gross exposures. The failure to recognize this possibility may be a source of misperceptions of risk in certain contracts and transactions. Several initiatives were introduced in the aftermath of the turbulence in the late 1990s and have helped to facilitate bilateral and multilateral netting, but they were for specific instruments (RepoClear, for example). Because finance is innovative and new instruments are created continually, market participants and officials will no doubt face related challenges from time to time. An example that has become increasingly relevant is the desirability of introducing netting arrangements for exposures associated with credit derivatives, for which there are a host of still unsettled legal-enforceability concerns (as discussed in Chapter 10).
The various legal and regulatory environments in which OTC derivatives transactions are conducted present uncertainties, too, owing to the high pace of innovation, the relatively limited extent of legal precedent, the cross-border nature of OTC derivatives markets, and the supervisory and regulatory framework. Legal risks include the possibility that a counterparty may walk away from obligations, or may cherry pick; it may dispute the terms of an agreement; it may claim that it did not understand the agreement; and it may claim that it did not have the authority to enter into the agreement.
In the United States, whether certain types of swaps are subject to CFTC approval and oversight was legally ambiguous (see Box 9.4 for reasons) until recently, when new legislation was introduced. This legal uncertainty contributed to reluctance to standardize swap contracts and to centralize clearing (Folkerts-Landau and Steinherr, 1994). Some market participants believed in the late 1990s that bankruptcy procedures needed to be modernized to strengthen the legal certainty of risk-mitigation methods and the definitions of what constitutes a default, which is particularly relevant for the development of credit derivatives. For example, some saw the need to extend creditor-stay exemptions under U.S. bankruptcy law beyond swaps and repurchase transactions to other OTC derivatives contracts. Legislation is still pending in the U.S. Congress that would strengthen and clarify the enforceability of early termination and closeout netting provisions and related collateral arrangements in U.S. insolvency proceedings and is expected to be considered in mid-2005 (Cunningham and Cohn, 2005).
Many jurisdictions, other than the United States and the United Kingdom, are ill-suited for effective modern risk management. For example, collateral may afford limited protection in bankruptcy (unless the collateral is held in the United States or the United Kingdom). Legal staffs at major dealer and market-making institutions see significant legal uncertainties associated with the use of collateral in advanced countries (Canada, Italy, and Japan). While the legal and regulatory environments for OTC derivatives are complex in the United States and the United Kingdom they are considerably more complicated elsewhere. The same instrument might be legally defined as a swap transaction in one country, an insurance contract in a second country, and a pari-mutuel betting instrument in a third country. Market participants are making strong efforts to mitigate the legal risks, but the private sector can accomplish only so much because contracts must ultimately be enforceable in a legal system.
Strengthening the Stability of Modern Banking and OTC Derivatives Markets
Market participants and officials acknowledge that issues in OTC derivatives markets need to be dealt with, and proposals and initiatives have been advanced as a result of experiences in the late 1990s and the early 2000s. Some progress has already been made, and the lessons of recent experience are likely to motivate further actions. However, the available evidence suggests that many recognized problems have yet to be adequately addressed. Insufficient progress has been made in implementing reforms in risk management, including counterparty, liquidity, and operational risks.99 Relatively less attention has been focused on removing legal and regulatory uncertainty. Given the limited progress to date, the implementation of further reforms is essential (Geithner, 2004a; 2004b).
Box 9.4.Sources of Legal Uncertainty in the U.S. Regulatory Environment
In the United States, legal uncertainties arise from concerns about (1) whether some OTC swap contracts (primarily those that are standardized) could be construed to be futures contracts and would thus be subject to the Commodity Exchange Act (CEA) and (2) whether certain types of mechanisms for executing and clearing OTC derivatives transactions could alter the status of otherwise exempted or excluded swaps. There are also ambiguities about which securities-based derivatives fall under the jurisdiction of the Securities and Exchange Commission (SEC) or Commodity Futures Trading Commission (CFTC), or may, in fact, be prohibited.
