The story of the hedge fund Long-Term Capital Management (LTCM) is the story of Icarus, a high-flyer who became one of the most spectacular failures financial markets have ever seen. The Russian Crisis is often referred to as the cause of LTCM’s financial downfall. However, the Russian Crisis was merely a trigger and indirect factor for LTCM’s collapse. The rise to liquidity across global fixed income markets arising from the Crisis paired with LTCM’s excessive leverage, illiquid investment positions and management mistakes shortly before and during the Crisis, caused its breakdown which was finally only halted by a bail-out organized by the US Federal Reserve Bank. This paper analyzes the contributing factors and interrelations between the two events and concludes with some implications for financial market participants and policy-makers from the two cases and empirical research on financial crises.
“ At times the mathematics of the models become too interesting and we lose sight of the models ’ ultimate purpose. The mathematics of the models are precise, but the models are not, being only approximations to the complex real world. [...] The practitioner should therefore apply the models only tentatively, assessing their limitations carefully in each application. ”
(Robert C. Merton, 1994)
These wise words from 1994 were spoken by Robert Merton, one of the masterminds of the late hedge fund Long-Term Capital Management (LTCM), likely still today the most notorious fund of its kind in the world. However, despite Robert Merton’s cautious approach to financial models, LTCM’s stellar success which was based on quantitative, computer-driven strategies came to a sudden end only a few years later when the fund’s equity melted down within weeks in the summer of 1998. Only an unprecedented rescue by the US Federal Reserve Bank prevented LTCM’s investors from a total loss of their investments and the world from a potential breakdown of financial markets. The demise of LTCM is often related to the so-called Russian Crisis, a debt, currency and banking crisis which sent shockwaves down financial markets in 1998. There are few papers relating to the events surrounding LTCM’s collapse likely due to the lack of available information on the positions taken by LTCM. This paper attempts to elucidate the backgrounds of the Crisis and LTCM’s failure individually, analyze the contribution of the Crisis to LTCM’s quasi-default and to relate this connection between macroeconomic events and hedge fund failure to the financial crises literature and the current crisis. The paper proceeds as follows: Section 2 presents some facts on the market environment LTCM was operating in during the late 1990ies, explains LTCM’s origins, its investment strategy and its performance from initial success to liquidation. Section 3 describes the evolvement of the Russian Crisis and Section 4 looks at the role of the Crisis for LTCM’s performance. In Section 5 implications of the Crisis and LTCM’s breakdown in the context of other more general research on financial crises are discussed and Section 6 concludes.
2. The Rise and Fall of LTCM
2.1. The Hedge Fund Environment in the Late 1990ies Relative to the Situation Today
Hedge funds have been enjoying increasing popularity for years now due to their alleged ability to combine attractive returns with reduced volatility and low correlation with traditional asset classes such as stocks and bonds. Being able to generate positive returns in both rising and falling stock and bond markets, hedge funds experienced a massive inflow of funds after the New Economy Crisis in 2001. However, already before that hedge funds witnessed a first boom period in the 1980ies. Then George Soros’ Quantum Fund hit the headlines with its hugely profitable bet in the European exchange rate crisis in September 1992. The industry gained the highest notoriety so far in their 50-year old history in the late 1990ies when LTCM spectacularly collapsed in 1998. When markets generally tumbled in 2001, notoriety changed into popularity.
LTCM was by far the largest hedge fund ever of its time with $1.25bn raised initially in 1994. In the third quarter of 1995 the total amount of capital invested in hedge funds globally was thought to be $100bn only, a tiny figure relative to the $1.4 trillion at the end of 2008 which shows the industry’s level of maturity at the time. The number of hedge funds had tripled from 1,000 at year-end 1992 to more than 3,000 in 1998. The beginning of the 1990ies had been a boom-and-bust time for hedge funds: while hundreds of new funds were founded, numerous also ceased to exist with poor performance being a major causal factor. The average hedge fund in 1998 had a size between $30-90m. The rapid growth of the hedge fund industry in the early to mid 1990ies was driven by more and more high-net-worth individuals in the world and by increasing interest from institutional investors who were attracted by the low correlation hedge funds offered relative to traditional assets and by their impressive Sharpe ratios as reported by Edwards (1999).
