Not for the first time, Soros’s dual personas caused trouble. His government interlocutors did not know how to interpret his views: Was Soros saying what would be good for the world, or was he saying what would be good for his portfolio? Sensing that his private discussions were not gaining traction, Soros went public with his Russia plan by publishing a long letter in the Financial Times on August 13; now it was investors’ turn to feel uncertain. Soros proposed that, as part of a package that would include new Western financing, the Russians should reduce the burden of the ruble-denominated GKOs by devaluing the currency by 15 percent to 25 percent. It was a sensible policy prescription, but to everyone in the markets it read as a public announcement that Soros was shorting the ruble. On the reasonable assumption that Soros was getting ready to repeat the sterling trade of 1992, investors ran for the exit; the day after Soros’s letter appeared, the yield on GKOs hit 165 percent. On Monday, August 17, facing an inexorable assault from the markets, Russia devalued the ruble and defaulted on its debts to foreigners.51
The truth, of course, was that Soros had not shorted the ruble. He had seen the inevitability of devaluation and even hastened the moment; but instead of selling the life out of the currency in the knowledge that politicians would do nothing to save it, he had attempted to make politicians behave differently. Indeed, far from being short the ruble, the Soros funds were vastly long—on top of the $1 billion exposure to Svyazinvest, they owned all kinds of Russian bonds and equities.52 As a result of the default and devaluation, Quantum and its sister funds lost 15 percent of their capital, or between $1 billion and $2 billion.53
For Stan Druckenmiller, whose lifework was the performance of the Quantum Funds, it was a bitter moment. For anyone who knew the size and manner of the loss, it made nonsense of the idea that Soros was a superpredator. But Soros himself processed the failure on an entirely different level. Looking back on the experience, he wrote, “I have no regrets with regard to my attempts to help Russia move toward an open society: They did not succeed but at least I tried.”54
Soros had come down from Mount Olympus like a messiah to save sinners. He had suffered crucifixion.
10
THE ENEMY IS US
For years the Bank of China building in Hong Kong was heavy and squat: It projected the granite solidity of the old-fashioned finance. Then the bank bought itself a face-lift from the Chinese-American wizard I. M. Pei, and in 1990 it emerged as a soaring, daring, giddy thing: a slim, stiletto-heeled goddess on the catwalk of Hong Kong’s financial district. The architect explained that his triangular towers were intended to evoke bamboo, their sectioned shafts of aluminum and glass tapering as they rose into the sunlight. But the comparison could be pushed further than Pei knew. Through the miracles of modern leverage, financial institutions were indeed growing as quickly as bamboo, sucking money out of the world’s burgeoning bond markets and pumping it back in again. But thanks to that same leverage, the new financial palaces were thin sided and hollow. An unexpected sideways blow could topple them.
For anyone who knew the two Bank of China buildings, the celebration that took place in September 1997 was ironically located. A confident young hedge fund called Long-Term Capital Management—the epitome of the new financial engineering that Pei’s structure evoked—picked the art gallery at the squat old premises to throw a party. To the south and the west lay Indonesia and Thailand, which were struggling with currency crises; ensconced in a hotel suite not far away, Malaysia’s prime minister was waging his campaign against speculators. But Long-Term Capital seemed to float above the region’s storm. Its small team of economists had earned a stunning $2.1 billion in profits the previous year; and its magic formula appeared to work irrespective of the turmoil roiling Asia.1 The fund’s prestige and prominence were reflected in the party’s guest list, which included an A team of private-sector bosses and financial officials who were in town for the IMF/World Bank annual meetings. After the champagne had stopped flowing, the economists returned to their hotel to find a fax of the front page of that morning’s Wall Street Journal. There, above the fold, the Journal reported their decision to return two fifths of their fund’s capital, $2.7 billion, to outside investors. In the new world of soaring leverage, Long-Term Capital had no need for so much client cash. By boosting its borrowing, it could maintain its towering portfolio on a thinner foundation. It could be ambitious and slender, like an I. M. Pei creation.2
Long-Term Capital Management’s founder, John Meriwether, had been one of the first executives on Wall Street to see the potential in financial engineering. As a rising star at Salomon Brothers in the mid-1980s, he had set out to transform the small trading group he managed into “a quasi-university environment.”3 Meriwether’s plan was to hire young stars from PhD programs and encourage them to stay in touch with cutting-edge research; they would visit finance faculties and go out on the academic conference circuit. He recruited Eric Rosenfeld, a Harvard Business School professor, then scooped up Larry Hilibrand, who had not one but two degrees from the Massachusetts Institute of Technology. By 1990 Meriwether’s team included Robert Merton and Myron Scholes, who would later win the Nobel Prize for their pioneering work on options pricing.
