More Money Than God_Hedge Funds and the Making of a New Elite

by Sebastian Mallaby

  12. These details on Amaranth’s positions, and many others that follow, are drawn from a lengthy report by the U.S. Senate Permanent Subcommittee on Investigations, which reports to the Committee on Homeland Security and Government Affairs. See U.S. Senate Permanent Subcommittee on Investigations, “Excessive Speculation in the Natural Gas Market,” June 25, 2007. The report is not flawless. It draws the conclusion that Amaranth’s failure makes the case for additional hedge-fund regulation, whereas the failure is better seen as an example of the market disciplining a rogue trader. Further, the report makes much of the total exposure accumulated by Amaranth in various futures contracts, not explaining that the net exposure matters more and that natural-gas futures are traded over the counter, making it impossible to know how much of the total market Amaranth accounted for. Nevertheless, the Senate investigators did collect a vast amount of valuable data and testimony on Amaranth’s natural-gas trading. In the judgment of the Senate report, “Amaranth’s large-scale trading was a major driver behind the rise of the January/November price spread from $1.40 in mid-February to $2.20 in late April, an increase of more than 50 percent.” The Senate report states, “On every trading day in May, Amaranth accounted for at least 55 percent of the open interest in the November 2006 contract…Put simply, Amaranth was too big for the market it had created.” Even allowing for the caveat that NYMEX is not the whole of the gas-futures market, Amaranth’s share of NYMEX trading is striking.

  13. At the time, Blackstone kept its withdrawal secret. A Blackstone official explains that publicity might have caused other investors to flee Amaranth, creating a run on the fund that might have provoked a freeze on withdrawals, trapping Blackstone’s money.

  14. Amaranth’s willingness to pay Morgan Stanley a large fee to get out of certain gas positions confirms the verdict that it had grown too big for the market. If it had been able to trade out of its positions easily, it would have done so. The Morgan Stanley evidence matters because Amaranth representatives have sometimes suggested that the fund was brought down not by its excessive size, but rather by conspiracies against the fund, ranging from predatory trading on the NYMEX in late August to J.P. Morgan’s opposition to the Goldman Sachs deal in September.

  15. Amaranth’s broad exposures were well known because the fund provided investors with monthly reports detailing returns and outlining how these had been generated. Hedge-fund transparency is generally considered a good thing, but there is a risk to it.

  16. U.S. Senate Permanent Subcommittee on Investigations, “Excessive Speculation in the Natural Gas Market.”

  17. A former trader at Amaranth comments, “They counseled Brian to get out. He needed to be ordered.” Another Amaranth official says, a bit uncertainly, “I don’t want to believe that Brian Hunter didn’t try to reduce his positions. Because we were told that he was trying, but there just wasn’t enough liquidity.” Yet a third Amaranth veteran describes extensive debates within the firm as to how quickly to cut the natural-gas exposure; these concluded in the view that it was unwise to pay a high price in order to exit precipitously. It is not clear that Amaranth could have saved itself by opting to exit quickly at all costs. If Hunter had cut his positions aggressively any time after April, he might well have taken losses so large as to put Amaranth out of business.

  18. Looking back on this period, one trader describes Hunter as “a menace.” Equally, the Senate report quotes numerous traders to this effect. For instance, one says, “Everyone in the market knew Amaranth killed MotherRock.” Amaranth denies it.

  19. The Amaranth veteran comments, “Remember, I said the guy fell in love. Maybe that’s what we’re talking about. Maybe it’s just another manifestation of the love…. I told you he thought this guy could do no wrong. And when he made that statement [to the Wall Street Journal] I’m sure he believed it.”

  20. An Amaranth official recalls that some of Hunter’s summer/winter positions were designed to hedge others, but that by September supposedly offsetting positions were going wrong simultaneously, suggesting that Amaranth was being targeted by rival traders.

  21. This dialogue comes from interviews with Winkler and Griffin and from a complaint filed by Amaranth against J.P. Morgan in the New York State Supreme Court on November 13, 2007.

  22. Many big banks run multiple computer systems, which would have made it hard to sync Amaranth data into all the relevant divisions.

  23. J. Tomilson Hill, vice chairman of the Blackstone Group, comments, “If Citadel had been big enough in 1998 to buy LTCM, the odds would have been much better that a deal would have gotten done.” J. Tomilson Hill, interview with the author, September 9, 2009.