Two generic investment strategies can be distinguished today as in 1998: Some hedge funds focus on speculative strategies (e.g. event-driven, distressed securities, global macro, or aggressive growth) whereas others pursue arbitrage strategies (e.g. market neutral, merger arbitrage). Nevertheless, also the isolation of arbitrage opportunities can represent a risky investment concept for investors because (i) hedge funds (arbitrage funds in particular) often rely on tremendous leverage and use derivatives extensively, (ii) the exact investment strategy is a secret and hence the investors’ risk exposure from its hedge fund investment relative to his overall investment portfolio is unclear (blind pool risk), (iii) the high proportion of performance-driven, variable income generally incentivizes hedge fund managers to follow risky strategies. In order to align interests and reduce the fear of excessive risk-taking, all hedge fund managers usually invest a significant amount of their own personal wealth into the fund alongside external investors. But due to the generally unregulated character of the hedge fund industry and the inherent risks for external investors, authorities in the US impose a minimum of regulation by limiting the circle of hedge fund investors to “sophisticated”, i.e. wealthy or institutional, investors considered to be aware of the risks facing them.
In 1998 it was unclear if hedge funds were able to outperform the traditional markets on a risk-adjusted post-fee basis and LTCM provided arguments for both sides in the course of its history as we will see in the next section. Today, there is a lot of evidence indicating outperformance, if not for the whole industry at least for the upper quantiles of the fund universe which are said to persistently outperform other asset classes. Yet, there is still an ongoing discussion as to whether researchers appropriately capture risk in their models. But even if hedge funds do not outperform on a net-net basis, their low betas can turn them into valuable additions for a diversified portfolio. The dark side of the medal, however, is the unfortunate causal relationship between increasing capital flows in the industry and decreasing returns, due to fewer investment opportunities chased by more and more funds. In the current crisis, it appeared disastrous that more and more hedge funds were following the same strategies, because liquidity quickly dried up in market segments dominated by hedge funds when the wind changed its direction; in the Russian Crisis LTCM alone was so dominating that liquidity evaporated in certain swap markets when only LTCM wanted to reverse its trading positions.
2.2. The Founders and the Concept
LTCM had been founded in 1993 in Greenwich, Connecticut, by former Salomon Brothers star bond trader and arbitrageur, John Meriwether, and two Nobel prize winning finance academics with guru status, Robert Merton and Myron Scholes who were eager to apply their valuation models in practice. Together they gathered an impressive list of professionals from diverse backgrounds around them including David Mullins, a former Vice Chairman of the Board of Governors of the Federal Reserve System. International market knowledge, profound trading skills and rigorous academic understanding were combined as foundation for a new, superior hedge fund business model. LTCM’s investment strategy was based on identifying market abnormalities in interest rate products like bonds or swaps or unravelling atypical volatility in market prices. By assuming short positions in overpriced securities and simultaneously going long in corresponding securities deemed undervalued, the fund should profit from market corrections of the differential between the two securities no matter in which direction the market generally developed (“convergence trades”). LTCM believed that partly as a result of the Asian Crisis in 1997 the yield differential between high- and low-yield bonds (or less and more liquid securities) was excessively wide and would narrow when the markets reassessed the risk. These arbitrage play opportunities were identified with the help of a quantitative strategy based on complex mathematical models, massive databases and vast computing power. Although designed as an arbitrage hedge fund in general considered less risky vis-à-vis speculative hedge funds, LTCM’s portfolio also included speculative bets.
Because a reduction of these spreads was usually small in magnitude, the positions had to be significantly leveraged in order to generate the expected returns on equity. LTCM operated with a leverage factor of 25 which meant that a return on capital that was only 1% higher than the cost of debt was already sufficient to reach impressive returns on equity of above 25%. This magnitude of leverage was also very high by industry standards. Nevertheless, LTCM’s management team claimed that the fund’s risk exposure was comparable to that of a broad stock market index such as the S&P 500. This statement relied on the fund’s risk management system which in turn was primarily based upon the relatively new concept of Value-at-Risk (VaR). The concept was considered to be very sophisticated at the time because unlike previous models’ inherent and sole focus on past volatility, it relied on daily estimates of prospective return variability and allows quantifying risk while it is being taken.