In the mid-1980s, most Salomon partners had not gone to college, much less a PhD program.4 The personification of the firm’s trading culture was Craig Coats Jr., a tall, handsome, charismatic stud believed by many to be the model for the hero in Tom Wolfe’s The Bonfire of the Vanities. Coats ran Salomon’s government-bond trading the old-fashioned way: While Meriwether’s professors debated whether the relationship between two bonds was out of its normal range, or whether the volatility of a bond price was likely to decelerate, Coats’s main tool was a firm belief in his own instincts. But remarkably quickly, the superiority of Meriwether’s professors became obvious. By the end of the 1980s, the small quasi faculty accounted for 90 percent of the profits at Salomon. Coats left Salomon Brothers after a big trading loss; Meriwether, for his part, was elevated to the position of vice chairman. Quantitative precision had triumphed. It was the end of anti-intellectualism on Wall Street.5
There was a problem with this victory, however. If Meriwether’s hundred-strong department could generate tens of millions of dollars, what was the purpose of Salomon’s six thousand other employees? Meriwether’s lieutenants complained that the fruits of their brilliance were being spread to undeserving corners of the firm. Larry Hilibrand, the double-MIT lieutenant, campaigned to shutter the fancy internal catering service that Salomon maintained for the benefit of its investment bankers, even suggesting that the investment banking division should be closed entirely. Eventually this push bore fruit in a secret deal: To keep the rocket scientists happy, Salomon’s overlords granted them a fixed 15 percent of their group’s profits. By securing a guaranteed performance fee, Meriwether had created a hedge fund within a bank. Creating a hedge fund instead of a bank was merely a step away for him.
The event that forced that extra step was not of his own choosing. In 1991, the Treasury bond scandal that embarrassed Michael Steinhardt and Bruce Kovner triggered a full-blown crisis for Salomon Brothers. The firm’s government-bond trader, Paul Mozer, had cheated repeatedly in the auctions; Meriwether, who was responsible for overseeing Mozer, resigned from the firm and paid a $50,000 fine imposed by the Securities and Exchange Commission.6 After scouting about for opportunities, Meriwether resolved to set up on his own. He would reassemble his team of rocket scientists and would do it without the unnecessary trappings of a big bank: Functions like marketing, clearing, settling, and operations would be outsourced, so that there would be no need to spread the professors’ trading profits through undeserving back-office departments. The way Meriwether saw it, he was inventing a new kind of financial institution for a new age. A world in which a small brotherhood of academics could earn more than a large bank required a fresh kind of setup. It required “Salomon without the bullshit.”7
IN FEBRUARY 1994, MERIWETHER
LAUNCHED LONG-TERM Capital Management. He brought along Eric Rosenfeld, Larry Hilibrand, Robert Merton, and Myron Scholes; in all, eight members of his Salomon brain trust joined in setting up the company. The professors leased space at 600 Steamboat Road in Greenwich, Connecticut, in the same four-story building overlooking the Long Island Sound to which Paul Tudor Jones had moved recently. Instead of New York suits and ties, they showed up for work in golf shirts and chinos. Sometimes in the lunch hour, bluefish could be spotted jumping out of the sound, and a team of eager quants would arm themselves with fishing rods and race out in hot pursuit of them.
