29.1.1 Amaranth Advisors, LLC

Amaranth Advisors, LLC, was a self-described multistrategy hedge fund investing across asset classes and strategies. Multistrategy hedge funds are generally designed to employ a variety of investment strategies. However, Amaranth's fall from glory came from an extremely concentrated bet in the energy markets. Although Amaranth was technically a multistrategy hedge fund with positions across multiple asset classes, by 2006 it had devoted a large proportion of its risk capital to natural gas trading.

Amaranth was founded in 2000 by Nicholas Maounis and was headquartered in Greenwich, Connecticut. The founder's original expertise was in convertible bonds, but the fund later became involved in merger arbitrage, long/short equity, leveraged loans, and energy trading. As of June 30, 2006, energy trades accounted for about half of the fund's capital and generated about 75% of its profits.

Although Amaranth was based in Greenwich, its star trader, Brian Hunter, was based in Calgary, Alberta, Canada. The Wall Street Journal reported that Amaranth's head energy trader sometimes held “open positions to buy or sell tens of billions of dollars of commodities.”¹ Due to the hedge fund incentive fee structure, Amaranth had a big incentive to take big bets. This is also detailed in the Wall Street Journal article: “At Amaranth, star energy trader Brian Hunter won an estimated $75 million bonus after his team produced a $1.26 billion profit in 2005. Like many others at the fund, he had to keep about 30% of his pay in the fund. The fund's chief risk officer, Robert Jones, got a bonus of at least $5 million for 2005, say people familiar with the bonuses.”

Amaranth engaged in a commodity futures strategy of establishing long positions in winter delivery contracts for natural gas and going short the non-winter contracts. Understanding the nuances of natural gas trading requires some background. The key economic function for natural gas in the United States is to provide for heating demand during the winter in the northern United States, although natural gas is also a key energy source for generating electricity for air-conditioning demand during the summer. There is a long injection season, from spring through fall, in which natural gas is injected and stored in caverns for later use during the long winter season. By the end of February 2006, Amaranth held nearly 70% of the open interest in the November futures contracts on the New York Mercantile Exchange (NYMEX) and nearly 60% of the futures for January.

The size of those positions in natural gas contracts led to Amaranth's collapse. But Amaranth's fall did not happen overnight. After starting 2006 with $7.5 billion in investor assets, the fund soared to $9.2 billion but then eventually tumbled to less than $3 billion. By the end of May 2006, at least some of Amaranth's traders and officers were aware of the firm's predicament, as it had lost approximately $1 billion in that month alone. Still, Amaranth continued to put more money into its natural gas bets. By the end of August, Amaranth was still pumping in more money to hold its positions, but the market continued to move against the fund. When natural gas prices fell in September 2006, Amaranth found itself losing a substantial amount of money on its very large positions. These losses led Amaranth to scramble to transfer its natural gas futures contracts to third-party financial institutions over the weekend of September 16. Initially, Merrill Lynch agreed to take on 25% of Amaranth's positions for a payment of about $250 million. Amaranth lost another $800 million on Tuesday, September 19. The next day, on September 20, Amaranth succeeded in transferring its remaining energy positions to Citadel Investment Group and to its prime broker, JPMorgan Chase, at a $2.15 billion discount to their mark-to-market value, using the previous day's prices. In total, Amaranth lost $6.6 billion in a matter of months.

Prime brokers generally treat hedge funds as their most favored clients (MFCs). But when it comes to a quickly eroding hedge fund, MFC status can change quickly, and a hedge fund can find that prime brokers exacerbate the hedge fund's problems rather than alleviate them. This appears to be part of the story for Amaranth. On Monday, September 18, just when Amaranth thought it had a rescue plan negotiated with Goldman Sachs, its prime broker, JPMorgan, refused to release the collateral that Amaranth had deposited with JPMorgan. This effectively killed any potential bailout of Amaranth.

What are the lessons to be learned here? A multistrategy fund should clearly not have 50% of its capital dedicated to one specific market or trade, especially if that fund espouses to be well diversified. Investors should watch for signs that a fund has concentrated positions, such as may be indicated by a volatility of monthly returns exceeding 10%. Fund management should closely supervise all traders, even those star traders who have previously earned billions of dollars in profits.