LTCM’s hedge fund team and concept were perceived to be so promising that the management around John Meriwether was able to impose very generous investment conditions for itself. Apart from above-average performance and management fees, the external investors also had to agree to an unusually strict lock-in clause which foresaw that investors were required to commit their funds for at least three years. Nevertheless, the young fund was able to attract some of the top financial institutions from around the globe like the United Bank of Switzerland which ended up losing significant amounts ($700m in the third quarter of 2008 alone, albeit not nearly as much as it lost in the current crisis).
2.3. Performance Until the Russian Crisis
Solid reputation of the individual team members, a convincing strategy and good contacts to regulators provided the basis for a head start. Scale effects, leverage and substantial investments in infrastructure and manpower settled the matter. In its first three years of existence between 1994 and 1996 LTCM appeared to have constructed a money machine; return on equity was 20% in 1994, 47% in 1995 and 45% in 1996. The fund outperformed the US stock market (which was rallying at the time) and blue chip corporate bonds. Of course, this performance is based on returns only; LTCM’s risk-adjusted returns are said to have been not nearly as good. The conditions during LTCM’s first years of existence were very convenient. Efficient markets showed linear behaviour and statistical analysis could identify trends relatively easily. In particular, LTCM’s bets on interest rate convergence for corporate and government bonds among the different member states of the European Monetary System proved to be very profitable until 1997 when rates had narrowed substantially as the introduction of the Euro as a common currency on January 1, 1999 had come closer.
Roger Lowenstein mentions in his book “When Genius Failed” that LTCM’s return on overall capital was only 2.45% in 1995. The high return on equity was thus the result of extremely low cost of debt. LTCM’s hard bargaining paid off because banks did not want to be left out of business with the fund. Thus, the fund’s equity grew from $1.25bn in 1994 to $7bn at the end of 1997. As a result of the lower leverage level arising from the increased equity base and increasing competition by copiers, in 1997 return on equity dropped to 17% while US stocks gained 33% in that year. In response, the fund returned $2.7bn to its investors because “investment opportunities were not large and attractive enough” bringing the leverage ratio back up to 28. Troubles began not long thereafter in May 1998 when a downturn in the mortgage-based security markets (today a well known predictor of crises) decimated LTCM’s equity by 16% increasing the leverage ratio to 31.
 The step was not totally unprecedented since institutions other than hedge funds had been backed before like after Penn Central’s default in 1970 or the encouragement of lending to brokerage firms after the stock market crash in 1987.
 37% of all hedge funds existing at some point during 1989 and 1996 did no longer exist at the end of the period. 60% of hedge funds disappeared within three years according to Edwards (1999).
 Sections 3(c)1 and 3(c)7 of the Investment Company Act of 1940.
 While the office was located in Connecticut, the legal structure involved the creation of two partnerships, one on the Cayman Islands and one in Delaware.
 A prominent example for such a mis-match spread is the (usually minimal) pricing difference between on-the- run (the most recently issued and most liquid) and off-the-run US treasury bills. Credit spreads between same maturity corporate and treasury bonds were also compared to their historical means and evaluated as either too large or too small which then lead to an according dual investment position in the belief that convergence to the historical mean would eventually set in.
 This calculation is based on the following formula: re = rce + L * (rce - cd) where re is return on equity, rce return on capital employed, L represents the leverage ratio (debt divided by equity) and cd is cost of debt.
 In 1998 a third of hedge funds did not borrow at all, of those that borrowed 54% borrowed not more than the equity they had and for those who borrowed more leverage ratios rarely exceeded 10 (Edwards, 1999).
 VaR is a risk measure of the worst loss on a specific financial portfolio under normal market conditions which emerged in the late 1980ies and is widely used as a result of the impetus to its use from the world-wide adoption of the Basel II Accord. Formally, VaR is defined as a threshold value such that the probability that the mark-to- market loss on a given portfolio over a given time horizon exceeds this value, is the specified probability.
 Typically hedge funds charged 20% of profits plus 1% of assets under management; LTCM charged a 25% performance fee and a 2% asset management fee.
 John Meriwether in the Washington Post.