Uncertainties about the standing of swap agreements emerged in connection with the CFTC’s Swap Exemption. The Futures Trading Practices Act of 1992 granted the CFTC authority to exempt certain instruments from the CEA (and from the requirement to trade on an exchange). In 1993, the CFTC issued the Swap Exemption, which excludes any swap agreement that meets certain criteria from the CEA. These criteria restrict the design and execution of transactions and are meant to prevent unregulated exchange-like markets for swaps. To qualify for the exemption, a swap (1) must be concluded between eligible swap participants; (2) cannot be standardized as to the material economic terms; (3) cannot be part of a central clearing arrangement; and (4) cannot be traded through a multilateral transaction execution facility.1 Uncertainties in the interpretation of these conditions have, however, emerged. The rise of electronic trading has blurred the line between bilateral and multilateral trading, and the advantage of centralized clearing systems has become widely recognized as trading volumes have increased and a wider range of users have entered the market.2 As a result, the limits of the swap exemption have come to be viewed as impediments to further development of the swaps market and in particular seem to be inhibiting the introduction of electronic trading platforms and clearing arrangements to mitigate risks.
Regulatory uncertainties may have restricted the types of OTC derivatives contracts that are written. Ambiguities about the extent of CFTC or SEC jurisdiction to regulate certain securities-based derivatives, such as equity swaps, credit swaps, and emerging country debt swaps, are largely the legacy of the 1974 amendment to the CEA that gave the CFTC exclusive jurisdiction over all futures (on physical and financial commodities) without superseding or limiting the jurisdiction of the SEC. However, the broad definition of “commodity” in the CEA raised concerns that OTC markets for government securities and foreign currency would have been covered by the Act. Therefore—upon the Treasury’s request—an amendment (the Treasury Amendment of 1974) was inserted into the Act that excluded from it, among other things, transactions in foreign currency, government securities, and mortgages, “unless such transactions involve the sale thereof for future delivery conducted on a board of trade.” However, these amendments did not eliminate conflicts regarding each agency’s jurisdiction. Ambiguities and potential overlaps of CFTC and SEC jurisdictions remained, in particular, over novel financial instruments that have elements of securities and futures or commodity option contracts. Therefore, the Shad-Johnson Accord between the SEC and the CFTC was concluded in 1983; it explicitly prohibits futures contracts based on the value of an individual security (other than certain exempt securities).3 The Shad-Johnson Accord itself created some uncertainty, particularly about the status of swap agreements that reference “non-exempt securities,” such as equity swaps, credit swaps, and emerging market debt swaps.1For a detailed list of these swap conditions, see United States, President’s Working Group on Financial Markets (1999).2Folkerts-Landau and Steinherr, 1994; United States, President’s Working Group on Financial Markets, 1999.3Exempt securities include government securities and other securities that are exempt from many of the federal securities laws.
Balancing Private and Official Roles
Many of the instabilities identified above result from three areas of imperfection: (1) market discipline; (2) risk-mitigating infrastructures; and (3) official rule making and oversight. Elements of all three failed to prevent the buildup and concentration of counterparty exposures in 1998 and the more general governance lapses in 2001–2004. Strengthening market stability requires improvements in each of these three areas, but consideration should also be given to altering the balance of the roles of the private and public sectors in ensuring market stability, in particular in tilting the balance toward greater reliance on effective market discipline. It is in the general public’s interest to have these markets function as smoothly as possible most, if not all, of the time. More generally, what is the appropriate balance of market discipline on the one hand, and official oversight on the other, for ensuring the smooth functioning of OTC derivatives markets?
In attempting to strike this balance several factors are relevant. The authorities in the mature markets, primarily through Group of Ten efforts, have collectively adopted an approach that places as heavy a reliance on market discipline as is feasible, while recognizing the limits to private discipline, including those emanating from moral hazard, information asymmetries, and other externalities. The desirable degree of official involvement evokes less agreement. Nevertheless, a strong case can be made for relying more heavily on market disciplining mechanisms, provided they can be made more effective. There may also be areas of complementarities and scope for constructive engagement. One such area is disclosure; more voluntary and some involuntary disclosure might greatly improve the effectiveness of risk management and market discipline through greater financial stakeholder awareness. However, only the right kind of disclosure would improve matters, and the international community is not clear about the optimal kind or frequency of information. There are also trade-offs. For example, more official oversight or regulation, by creating the impression that officials are monitoring, can create moral hazard by diminishing private stakeholder incentives to monitor and influence business decisions and reduce management incentives for risk-taking. Striking the right balance needs to take these interrelated effects into account.