Stripped down to its essentials, Long-Term Capital’s approach to the bond market recalled A. W. Jones’s innovation.8 Just as a Jones manager might buy Ford shares and short Chrysler, believing Ford’s management to be superior, so Meriwether’s team would buy one bond and short a similar one, believing the first bond’s cash flow to be more promising. The hedging out of market risk worked better with bonds than with stocks. A Jones manager—or for that matter, a long/short stock picker at Julian Robertson’s Tiger—might think Ford’s managers were better than Chrysler’s, but it was only an opinion. It involved judging the characters of the managers, eying the designs of their new cars, collecting the gossip on the morale of their sales teams. But bond investing was a different game: There was just a loan, an interest rate, and a promise to repay on a date certain. A bond analyst could find you two securities that were almost indistinguishable: The issuer was the same, the principal would be repaid in the same year, the legal documents describing the investor’s rights were word-for-word identical. If one of these bonds was trading for less than the other, you could buy the cheap one and short its overpriced pair. This was arbitrage, not simple investing, and it promised almost certain profits.
The quintessential LTCM trade started with the fact that newly issued Treasury bonds, known as “on-the-run” Treasuries, were bought and sold with great frequency. Traders who valued this liquidity were willing to pay a premium over marginally older, less frequently traded “off-the-run” Treasuries. But during the bond’s lifetime, the premium disappeared; the payout on a 30-year bond and a 29½-year bond were bound to come together by the time they hit their repayment dates. Meriwether’s team could simply sell the overpriced new bonds; buy the cheaper, older ones; and then wait patiently for the inevitable convergence. In ordinary times, admittedly, the profits from this strategy were barely enough to offset transaction costs. But when the market was panicky, the liquidity premium could balloon: Skittish traders wanted to own bonds they could sell in a hurry, and they were prepared to pay for the privilege. Meriwether’s lieutenants waited for these moments of panic, then put on the convergence trade. Larry Hilibrand, the LTCM partner who had campaigned against Salomon’s fancy catering unit, compared markets to Slinkies. They would always spring back. You just had to wait until the panic subsided.
Another classic Meriwether trade involved the Italian bond market.9 Italy’s cumbersome tax rules deterred foreigners from investing in the country’s bond market; as a result, demand was suppressed and the bonds were a bargain. A foreigner who figured out how to get around the tax obstacle could buy the bonds and collect a yield of, say, 10 percent. Then he could hedge the position by borrowing lire in the international money market at perhaps 9 percent, pocketing the 1-percentage-point difference. And the solution to the tax problem was hiding in plain sight. The trick was to go into partnership with a bank that was unencumbered by the tax issue because it was registered in Italy.10 Starting when they were still at Salomon Brothers and continuing into their first years at LTCM, Meriwether’s team seized on the Italian trade with an enthusiasm unrivaled on Wall Street. During LTCM’s first two years of trading, Italy contributed around $600 million of the firm’s $1.6 billion profits.11
LTCM was not the only player to crowd into Italy. Other banks and hedge funds followed; meanwhile, the Italian government got rid of the tax obstacle. But as the original trade ceased to be profitable, LTCM had fun in other corners of the Italian market. Italy’s retail investors were plunking their savings in a particular type of government bond that corner-shop banks sold to them, and because of this captive group of purchasers, the bond in question was clearly overvalued. Long-Term sold the bond short, then used the proceeds from those sales to buy lire bonds with higher interest rates, pocketing the spread as usual.12 Moreover, Italian savers were gradually waking up to the attractions of mutual funds and were switching their money away from direct bond purchases. As a result, demand for the bonds that LTCM had shorted went down, adding to LTCM’s profits.
Meriwether and his partners scoured the world for this sort of opportunity. They spotted probable convergences in all kinds of settings: between different bonds of the same maturity, between a bond and the futures contract that was based on it, between Treasury and mortgage-backed bonds or between bonds in different currencies.13 The common theme was that market anomalies occur when the behavior of investors is distorted—whether by tax rules, government regulation, or the idiosyncratic needs of large financial institutions. French insurance companies, for example, needed to buy French government bonds of particular maturities—not because they thought those maturities represented a bargain, but because they needed assets that matched the maturities of their promises to insurance customers. Similarly, in October 1996 a Federal Reserve ruling induced U.S. banks to issue lots of bonds and swap the payments on them into a floating rate; this flood of issuance depressed the fixed rate available in the swaps market. In each of these cases, LTCM took the other side—effectively trading against people who were buying or selling because institutional requirements compelled them to do so. By being the flexible player with the freedom to mirror the quirks of the inflexible ones, Long-Term provided liquidity to the markets. French insurers and American banks fulfilled their institutional imperative at a better price than they would otherwise have done. Meanwhile LTCM itself reaped fabulous profits.