  24. This account is based on interviews with Ken Griffin and other Citadel staff members, as well as with Karl Wachter and Charles Winkler of Amaranth.

  25. One popular regulatory response to the growth of leveraged trading is to push over-the-counter derivatives such as swaps onto exchanges. Although this response is generally reasonable, it should be noted that most of Hunter’s gas exposure was on an exchange, and further that the exchange authorities were ineffectual in limiting his excessive trading. By contrast, the discipline of the market proved brutally effective.

  CHAPTER FIFTEEN: RIDING THE STORM

  1. John Gittelsohn, “High Roller of Home Loans,” Orange County Register, May 20, 2007.

  2. Mark Pittman, “Bass Shorted ‘God I Hope You’re Wrong’ Wall Street,” Bloomberg, December 19, 2007.

  3. Michael Litt, interview with the author, July 2, 2009. The BIS report was “The Recent Behavior of Financial Market Volatility” (BIS paper 29, August 2006). For an account of FrontPoint’s portfolio manager, Steve Eisman, see Michael Lewis, “The End,” Portfolio, December 2008. It should be noted that the BIS was actually among the few official institutions to warn of the risk of a financial crisis, even though this warning was not apparent to Litt.

  4. From inception in December 2006 to mid-October 2007, Bass’s dedicated mortgage fund was up 463 percent. FrontPoint ran multistrategy funds, so the mortgage bets were diluted. Nevertheless, in 2007 FrontPoint’s low-volatility multistrategy fund was up 23 percent and its midvolatility version was up 44 percent.

  5. The following account of Paulson’s subprime trade is reconstructed from conversations with John Paulson and Paolo Pellegrini and from a report produced by Paulson and Company. John Paulson, interview with the author, July 15, 2009; Paolo Pellegrini, interview with the author, July 2, 2009; Paulson and Company, “Paulson Credit Opportunities, 2007 Year End Report.”

  6. Pellegrini interview.

  7. Paulson and Pellegrini soon realized their error. Their research showed that the percentage of mortgage loans extended on the basis of limited documentation had risen from 27 percent in 2001 to 41 percent in 2005, but it also showed that refinancing was covering up the problem of poor loan quality. Between 1998 and 2006, at least half of subprime mortgages were refinanced within five years.

  8. This is the conversation as remembered by John Paulson. Paulson interview.

  9. Paulson’s plan was to buy insurance on $12 of subprime mortgages for every $1 he had in his fund, so a $600 million fund involved buying insurance on $7.2 billion of mortgages. The cost of this insurance was about 1 percent of the value of the mortgages, so 12 percent of the value of the fund. But Paulson earned 5 percent from the interest on the free cash in the fund, so that the net cost of putting on the bet was 7 percent of the fund’s assets.

  10. Gregory Zuckerman, The Greatest Trade Ever: The Behind-the-Scenes Story of How John Paulson Defied Wall Street and Made Financial History (New York: Broadway Books, 2009), p. 197.

  11. Zuckerman, The Greatest Trade Ever, p. 208.

  12. In 2008, according to Hedge Fund Research, asset-backed hedge funds were down 3 percent, a respectable showing given the carnage that surrounded them. Although the rest of the hedge-fund industry suffered a hard year in 2008, it was not because it fell for subprime mortgages.

  13.
In perhaps the clearest example of this folly, UBS vacuumed up $50 billion worth of AAA mortgage bonds, confident that AAA paper would always pay back; in 2007 alone, this decision accounted for $12.5 billion of losses.

  14. Paul Muolo and Matthew Padilla, Chain of Blame: How Wall Street Caused the Mortgage and Credit Crisis (Hoboken, NJ: John Wiley & Sons, 2008), p. 190. Other accounts confirm O’Neal’s determination to raise Merrill’s ranking in mortgage securitization. See Bradley Keoun and Jody Shenn, “Merrill Loaded for Bear in Mortgage Market That Humiliated HSBC,” Bloomberg, February 12, 2007.

  15. William D. Cohan, House of Cards: A Tale of Hubris and Wretched Excess on Wall Street (New York: Doubleday, 2009), p. 281.

  16. “Investors who sought to take advantage of the inimitable risk management reputation of Bear Stearns found themselves in a highly complex hedge fund investment program that relied on overworked junior personnel to manage a conflict reporting process required by federal law.” Administrative complaint against Bear Stearns Asset Management filed by the Commonwealth of Massachusetts, quoted in Cohan, House of Cards, p. 302.