If market discipline should carry the heaviest load, the limits (natural or otherwise) to what the private sector can achieve on its own must be identified more precisely against the background of existing rules of the game (supervisory and regulatory frameworks, including financial safety nets). An example of such a limitation is the coordination failure that apparently occurred in organizing the private rescue of LTCM. By some accounts, in the days before it became clear that LTCM might default on some of its contracts, several large institutions apparently tried to organize a group of institutions to take over the hedge fund. While some were willing to put up substantial amounts of capital, in the end it was insufficient. Moreover, several institutions with financial interests nevertheless decided they would not be a party to such a partnership. This example of the free-rider problem represents a limit to the ability of the private sector to ensure the smooth functioning of markets—by coordinating private solutions—in the presence of market stress.
Reductions in information asymmetries may also be limited. LTCM was widely viewed as a large source of both trading revenues and information in 1996 and 1997. Each creditor institution essentially formulated an investment and trading strategy with LTCM that seemed desirable, given the limited information they had. In effect, institutions were involved in a dynamic game with LTCM and within the OTC derivatives markets. The institutions provided financing for LTCM’s trades in return for trading activity and a window on LTCM’s order flow and investment strategy. Although it is easy in retrospect to question why LTCM’s counterparts did not demand more information, in a competitive environment, cost considerations must have weighed heavily. Clearly, LTCM’s counterparties thought the cost of more information was too high, and walking away from deals was not in their interests. Moreover, they all thought they were receiving useful information from LTCM’s orders for trades.
Thus, situations can arise in which institutions in pursuit of self-interest can collectively produce market conditions that become unsustainable and harmful to them individually and collectively. That is, in the absence of a central, coordinating mechanism that enforces collective self-interest in market stability (such as on an exchange), individually desirable strategies, when aggregated, can produce bad market outcomes (the Prisoner’s Dilemma of Chapter 3). Perhaps private information sharing and coordination could have made the LTCM game end without a severe disruption, but so could more effective official refereeing. The challenge is to have a framework that more effectively prevents these situations from arising, thus the responsibilities for strengthening areas with potential instabilities need to be assigned.
Strengthening Incentives for More Effective Market Discipline
In some cases, the entity that should be responsible for strengthening a vulnerable area is obvious. Clearly, private financial institutions are responsible for managing individual private risks, within the regulatory and supervisory framework. Well-known improvements (as discussed above and documented in several reports issued since the LTCM crisis and more recent Enron problems in commodity derivatives) can be made in risk-management and control systems to enhance the likelihood that institutions will remain well capitalized and profitable and thereby help to avoid instability, even in times of stress. The fact that market participants have not moved as quickly as might have been expected to improve risk management systems—given the virulence of the turbulence in the autumn of 1998—suggests that designing and implementing new systems to deal with the complex and evolving risks involved in OTC derivatives is a difficult challenge.
Moral hazard, perhaps associated with national histories of market interventions, may be another factor hampering the effectiveness of market discipline. The risk to financial stability arising from banks’ OTC derivatives activities may also be influenced by access to financial safety nets, which, by imparting a subsidy element, can influence the pricing of risk and thereby lead to overextensions of credit both on and off balance sheet. Access to safety nets (including central bank financing) can give rise to incentives to take additional risks that can lead to the buildup of large, leveraged exposures that, when suddenly unwound, can precipitate a financial crisis of systemic proportions. Moreover, interventions during one stressful episode that limit losses can sow the seeds of the next buildup of exposures. These influences may have dampened the strong signal that institutions might have received from the turbulence that followed the near collapse of LTCM.
