When Genius Failed: The Rise & Fall of Long-Term Capital Management

Roger Lowenstein, “When Genius Failed: The Rise & Fall of Long-Term Capital Management” Random House 2000, 236 pps  

I have probably read over 30 books dealing with the financial crisis of 2008, and in most of that reading there is some reference to the 1998 failure of Long Term Capital Management (LTCM) and the contributions to that failure by its founder, John Meriwether.  So, even though I thought such a relatively old event had little to teach me about the current crisis, I read it anyway.  Much to my surprise, the reading experience of this book is much like reading about more recent failures such as Lehman’s.  It becomes evident to me that the rewards of the financial industry are such that the lessons of these past failures are affordably best ignored for the players; at least as long as they can continue to privatize the profits and socialize the losses.  The dramatic story of LTCM’s fall is a chilling harbinger of the crisis that would strike all of Wall Street, from Lehman Brothers to AIG, a decade later. In his new Afterword, Lowenstein shows that LTCM’s implosion should be seen not as a one-off drama, but as a template for market meltdowns in an age of instability—and as a wake-up call that Wall Street and government alike tragically ignored.  Tragic--at least for the taxpayers; not so much for those that kept the proceeds.

The financial industry approach to investment using computer modeling has been, and still is, based on the Black-Scholes Option model.  This model at the beginning of 1998 predicted that the probability of LTCM to lose all its capital in one year would be a statistical freak occurring one in every ten to the twenty-fourth power times—a so-called ten-sigma event. 

When it was founded in 1993, Long-Term was hailed as the most impressive hedge fund in history.  When LTCM crashed they had positions in securities with over a trillion dollars of face value. At the time, many investment banks were losing hundreds of millions of dollars on similar market bets. There was concern at the US Federal Reserve that if LTCM defaulted on their contracts it would cause chaos and a market crash. This in turn could place the US economy in danger. So the Fed organized a bail-out. A consortium of banks and trading houses put four billion dollars into LTCM. In return the consortium took possession of LTCM's market positions. The investors that had money in LTCM got ten cents on their invested dollar. The partners were largely wiped out. Once LTCM's market positions were unwound the rescue consortium made money on their investment.  LTCM's mistakes were made fatal by massive leverage and lack of liquidity. If not for their huge leverage, LTCM could have survived these mistakes, or at least survived without such breathtaking losses.  At least this failure was not paid for by taxpayers, unlike 2008.

In the epilogue Lowenstein summarizes LTCM's losses:

Investment                                                                                                                    LOSS                                                                                                                                     

Russia and other emerging markets                                                                                $430 million

Directional trades in developed countries (such as shorting Japanese bonds)                  $371 million

Equity pairs trading                                                                                                         $286 million

Yield-curve arbitrage                                                                                                       $215 million

Standard & Poor's 500 stocks                                                                                          $203 million

High-yield (junk bond) arbitrage                                                                                     $100 million

Interest swaps                                                                                                              $1.6 billion

Equity volatility bets                                                                                                     $1.3 billion

My Notes:

Pg. xiii:  On September 23, 1998 the chiefs of Bankers Trust, Bear Stearns, Chase Manhattan, Goldman Sachs, J.P. Morgan, Lehman Brothers, Merrill Lynch, Morgan Stanley Dean Witter, and Salomon Smith Barney gathered in the New York Federal bank to consider a rescue of LTCM.  Joining them were the chairman of the New York Stock Exchange and representatives from major European banks.  LTCM was on the brink of failing.  For four years LTCM had been the envy of Wall Street.  The fund had racked up returns of more than 40 percent a year, with no losing stretches, no volatility, seemingly no risk at all.  This fund had amassed $100 billion in assets, virtually all of it borrowed—borrowed from the bankers now gathered.  Along with this massive indebtedness was the additional problem of the thousands of derivative contracts, which had endlessly intertwined LTCM with every bank on Wall Street.  These contracts, essentially side bets on market prices, covered more than $1 trillion worth of exposure.  If LTCM defaulted, all of the banks in the room would be left holding one side of a contract for which the other side no longer existed.

Pg. 24:  As far as securities law is concerned, there is no such thing as a hedge fund. In practice, the term refers to a limited partnership, at least a small number of which have operated since the 1920s; they are private and largely unregulated investment pools for the rich.  In return for this freedom from regulation, the funds must limit access to a select few investors.  By law, funds can sign up no more than ninety-nine investors, people, or institutions each worth at least $1 million, or up to five hundred investors, assuming that each has a portfolio of at least $5 million.  Presumably, millionaires know that they are doing; if not, their losses are nobody’s business but their own.

Pg. 27:  Meriwether planned to initially raise a colossal $2.5 billion to start his fund.  Everything about LTCM was ambitious.  Its fees would be considerably higher than average at 25 percent of the profits, in addition to a yearly 2 percent charge on assets.  (Most funds took 20 percent of profits and 1 percent on assets.)  Moreover, the fund insisted that investors commit for at least three years, an almost unheard-of lockup in the hedge fund world.  To better attract investors, Meriwether recruited professors Merton and Scholes to the fund, each said to be on the shortlist of Nobel candidates.  He also hired David Mullins, vice chairman of the Federal Reserve and second in the Fed’s hierarchy to Alan Greenspan, the Fed chairman.  LTCM opened for business in February 1994 with $1.25 billion—well short of J.M’s goal but still the largest start-up ever.