Risk managers at Amaranth may have allowed Hunter to take more risk than was justified simply because of the size of his prior trading gains. The geographic separation between Hunter and Amaranth may have also been a factor, as Hunter ran a small office in Alberta, far from Amaranth's Connecticut home. This separation could have made it difficult to discover any concealed dealings. Hunter, however, had a checkered past, which should have given some investors pause. It was said that in his prior career at Deutsche Bank, Hunter's trading privileges were revoked after a series of crippling losses on energy trades.

Hedge funds often operate in a lightly regulated environment. The lack of regulation over Amaranth was compounded by the fact that its trading schemes were conducted in the complex and uncoordinated world of natural gas markets. For example, the U.S. regulatory umbrella covering energy trading has had a noteworthy gap in coverage. The exchange-traded futures markets are explicitly regulated by the Commodity Futures Trading Commission (CFTC), and the physical natural gas markets are explicitly regulated by the Federal Energy Regulatory Commission (FERC). However, over-the-counter energy derivatives trading has not been subject to the same regulatory scrutiny, and it was on such platforms that Amaranth carried out a substantial portion of its trading. Amaranth traded many of its natural gas positions on the Intercontinental Exchange (ICE) as opposed to the New York Mercantile Exchange (NYMEX). Both exchanges trade natural gas contracts; however, under what is known as the Enron loophole, electronic exchanges like the ICE are not as highly regulated as physical exchanges like the NYMEX.

The risk management process at Amaranth may have understated the probability of long-term and large unidirectional price changes. When risk measures, such as value at risk, are estimated based on historical data that reflect an abnormally calm period of price changes, they understate reasonable anticipations of future risk. For example, estimations of risk based on historical data during periods of short-term price changes that are uncorrelated or mean-reverting typically understate the potential magnitude of longer-term price changes over periods of positive autocorrelation, or trending. Due to the larger position sizes, leverage magnifies the impact of underestimating underlying risks. Therefore, positions using leverage should be based on worst-case scenario losses rather than volatility estimates from a recently calm market. For example, if data over a longer time interval, such as a complete market cycle, show a maximum drawdown of 10%, then levering similar positions at rates of more than 10 times capital is likely to lead to a complete loss of investor capital during the next down cycle for that market. Also, Amaranth apparently had no formal stop-loss or concentration limits. High leverage, unanticipated large unidirectional moves, and insufficient processes to control risk can cause fund failure.

Amaranth illustrates another problem with large positions relative to the liquidity of the underlying markets. As a fund experiences financial stress, its need to liquidate positions can cause further adverse price movements in markets. Note that the process of liquidating long positions in illiquid markets can drive prices down (and liquidating short positions in illiquid markets can drive prices up), fueling further losses and further liquidations. The problem is especially severe when, as is often the case, the events leading to a fund's trouble coincide with illiquidity in the market. Adding to the problem is that during periods of market stress and the accompanying need for funds to receive credit to ride out the adverse price movements, financial institutions often experience similar stress and restrict credit.

29.1.2 Long-Term Capital Management

World financial markets faced a dramatic crisis in 1998 when a Greenwich, Connecticut, hedge fund named Long-Term Capital Management (LTCM) collapsed. At the time, LTCM was considered one of the largest and best-managed hedge funds in the world.

LTCM was founded in 1994 by several executives from Salomon Brothers Inc., and its board included two Nobel laureates in economics. Its troubles began in May 1998 with huge losses in its mortgage-backed arbitrage portfolio. LTCM's net worth shrank from a peak of $5 billion to about $2.3 billion in August 1998 and to just under $400 million by the end of September.² How did this huge fund with its stellar management team collapse so quickly when much of the world's economy was performing quite well?

LTCM focused on relative value strategies, such as fixed-income arbitrage. The premise for these trades was the expectation that the spread in prices or rates between two similar securities would converge over time. LTCM would buy the cheaper security and short the more expensive security and wait for the spread between the two similar securities to narrow before closing the trades.