Official sector incentives encouraging the private sector to improve its ability to monitor itself, and to improve the effectiveness of market discipline, may now be required. As emphasized in IMF (1999), one way of improving the ability of private incentives to effectively discipline behavior is for the private and public sectors to jointly identify possible inconsistencies arising from the complex interplay of both private and regulatory incentives as they affect private decisions. Inconsistencies between private and regulatory incentives—for example, inconsistencies between internal models for allocating capital and regulatory capital requirements—could thus be rectified to alter behavior in ways that preserve efficiency and promote market stability.
Reducing Legal and Regulatory Uncertainty
There also seems to be an obvious assignment of responsibilities in the area of legal and regulatory uncertainty. The official sector and national legislatures can reduce legal and regulatory uncertainty. Legal or regulatory uncertainties that can be clearly identified should be addressed as soon as possible. Three areas immediately come to mind: the regulatory treatment of swaps and the implications for using private clearinghouses; closeout procedures; and netting. In each of these cases, reducing uncertainty could have the adverse consequence of actually increasing risk-taking. To ensure that measures to reduce legal and regulatory uncertainty actually strengthen financial stability, the measures should be linked to measures to address those features of OTC derivatives, institutions, and markets that most clearly pose risks to market stability. For example, legal certainty of closeout and netting would implicitly provide OTC derivatives creditors seniority over general creditors if a counterparty defaults, possibly giving rise to incentives to engage in riskier activities. To counteract such incentives, the extent of legally sanctioned closeout of contracts and permitted netting of exposures could be made contingent on key structural reforms that enhance stability. In this example, trading arrangements along the lines of a clearinghouse could be treated more favorably with respect to closeout or netting. More generally, the public sector should consider how steps to strengthen the legal infrastructure could help promote structural improvements in OTC derivatives markets. With these provisos in mind, the following proposals could potentially reduce the risk of market instability.
First, in the United States, the agencies supervising institutions and regulating markets (including the Federal Reserve System, the Treasury, the SEC, and the CFTC) agree that financial swaps should be exempt from CFTC supervision and regulation. The 1999 report by the President’s Working Group on Financial Markets on regulation of OTC derivatives recommended removing this uncertainty through legislative reforms that would grant swaps an exemption from potential CFTC oversight.100 This was well received by the private sector, and efforts led to revisions to legislation that would clear the way for serious private consideration of reorganizing OTC derivatives markets, including taking advantage of many of the risk-mitigating possibilities of a clearinghouse structure. In 2000, certain features of the legislation that was introduced and considered in the U.S. Congress raised some concerns from the U.S. Federal Reserve Board, the Treasury, and the SEC.101 As of May 2005, legislation to address these issues was still pending.102
If the legal obstacles are removed, a clearinghouse arrangement for OTC derivatives could mitigate risks associated with simple swaps—by handling clearing and settlement, formalizing and standardizing the management of counterparty risk through margin requirements, and mutualizing the risk of counterparty default—and thereby reinforce market discipline and encourage self-regulation.103 Would market participants need official encouragement to use a private clearinghouse? On the one hand, some market participants have expressed considerable skepticism about such an arrangement, and the clearing arrangements attempted thus far (such as SwapClear) have attracted little activity, in part because they are perceived to be costly, and they impose regulatory capital requirements. On the other hand, some market participants see a central clearinghouse as inevitable in view of the considerable operational difficulties of managing an OTC derivatives business, the challenges of managing credit risk on a bilateral basis, and the legal uncertainty of the OTC environment. In any case, if the regulatory environment is liberalized and the legal environment is clarified, the adoption of private electronic trading arrangements for swaps and other OTC derivatives (already in evidence since 2000 and continuing in early 2005) may well give rise to private clearinghouses.
Second, closeout procedures for derivatives contracts have proven to be impractical and ineffective in some jurisdictions and under some market circumstances. Had they worked effectively, some of the adverse market dynamics precipitated by the LTCM crisis might have been avoided. The uncertainty of the applicability of closeout procedures might be clarified by the appropriate regulatory and legal bodies, including at the Group of Ten level if the contracts involve more than one legal jurisdiction. Inaction could mean that virulent dynamics will not be avoided the next time rumors of default in OTC derivatives markets surface.