Pg. 42:  By leveraging one security, investors potentially give up control of all of their others.  So, although their securities might be unrelated, because they and many other investors owned them, they implicitly linked them in times of stress.  And when armies of financial soldiers are involved in the same securities, borders shrink.  The very concept of safety through diversification—the basis of LTCM’s security—would become meaningless. The real culprit in most financial crisis is leverage, not liquidity.  Whenever markets plunge, investors are stunned to find that there are not enough buyers to go around.  The mistake is in thinking that markets have a duty to stay liquid or that buyers will always be present to accommodate sellers.  If you aren’t in debt (leveraged), you can’t go broke and can’t be made to sell, in which case liquidity is irrelevant.  But a leveraged firm may be forced to sell, lest fast-accumulating losses put it out of business. 

Pg. 63:  Notice that there is a key difference between a share of IMB (or an infectious disease) and a pair of dice.  With dice, there is risk—you could, after all, roll snake eyes—but there is no uncertainty, because you know (for certain) the chances of getting a 7 and every other result.  Investing confronts us with both risk and uncertainty.  There is a risk that the price of a share of IBM will fall, and there is uncertainty about how likely it is to do so.  So many variables—political, economic, managerial, competitive factors—can affect the result that the uncertainty all but overwhelms us.

Pg. 68:  The LTCM professors computer model treated the volatility of a security like an inherent unchanging trait.  You or I might assume that the market fluctuations of so many yesterdays are so much noise—arbitrary, not necessarily likely to recur, and best forgotten.  But to Black, Scholes, and Merton—and to LTCM—these fluctuations were invested with deep predictive significance.  Each tick of the market up or down was viewed as latent with an unerring forecast of future risk.  This implied that markets were efficient and rational at every step.  Merton carried the assumption a step further.  He assumed that volatility was so constant that, for example a share of IBM, would never plunge directly from 80 to 60 but  would always stop at 79 ¾, 79 ½, etc., along the way.  Therefore at each infinitesimal moment, traders would readjust the price of option on IBM keeping them in synchrony with the price of the stock.  And traders who owned both could—by nimbly buying or selling—keep their portfolio in a risk-free state of balance.  This assumption may approximate real markets when they are calm—but only then.

Pg. 72:  There is a reason why financial markets run to extremes more often than coin flips.  The coin doesn’t remember that it landed on tails three times in a row; the odds on the fourth flip are still fifty-fifty.  But market trends will sometimes continue just because traders expect (or fear) that they will.  Black-Scholes model assumes a rational efficient market without the human element that so often, in real life, overwhelms the market.  Even the Federal Reserve was endorsing such programs at the time of LTCM; these programs went by the generic name “Value-at-Risk.” 

Pg. 102:  Regulation T sets a limit on broker loans for stocks, also known as margin.  For the past twenty-five years, the Fed has set the maximum margin loan at 50 percent of the total investment.  LTCM got around this by building its equity positions without actually buying actual securities.  Rather, the fund entered into derivative contracts that mimicked the behavior of stocks.  If, for instance, LTCMM wanted to earn the return on $100 million of CBS stock over a three-year period, it would strike a “swap” contract with, say Swiss Bank.  LTCM would agree to make a fixed annual payment, calculated as an interest rate on $100 million.  And Swiss Bank would agree to pay LTCM whatever profit would have been earned had LTCM actually purchased the stock.  This is not new; Wall Street has been using equity swaps to get around Reg T for about a decade.

Pg. 113:  Long-Term’s plan was to return at the end of 1997, al profits on money invested during 1994, its first year, and to return all money (principal and profits) invested after that date.  They were determined to keep all the future profits for themselves as this return of money invested excluded the partners and employees and partially excluded some big strategic investors such as the Bank of Taiwan.  Jimmy Cayne, chief executive of Bear Stearns, Long-Term’s clearing broker, got an exception, too.  The excluded were angry (but lucky).

Pg. 123:  Early in 1998, Long-Term began to short large amounts of equity volatility.  More than any other, ‘equity vol’ was Long-Term’s signature trade, and it set the fund ineluctably on the road to disaster.  Equity vol comes straight from the Black-Scholes model.  It is based on the assumption that the volatility of stocks is, over time, consistent.  The stock market, for instance, typically varies by about 15 percent to 20 percent a year.  Now and then, the market might be more volatile, but it will always revert to form—or so the mathematicians at Long-Term believed.  It was guided by the unseen law of large numbers, which assured the world of a normal distribution.  This flowed from the professors Mertonian view of markets as efficient machines that spit out new prices with all the random logic of heat molecules dispersing through a cloud.  And when the models told them that the markets were mispricing equity vol, they were willing to bet the firm on it.  However, there is no direct way of making this bet.   But there is an indirect way.  Remember that, according to the Black-Scholes formula, the key element in pricing an option is the expected volatility of the underlying asset.  As the asset gets jumpier, the price of the option rises.  Therefore, if you knew the price of an option, you could infer the level of volatility the market was expecting.  The options market appeared to be anticipating volatility in the stock market of roughly 20 percent.  Long-Term viewed this as incorrect, because actual volatility was only about 15 percent.  Thus, they figured that option prices would sooner or later fall.  So Long-Term began to short options—specifically, options on the Standard & Poor’s 500 stock index.  Then on August 17, 1998 Russia declared a debt moratorium.  Long-Term, which had calculated with such mathematical certainty that it was unlikely to lose more than $35 million on any single day, dropped $553 million—15 percent of its capital—on that one day.