LTCM used extensive leverage, based apparently on management's confidence that the fund's models could successfully identify mispriced securities and that large mispricings were virtually sure to be corrected on a timely basis. LTCM's massive degrees of leverage, including a leverage ratio of 25 to 1 in its cash positions, grew to over 50 to 1 in 1998. In addition, the fund had gross notional amounts of futures contracts that exceeded $500 billion, swap positions that exceeded $750 billion, and other derivative positions that exceeded $150 billion.³ The leverage ratio implied by these positions approached 300 to 1.

LTCM's already weakened positions were pummeled in August 1998 when the Russian government defaulted on the payment of its outstanding bonds. As a result of the Russian bond default, there was a sudden and drastic liquidity crisis that caused spreads to widen across a broad range of markets rather than contract, as LTCM's models had predicted. The fund's positions quickly accumulated large losses that led to a margin call from its prime broker. LTCM was forced to liquidate some of its positions. But liquidating in the midst of illiquid market conditions caused spreads to widen further. LTCM's losses continued to grow, which in turn led to more margin calls as its finances spiraled downward.

The situation for LTCM was bleak, and large financial institutions feared that if it were forced to liquidate the majority of its portfolio, there would be a systemic impact in the financial markets. The U.S. Federal Reserve Bank stepped in with three rate reductions within six months, but this action did not save LTCM. Finally, on September 23, at the neutral site of the Federal Reserve Bank of New York, 14 banks and brokerage firms met and agreed to provide capital infusions totaling $3.6 billion to LTCM. In return, the consortium of banks and brokerage firms received 90% ownership of the fund.

Although the cause of LTCM's demise was clear, the real question is: How was LTCM able to procure such a huge amount of credit that it could leverage its cash positions at a 25 to 1 ratio and its derivative positions at almost a 300 to 1 ratio? It was simple: LTCM never disclosed its full positions, because its counterparties did not demand information on the size of its total positions or its total credit exposure. As a result, LTCM was able to amass tremendous positions and credit.

It should be noted that LTCM's spread trades would have performed well if LTCM had had more time to work its way out of the liquidity crisis that gripped the markets. It was not that LTCM had poor trade ideas; on the contrary, its valuation models were robust. Instead, the problem was the inability to weather a major liquidity crisis due to the use of too much leverage. When the Russian bond default occurred, there was a flight to quality; LTCM's relative value positions diverged instead of converging and, with its very large amounts of leverage, LTCM collapsed.

29.1.3 Carlyle Capital Corporation

Another swift and stunning reversal of fortune was experienced by Carlyle Capital Corporation (CCC), which was created by the Carlyle Group in 2007. The Carlyle Group is one of the most successful alternative asset firms in the world, managing as of 2014 more than $200 billion in assets for some of the world's most sophisticated clients. The Carlyle Group created CCC as part of its efforts to diversify its business and to give public shareholders a way to get exposure to some of its funds.

Unlike most such funds, CCC was listed on Euronext Amsterdam. Therefore, it was available to the public starting with its listing in July 2007. Partners of the Carlyle Group retained a 15% ownership in CCC. Unfortunately, within weeks of its public listing, the Carlyle Group was forced to make its first bailout of the fund with additional injected capital. The fund's demise came just eight months later and was a surprise to many. How could such a new and publicly traded fund with such a prestigious parent firm collapse so quickly?

Carlyle Capital Corporation's strategy was not fraught with complex derivatives or secretive black-box trading schemes. Its strategy was simple: It borrowed money at low short-term interest rates and invested this borrowed capital in long-term AAA-rated mortgage bonds issued by Freddie Mac and Fannie Mae. Fannie Mae, Freddie Mac, and Ginnie Mae are nicknames for three large U.S. government–sponsored companies with the corresponding acronyms FNMA, FHLMC, and GNMA. These agencies bought mortgages and issued bonds backed by an explicit or implicit guarantee by the U.S. government. Debate existed as to whether the full credit risk of the bonds issued by the agencies was unambiguously backed by the U.S. government, but the bonds were considered safe enough by the rating agencies to receive AAA ratings.