  17. Hedge fund subsidiaries of other banks also fared poorly. UBS’s Dillon Read Capital Management and Royal Bank of Scotland’s Greenwich Capital were both wound down in 2007 following losses on subprime securities. Much as happened at Bear Stearns, UBS injected capital into its failed funds, took their losses onto its balance sheet, and then found itself in need of a government bailout. Of UBS’s $19 billion in losses in 2007, Dillon Read accounted for $3 billion. Meanwhile, the Royal Bank of Scotland had to take RBS Greenwich’s losses onto its balance sheet, contributing to the bank’s later collapse into the arms of the UK government.

  18. These figures come from Paulson and Company, “Paulson Credit Opportunities, 2007 Year End Report.” The cumulative figure is reached by compounding the 20 percent return in 2006 with the 590 percent return in 2007.

  19. This story and the ensuing account of the Sowood transaction is reconstructed from interviews with Ken Griffin, Gerald Beeson, and Adam Cooper of Citadel. Ken Griffin, interview with the author, July 9, 2009; Gerald Beeson, interview with the author, June 30, 2009; Adam Cooper, interview with the author, June 30, 2009.

  20. Kyle Bass of Hayman Capital wrote in an investor letter dated July 31, 2007: “What is truly remarkable about this particular situation is the fact that Jeff Larson, the former manager of the $30 billion Harvard Endowment, is the principal Manager at this firm. Sowood was renowned as being a ‘best-in-class’ fund.” Kyle Bass, letter to Hayman Capital investors, July 30, 2007. See also Jenny Strasburg and Katherine Burton, “Sowood Funds Lose More Than 50% as Debt Markets Fall (Update 4),” Bloomberg, July 31, 2007.

  21. Gregory Zuckerman and Craig Karmin, “Sowood’s Short, Hot Summer,” Wall Street Journal, October 27, 2007.

  22. Some press accounts note that Larson and Griffin spoke on Friday, July 27. But Griffin, Cooper, and Beeson separately recall that the key phone conversation was on Sunday.

  23. Jeff Larson, letter to Sowood investors, July 30, 2007.

  24. Cohan, House of Cards, p. 381.

  25. Jim Cramer, “Street Signs,” CNBC, August 3, 2007.

  26. A quant firm could believe both in stock momentum and in momentum reversal. Both effects could exist, but on different time horizons.

  27. For instance, Black Mesa, a small quantitative hedge fund based in New Mexico, reported in an investor letter that a pattern of liquidation started on July 25, 2007, and lasted through Friday. “The losses were found not to be attributable to common market risks,” Black Mesa reported. “The losses were in our proprietary factors or, in other words, attributable to risks to which we deliberately expose ourselves.”

  28. Many in the quant industry suspect that the storm of deleveraging was started by Bruce Kovner’s Caxton Associates. This is not quite right. It is true that Kovner assembled his portfolio managers on the evening of Sunday, August 5, and instructed them to cut risk. But the meeting did not include Aaron Sosnick, who managed the capital that Caxton committed to statistical arbitrage. Rather, Sosnick had cut his leverage substantially in the previous several days, so was not selling aggressively on Monday, August 6, the start of the quant quake. Bruce Kovner, interview with the author, October 14, 2009.

  29. Quant equity hedge funds in the summer of 2007 seem to have been leveraged between six to one and eight to one. They sometimes described this as leverage of “three to four,” meaning three to four times on the short side and the same amount on the long side, giving a total leverage of six to eight.

  30. Clifford Asness, “The August of Our Discontent: Questions and Answers about the Crash and Subsequent Rebound of Quantitative Stock Selection Strategies,” working paper, September 21, 2007.

  31. Cliff Asness explains, “By and large much of quantitative investing is about common sense and discipline, rather than about esoteric math and computer algorithms…. The computers help us process the data and maintain a diversified and disciplined approach…. It’s about good investing done broadly and without the often dangerous influence of tick-by-tick human emotion. Our strategies are not ‘black boxes.’” (Clifford Asness, “The August of Our Discontent: Questions and Answers about the Crash and Subsequent Rebound of Quantitative Stock Selection Strategies,” working paper, September 21, 2007.) In an e-mail to investors, Jim Simons wrote, “While we believe we have an excellent set of predictive signals, some of these are undoubtedly shared by a number of long/short hedge funds.” (Jim Simons, e-mail to Renaissance Technologies investors, August 9, 2007.)