Long-Term Capital Management’s success showed how lucrative this game was. Even after subtracting its 2 percent management fee and its hefty 25 percent performance fee, LTCM returned 19.9 percent in its ten months of trading in 1994, followed by 42.8 percent in 1995 and 40.8 percent the year after. It generated these returns, moreover, without riding the markets: The gains from its convergence trades were not correlated with any stock or bond index. Small wonder that LTCM had no difficulty raising capital. To its office in Greenwich it added bureaus in London and Tokyo; having launched with a slim staff of 41 people, Long-Term employed 165 by the time of the Bank of China party in the fall of 1997.14 Eric Rosenfeld, whom Meriwether had plucked from Harvard just a decade earlier, built a ten-thousand-bottle wine cellar, stocked directly from France. An LTCM partner named Greg Hawkins kept thoroughbred horses.15 Meriwether himself bought Waterville, an enchanting golf course in County Kerry, Ireland, to which he invited other Wall Street heavyweights to cement his business relationships. “Everybody was enamored with their intellect,” a Merrill Lynch salesman remembered. “It was like Kennedy’s inner circle—Camelot! They have the best and the brightest.”16
Of course, within a year of the Bank of China party, LTCM had blown up. Not for the first time, the newfangled finance turned out to be fragile, with large and unappreciated risks for the entire world economy.
IN HIS BEST-SELLING ACCOUNT OF LONG-TERM CAPITAL Management’s brief life, Roger Lowenstein portrays the fund’s demise as a punishment for hubris. This is ultimately correct, but it is not as though the firm was crass about its risk taking. Meriwether and his partners were not gambling irresponsibly with OPM—Wall Street’s contemptuous acronym for “other people’s money.” To the contrary, most of the partners invested nearly all their earnings in the business, year after year; and by ejecting $2.7 billion of outsiders’ capital in the fall of 1997, they ensured that nearly a third of the remaining fund was their own savings. Unlike many financiers who reduce their institutions to ruin, Long-Term Capital�
�s partners had every incentive to be prudent.
It is true that LTCM operated with extremely high leverage. Indeed, leverage was the very essence of the firm: The pricing anomalies it found were too small to be worth much without the multiplier of borrowed money. In 1995, for example, Long-Term’s return on assets, at 2.45 percent, was modest; but leverage transformed an indifferent return on assets into a spectacular return on capital—2.45 percent became 42.8 percent. The leverage was safe, Meriwether reasoned, because LTCM hedged out nearly all the risks in its trades. Soon after it opened, for example, the firm created a $2 billion position in on-the-run and off-the-run Treasuries. The exposure might have sounded daunting for a $1 billion fund, but Meriwether’s team calculated that betting on convergence was one twenty-fifth as risky as owning either bond on its own. The firm’s $2 billion position was the equivalent of an $80 million position for an unhedged investor.
LTCM was one of the first hedge funds to quantify its risk mathematically.17 Macro traders like Druckenmiller kept their exposures in their head; they had a feel for how much a market might swing and how much they could lose on any major position. Long-Term used a technique developed in the 1980s known as “value at risk,” which was essentially a formalization of the macro traders’ mental computations. LTCM worked out the volatility of each position, then translated that finding into a dollar amount that could be lost in normal circumstances. For example, Long-Term might buy one Italian bond and short another one, betting that the gap between them would converge; by studying the history of this trade, it might discover that, on ninety-nine days out of a hundred, the worst that was likely to happen was that the gap would widen by ten basis points.18 If the trade was sized such that a one-basis-point widening cost LTCM $10 million, the fund’s loss from a ten-basis-point move would be $100 million.
More Money Than God_Hedge Funds and the Making of a New Elite Page 26