Carlyle Capital Corporation's investment strategy was to make money on the difference between the cost of funding the AAA-rated mortgage securities and the interest received on them. It used aggressive leverage; for every $1 in capital it raised from investors, CCC borrowed about $31 to invest in the U.S. housing agency bonds. Its aim was to use this leverage to amplify the narrow spread between the return on its assets and the cost of its funding to generate an annual dividend of around 10%. It used only about $670 million in cash equity to finance its $21.7 billion portfolio of securities issued by Freddie Mac and Fannie Mae.

Although Freddie Mac and Fannie Mae securities were considered almost certain to be repaid, their value plummeted dramatically in February and March 2008, as investors worldwide shunned risk of any type and as the U.S. housing market continued to suffer. Carlyle Capital Corporation's financial health was vulnerable because of the highly leveraged nature of its investment strategy.

Dire funding problems first emerged on March 5, 2008, when CCC said it had been unable to meet margin calls from four banks on short-term repurchase agreements. Just two days earlier, the chief executive, John Stomber, had told investors on a conference call that the fund wasn't seeing increased margin pressure from its lenders. But within a week, the margin calls had reached more than $400 million, and lending banks had seized about three-quarters of CCC's assets. Efforts to put in place a standstill agreement with banks holding the remaining assets failed late in the week of March 10.

The substantial decline in value of Fannie Mae and Freddie Mac securities came to a head with margin calls from Deutsche Bank, JPMorgan Chase, and other lenders that reached more than $900 million. At that point, the lenders began to seize the fund's collateral and its chief assets, the AAA-rated mortgage-backed securities. The share price for CCC declined swiftly, from its public offering price of $20 in July 2007 to $0.31 in March 2008. During the week of March 10, 2008, CCC declared that it would wind up its operations and further stated that there would be no money left for shareholders.

CCC's collapse was a casualty of a liquidity crisis that led to more than $50 billion in losses at major investment banks. What was the lesson here? No security is safe from a liquidity crisis. The securities purchased by CCC were presumed to be U.S.-government backed and immune from credit risk. But as the credit crisis took hold, even these safe investments declined substantially in value. Indeed, as history played out later in 2008, Fannie Mae and Freddie Mac were taken over by the U.S. government, and the implicit guarantee of their bonds was finally made explicit. Even the Carlyle Group, with its clout and reputation, could not negotiate a grace period from its bankers to save CCC. Prime brokers and bankers have no patience or compassion when it comes to declining collateral values.

29.1.4 Declining Investment Opportunities and Leverage

Some funds choose to use aggressive levels of leverage from the start. Other funds drift into dangerous levels of leverage due to declining investment opportunities in a previously lucrative strategy or market. Declining profits can lead fund managers to take on increasing leverage and pursue riskier trades to maintain performance levels. This section reviews the mechanics of leverage and returns.

Return on equity (ROE) is profit after financing costs, expressed as a percentage of equity. Return on assets (ROA) is profit before financing costs (and taxes), expressed as a percentage of assets. ROE can be expressed as a function of ROA, leverage (L, which is defined here as the ratio of assets to equity), and interest costs on the financing (r):

(29.1)

In Equation 29.1, the rightmost term in brackets is the total interest expense from leverage, expressed as a percentage of equity. Note that without leverage, such that L = 1, ROE will be equal to ROA. Equation 29.1 illustrates the sensitivity of ROE to ROA (i.e., asset performance) when L is large as well as the sensitivity of ROE to financing costs (r) when L is large.

29.1.5 Behavioral Biases and Risk Taking

Examining the use, measurement, and dangers of leverage is reasonably straightforward. However, the search for profitable trades in markets with increasing competition raises more challenging issues. Financial markets with substantially increased competition are often described as too much money chasing too few opportunities. The dynamics of these environments deserve careful consideration, since these environments are ripe for market-driven collapses and fraudulent schemes that attempt to mask losses.