  32. Satya Pradhuman, director of research at Cirrus Research, identified 148 companies with market capitalizations between $2 billion and $10 billion and 473 companies with market capitalizations between $250 million and $2 billion in which large quant funds had ownership stakes exceeding 5 percent. See Justin Lahart, “How the ‘Quant’ Playbook Failed,” Wall Street Journal, August 24, 2007.

  33. Scott Patterson, “A Hedge-Fund King Is Forced to Regroup,” Wall Street Journal, May 26, 2009.

  34. Asness, “The August of Our Discontent.”

  35. Cliff Asness comments, “Most of our lives are about automated quant trading. But when you have a conflagration of this size, having good intelligence, having good contacts on the Street, those things really matter.” (Cliff Asness, interview with the author, July 9, 2009.) Similarly, Sushil Wadhwani, who was running his systematic funds in London, recalls, “I remember Friday morning…. It was a question of either someone came in with a bailout or they delevered.” (Sushil Wadhwani, interview with the author, July 28, 2009.)

  36. Here was yet another example of a hedge fund managed under the umbrella of an investment bank going wrong. The fact that the parent bank bailed out the hedge fund, as had happened at Bear Stearns, showed why the fund managers may have been less vigilant than their counterparts at independent funds with no deep-pocketed parents. J.P. Morgan analyst Stephen Wharton brought up this issue on Goldman’s conference call, organized to announce the recapitalization. “I mean do you feel there is some moral hazard being introduced here in terms of how investment banks are reacting to problematic hedge funds managed by their asset management arms?” Naturally, Goldman rejected the comparison, pointing out that it was providing $2 billion of the $3 billion recapitalization, with the rest coming from outside investors. Goldman Sachs conference call, final transcript, Thomson StreetEvents, August 13, 2007.

  37. See Amir E. Khandani and Andrew W. Lo, “What Happened to the Quants in August 2007?” working paper, November 4, 2007; Richard Bookstaber, A Demon of Our Own Design (Hoboken, NJ: John Wiley & Sons, 2007). See also Richard Bookstaber, “What’s Going On with Quant Hedge Funds?” (available at http://rick.bookstaber.com/2007/08/whats-going-on-with-quant-hedge-funds.html).

  38. “I have said before that ‘there is a new risk factor in our world,’” Cliff Asness wrote in the wake of the quant quake. “But it would have been more accurate if I had said ‘there is a new risk fac
tor in our world and it is us.’” See Asness, “The August of Our Discontent.”

  39. Cliff Asness and Adam Berger, “We’re Not Dead Yet,” Alpha, November 2008.

  40. “It’s hard to prove that if there are a lot of people in a space, returns get worse. You can look at the performance and conclude that, like Lo does. Increased assets in a space, performance goes down—it sounds reasonable. But if you try to actually demonstrate is by building a portfolio and saying, ‘This is a portfolio I would like to get but can’t because the market is slipping away from me,’ you can’t quite do that. At Thales we find we can trade the same model with a slight variation: one that sets trades on a two-to three-day time scale and one that trades on a five-day time scale. They don’t even interfere with one another even though these are two models that are almost the same. The likelihood that our particular model is being interfered with by Shaw or Caxton or Citadel seems low.” Marek Fludzinski, interview with the author, June 25, 2009.

  41. Asness, “The August of Our Discontent.”

  42. Mike Mendelson, interview with the author, July 9, 2009.

  CHAPTER SIXTEEN: “HOW COULD THEY DO THIS?”

  1. In 2008, Kynikos was up more than 60 percent. Relative to the market’s performance, however, the 2007 return of over 30 percent was even better. Kynikos takes a performance fee based on its returns relative to the market benchmark.

  2. This remark and much of the telephone exchange between Chanos and Schwartz was reported by Gary Weiss and confirmed by the author in an interview with Chanos. See Gary Weiss, “The Man Who Made Too Much,” Portfolio, February 2009. Weiss also reports that Schwartz disputes the timing and detail of the call, saying it took place one day earlier, on Wednesday. However, Chanos remembers receiving the call on the way to see Bernstein, and Bernstein confirms that the dinner was on Thursday and that Chanos told him then about the phone call. James Chanos, interview with the author, September 23, 2009; Carl Bernstein, interview with the author, September 28, 2009.