Third, netting arrangements, another risk-mitigation technique, can help reduce gross creditor and debtor counterparty positions to a single bilateral credit or debit with each counterparty. Uncertainty about the legality and regulatory treatment of these arrangements can give rise to situations of heightened credit risk. Further and stronger efforts should be made to strengthen the legal basis for netting.
Coordinated Improvements in Disclosure
Coordination is particularly necessary in the area of information disclosure. In finance, information is a source of economic rents. Natural limits determine how much information will be voluntarily provided publicly, or even privately, to establish counterparty relationships. Therefore, the private sector is unlikely, on its own accord, to provide the right amount and kind of information to counterparties, the markets, and authorities, unless it has incentives to do so. Accounting standards and prudential rules require certain forms of disclosure. However, information in 1998 and the early 2000s was insufficient for private counterparts, supervisors, or those responsible for market surveillance to reach the judgment that vulnerabilities were growing in the global financial system. The public sector has a strong role to play in providing incentives for greater disclosure to the markets, and greater information on a confidential basis to the official sector.
While in principle creditors have incentives to demand adequate disclosure from their counterparties, these incentives can be undermined by competitive pressures and concerns by their counterparties that confidentiality might not be protected. To overcome this weakness, the primacy of credit risk management can be emphasized, along with the autonomy of risk management within organizations to make confidentiality credible, in line with proposals by both private and official groups. The public sector role could be limited to assessing and monitoring the quality of risk management and control systems more systematically and thoroughly, and to defining how information is used, plus ensuring that counterparty disclosure is adequate. The counterparty market discipline imposed by creditors could also be strengthened significantly through better pricing and control of the terms of access to credit.
The challenges in improving public disclosure are formidable. The shift in the boundary between private and public information could, by reducing the private information advantage, lessen intermediation activity. The potential consequences for market functioning need to be weighed against the benefits for market participants of more information on risk concentrations. In addition, it will be difficult to guarantee confidentiality, and even more difficult to develop a consensus on what can usefully be disclosed, in what form, to whom, and how often. For these reasons, an eclectic, innovative approach is needed to address these challenges and pitfalls. Supervisors might promote and facilitate more exchange-like OTC market structures, such as clearinghouses and electronic trading and settlement systems, which would support greater transparency and potentially serve as a nexus for information. Supervisors and regulators could facilitate the adoption of such facilities by regulating them lightly and by devising arrangements for multilateral clearing of contracts that are already covered under bilateral master agreements. Addressing this challenge requires a continuous and close process of cooperation between public and private sectors to strike the right balance between financial market efficiency and stability.
Private and Public Roles in Reducing Systemic Risk
Both the private and public sectors must work to reduce systemic risk. Private market participants can—by developing and implementing effective risk management and control systems and risk-mitigation tools—individually ensure their own viability and soundness even in extreme circumstances. Well-managed and highly capitalized financial institutions are important components of the first lines of defense against systemic financial problems. Improved risk management and control would reduce the potential for excessive risk-taking and the buildup of vulnerabilities at individual institutions, and highly capitalized institutions are better able to absorb losses when they occur. If all institutions succeeded in accomplishing these objectives, the effectiveness of the first line of defense against systemic risk—market discipline—would be strengthened. If a chain is only as strong as its weakest link, no financial institution can be assured of dealing in OTC derivatives markets with counterparties that are managing their risks well unless all of the systemically important financial institutions substantially improve their risk management systems. Thus, there is some—albeit not strong—incentive for collective private action centered around improving risk management and the financial infrastructure of these systemically important markets. Such collective private action would support the efforts of the ISDA, the Group of Thirty, the Counterparty Risk Management Policy Group II (CRMPG II), and other groups dealing with corporate governance lapses.104 These efforts should be intensified and accelerated.