Firms seeking to maintain trading profits in increasingly competitive markets may become overexposed to particular risks. First, when profitable trading opportunities are rare, fund managers may concentrate positions in particular bets that they perceive as having value. Second, managers may be more likely to invest in opportunities that they have misunderstood. As competition increases and as truly valuable opportunities become rarer, it is increasingly likely that managers will expand their search for opportunities until they find a bet that appears unusually attractive. But even an overpriced opportunity can appear attractive to an analyst when that analyst errs by missing or misunderstanding the true risks. In a behavioral sense, an analyst may be biased toward concluding that an opportunity is truly attractive rather than interpreting the finding as the result of an error in evaluation. The result is that the fund may become overexposed to risks of which it is unaware.

Behavioral finance provides explanations of why these concerns are valid. Behavioral finance studies the potential impacts of cognitive, emotional, and social factors on financing decision-making. For example, confirmation bias is the tendency to disproportionately interpret results that confirm a previously held opinion as being true. The previous discussion provided an example. The analyst knows that profitability predictions are susceptible to error. But based on experience, the analyst believes that highly profitable trading opportunities can be identified. The analyst then examines numerous trading opportunities. Since the evaluations of each opportunity are subject to error, the analyst may locate an investment opportunity based on a false prediction of high profitability. A confirmation bias would cause the analyst to be too likely to decide that the reason for the finding is that the opportunity will be highly profitable rather than to conclude that the prediction was made in error.

The analyst's belief that profitable trading opportunities exist in a particular market may be a result of anchoring. Anchoring may be viewed in this context as a tendency to rely too heavily on previous beliefs. An analyst observing past successful searches for profitable trades may disproportionately expect that new opportunities will be found, despite knowing that market conditions have changed.

Confirmation bias and anchoring are examples of behavioral biases. Behavioral biases are tendencies or patterns exhibited by humans that conflict with prescriptions based on rationality and empiricism. Behavioral biases are generally viewed as important explanations for some investment behavior and often conflict with rational long-term investment decision-making. There are numerous types of behavioral biases that have been described and observed. In due diligence analysis, these biases should be presumed to affect both the fund managers being evaluated and the analysts charged with performing the analysis. For instance, analysts performing due diligence on a fund might desire to find that the fund represents a great investment opportunity and therefore be biased in favor of reaching a favorable conclusion. The primary methods of combating errors in decision-making due to behavioral biases are to rely on evidence, including academic research, and to use industry best practices that are based on careful analysis of empirical evidence and theory.

Analysts performing due diligence should be especially alert when analyzing funds using strategies that have become widely known as generating high returns. In these situations, increased use of leverage, increasing concentrations of positions, and reported returns out of line with competitors are valuable warning signals.

29.2.1 Quant Crisis, August 2007

During the first half of 2007, events in the U.S. subprime mortgage markets began affecting many parts of the U.S. financial industry, setting the stage for eventual turmoil in financial markets. By the first week of August 2007, several quantitative long/short equity hedge funds had sustained tremendous losses. Researchers from MIT subsequently conducted simulations of long/short equity portfolios and found evidence that the unwinding of fund positions began in July 2007 and continued until the end of the year.⁴ In 2008, Khandani and Lo analyzed what became known as the Quant Meltdown of August 2007 and proposed the unwind hypothesis to explain the events of the previous year. The unwind hypothesis suggests that hedge fund losses began with the forced liquidation of one or more large equity market-neutral portfolios, primarily to raise cash or reduce leverage. The subsequent price impact of this massive and sudden unwinding caused other similarly constructed portfolios to experience losses. These losses caused other funds to deleverage their portfolios, leading to a vicious spiral reminiscent of Long-Term Capital Management in 1998.

Some of the most successful equity hedge funds in the history of the industry reported record losses, although equity markets were only moderately affected by these troubles. Many of the funds suffering large losses were equity market-neutral and statistical arbitrage hedge funds.

The unwind hypothesis underscores the potential commonality of strategies among many quantitative equity market-neutral hedge funds. When large investors hold substantial positions in the same asset or similar assets, it is known as a crowded trade. Crowded trades are viewed as risky positions due to the relatively large potential for massive sell orders or buy orders placed by investors at approximately the same time. The pursuit of similar strategies across a set of very large hedge funds combined with rapid trading techniques facilitated by new technologies broadens the concern regarding crowded trades into concerns regarding crowded strategies.

29.2.2 The Flash Crash of 2010

On the afternoon of May 6, 2010, the United States' Dow Jones Industrial Average (the Dow) was down about 3% on the day based on macroeconomic concerns, such as the debt crisis in Greece. Suddenly the Dow dropped another 600 points (about 6%) in a period of five minutes, bringing the Dow to a loss of 900 points by 2:47 p.m. Then the market turned abruptly and began a rapid rally. By 3:07 p.m., the Dow had bounced back by about 600 points and leveled off to end the day where it had stood prior to the flash crash. During the day of the flash crash, the Dow experienced its second largest point swing to that date.

What caused this wild swing in prices? While there were initial rumors of trading errors, high-frequency traders, and technical glitches, news emerged that an intentional but very large E-Mini S&P 500 sell order may have triggered the crash. A report by the SEC pointed to “a large fundamental trader (a mutual fund complex) initiated a program to sell a total of 75,000 E-Mini S&P contracts (valued at approximately $4.1 billion) as a hedge to an existing equity position.”⁵ In April 2015, the U.S. Department of Justice attributed the large order to fraud and manipulation through spoofing and pursued criminal action against a British trader. Spoofing is the placing of large orders to influence market prices with no intention of honoring the orders if executed.

High-frequency trading firms and other traders jumped in and joined the selling of futures contracts. Arbitrageurs bought the depressed futures contracts and sold equities in the cash markets, driving down the cash equity market.

In the aftermath of the flash crash, circuit breakers were expanded. A circuit breaker is a decision rule and procedure wherein exchange authorities invoke trading restrictions (even exchange closures) in an attempt to mute market fluctuations and to give market participants time to digest information and formulate their trading responses. Despite their intended purposes, circuit breakers have also been argued to heighten risk due to concerns over illiquidity when exchanges suspend trading. These challenges and opportunities reflect the impact of advancing trading technologies on financial markets and their participants.

29.2.3 Knight Capital Group, 2012

Knight Capital Group was a global financial services firm engaged in market making, electronic execution, and institutional sales and trading. Knight was the largest trader in U.S. equities, with a market share of about 17% of equity exchange volume according to information supplied by the firm on its website in late 2011. One of Knight's major roles was receiving and executing order flow from large institutions, including investment companies, banks, and brokerage firms.

On Wednesday, August 1, 2012, Knight's trading caused a major disruption in the prices of about 150 companies listed on the NYSE. The problem was attributed to a technological breakdown, namely the installation of software that had caused Knight to enter millions of faulty trades in less than an hour. By the end of the day, Knight's losses from the error totaled $460 million.

Knight's technological issue appears to have stemmed from flaws in the oversight and management of a new technology deployed at Knight. According to an analysis by Nanex, a Knight algorithm appeared to have been repeatedly buying at the offer and selling at the bid, causing Knight to lose a small amount of money on each trade and resulting in a loss of huge sums due to the trades being repeated over and over again.

By Friday morning, Knight was in dire straits; as the Wall Street Journal explained, “In the span of two days, the company's market value had plunged to $253.4 million from $1.01 billion.”⁶ Major customers stopped doing business with Knight or were “dialing back” their trading through Knight. Ultimately, Knight reached an agreement with Getco LLC in December 2012 and was acquired for $1.4 billion in July 2013.

According to the New York Times, “The SEC blamed two ‘technology missteps’ for the trading fiasco on Aug. 1. It contends that Knight Capital failed to remove a defective function in one of its routers . . . [resulting] in a barrage of erroneous stock orders. . . . The regulator also contends that an automated e-mail identifying the error ahead of the market opening on Aug. 1 was sent to a group of employees. While these messages were not intended to be alerts, they provided a chance for the firm to fix the problem.”⁷ Knight agreed to pay a $12 million fine imposed by the SEC to settle charges that it had violated trading rules.⁸

The fine against Knight marked the first time that the SEC used a market-access rule against a trading firm; the new rule had been adopted in 2010 and required brokers and dealers with direct access to American exchanges to institute controls to protect the markets from such trading errors.⁹ Although regulatory efforts may mitigate some threats, risks from advanced trading technology remain and are relevant to all market participants. As in the case of Knight, rapid electronic trading can generate enormous benefits but can magnify even the smallest trading glitches. The problems at Knight disrupted markets and trade executions for several days. These events highlight the systemic risks posed by very large organizations, and indicate that the infrastructure supporting modern capital markets is highly technology dependent.