In addition to private actions to reduce systemic risk, authorities are responsible for ensuring financial stability through prudential regulations, banking supervision, and market surveillance. One strong step forward in the arena of prudential regulations would be for the Basel Committee on Banking Supervision to reconsider capital requirements for off-balance-sheet credit risks. While the Committee’s 2004 revised bank capital requirements (Basel Committee on Banking Supervision, 2004) go some way toward more effectively recognizing the risks in off-balance-sheet activities, the increasing sophistication of banks in arbitraging capital requirements and the dynamic nature of OTC derivatives exposures are likely to widen existing gaps in the measurement of banks’ overall credit exposures, and consequently in setting appropriate capital levels. The Committee should consider ways in which capital charges on OTC derivatives positions could more closely reflect the significant changes (positive and negative) that occur in a bank’s current and potential future credit exposures when market prices change. In this context, banks’ internal credit risk systems could be required to quantify off-balance-sheet credit exposures (both current and potential) as a basis for appropriate capital charges—subject to verification through effective supervision.
More generally, authorities face the difficult challenge of helping to ensure financial stability without encouraging risk-taking beyond some reasonable prudent level, without impeding financial innovation, and without unduly distorting market incentives. In principle, safeguards (including the financial safety net) promote a more desirable equilibrium than would be obtained without them. Perversely, the safeguards may also encourage excessive risk-taking. The challenge of keeping moral hazard to a bare minimum in the first instance requires authorities to engage in sufficient monitoring to ensure that the insured institutions and markets take appropriate account of the risks inherent in their activities. Banking supervision and market surveillance need to keep abreast of the changing financial landscape and the institutions that change it, and need to invest in developing analytical frameworks for understanding the changes.
This chapter draws on material in Schinasi and others (2000), some of which was also published in IMF (2000). The author is grateful to his colleagues for the joint effort at the time, and for their permission to draw on this publication.
See Box 9.1 “LTCM and Turbulence in Global Financial Markets,” for a description of the problems encountered by the hedge fund and the market turbulence associated with it.
See Glossary for definitions of special terms.
Among the exceptions, in Brazil all OTC derivatives transactions must be centrally registered. See United States, Commodity Futures Trading Commission (1999).
In the Basel Accord, the credit equivalent of an off-balance-sheet item (the basis for capital requirements) has two components: current replacement cost and add-ons designed to capture potential future credit exposure. Add-on ratios to notional value are specified in a matrix of contract types and remaining maturity. See Basel Committee on Banking Supervision (1995).
The ISDA and the Group of 30 are ongoing organizations with at least two decades of history each. ISDA, with membership of about 625 financial institutions, develops standards and serves as a forum for the discussion of legal and documentation issues surrounding OTC derivatives contracts. The Group of 30 consists of senior representatives of the private and public sectors and academia and explores international economic and financial issues; its influence has been most felt in global clearings and settlement processes. The Counterparty Risk Management Policy Group (CRMPG) and the Derivatives Policy Group (DPG) were much smaller groups, formed to tackle a limited range of issues. DPG consisted of six nonbank OTC derivatives dealers, and worked closely with the CFTC and the SEC in the mid-1990s to develop voluntary procedures for risk management, internal controls, and external reporting. CRMPG initially came together in early 1999, following the market turbulence of 1998, as a group of 12 internationally active commercial and investment banks with the objective of improving internal counterparty credit and market risk management practices. The group was reconstituted, with the encouragement of the president of the New York Federal Reserve Bank, in 2005 with 15 members as a response to growing concerns about the market implications of the rapid growth of unregulated hedge funds.
Federal Reserve Chairman Alan Greenspan noted (2005, p. 2), “To be sure, the benefits of derivatives, both to individual institutions and to the financial system and economy as a whole, could be diminished, and financial instability could result, if the risks associated with their use are not managed effectively. Of particular importance is the management of counterparty credit risks.”
See testimony by Federal Reserve Chairman Greenspan, SEC Chairman Levitt, and Treasury Secretary Summers (United States Joint Senate Committees on Agriculture, Nutrition, and Forestry and Banking, Housing, and Urban Affairs, 2000).
See CRMPG II (2005).