All three cases in this section emphasize not only the seriousness of technological failures but also the speed with which investors can lose money and the speed with which computer-driven trading systems need to be able to detect and correct errors.

29.3.1 Bayou Management

Bayou Management perpetrated one of the boldest of all hedge fund frauds. Bayou started out as a legitimate hedge fund but quickly degenerated into outright fraud. The two principals, founder Samuel Israel III and CFO Daniel Marino, were both eventually sentenced to more than 10 years in prison.

The story of Bayou began in 1996, when, within a few months after the Bayou fund opened and started trading, Bayou sustained trading losses and began lying to customers about the fund's profits and losses. Bayou concealed its true volatility and losses by fabricating results. In 1997, with profits falling short of the amount principals had projected, Bayou transferred back into the fund a portion of the trading commissions that the fund had paid to Bayou Securities during that year. Bayou Securities was a separate broker that Bayou Capital had set up to process the trades from the fund, meaning it was a captive brokerage firm and earned commissions on the Bayou fund's trades. It is typically a bad practice for hedge fund managers to earn brokerage fees on the trades of the funds they manage, as this gives the fund managers incentive to increase the number of trades. Further, there is an incentive for the managers to direct trades to their securities firm without demanding the best execution and lowest commission rates.

Bayou did not disclose to its clients that the fund's performance was being bolstered by these rebates of commissions. Consequently, Bayou's clients were left with a false impression that the fund had made a profit after commissions. Trading losses continued into 1998, when the fund sustained a net trading loss of millions of dollars. Over the course of the year, Israel and Marino concealed their losses by making material misstatements to clients about the Bayou fund's performance and the value of clients' investments. Israel, Marino, and a former Bayou principal concocted false investment returns to report to their clients and applied those false results to create inaccurate year-end financial statements.

By December 1998, the Bayou fund's mounting losses could not withstand an independent audit. So Bayou dismissed the fund's independent auditing firm and created fictional auditor's reports, financial statements, and performance summaries. Marino, a certified public accountant, agreed to fabricate the annual audit of the Bayou fund to conceal the trading losses. He created a fictitious accounting firm, Richmond-Fairfield Associates, to pose as the independent auditor of the Bayou fund. But Marino was the sole principal of Richmond-Fairfield, and the firm had no other clients.

In 1999, the fund again suffered substantial losses. Bayou again concealed the loss by creating and distributing false performance summaries and financial statements that purportedly had been audited by Richmond-Fairfield Associates. The trading losses continued to mount, and fictional financial statements and summaries continued to be issued from 2000 to 2002 to create the appearance of modest, reliable, and believable growth. The performance summaries sent out by Bayou indicated that clients were earning 1% to 2% in net profits each month.

Throughout this time, Bayou actively solicited new investors and additional investments from current investors, raising tens of millions of dollars of additional capital. In January 2003, to attract more investors and capital, the managers liquidated the Bayou fund and created four successor funds: Accredited, Affiliates, No Leverage, and Superfund. While investor deposits peaked in 2003 at more than $125 million, the reorganization did not improve true performance. The new funds lost even more money from trading activities. In 2003, Bayou Superfund took in more than $90 million in capital but lost approximately $35 million through trading. However, according to its 2003 annual statement, Bayou Superfund had earned more than $25 million. Also, throughout this period, the managers of Bayou's funds continued to collect profit-sharing fees on the fraudulent gains, and Bayou Securities continued to earn millions of dollars in trading commissions. By 2004, Israel and Marino had stopped actual trading and transferred Bayou's depleted assets to Israel and other non-Bayou entities, effectively the managers' personal accounts. Nevertheless, the managers still sent periodic statements to investors describing profitable trades.

Things began to unravel for Bayou when, in May 2005, legal authorities from the state of Arizona seized $100 million. At the time, the Arizona authorities were investigating an unrelated financial fraud and became suspicious when they found that huge sums of money had been shifted between bank accounts in different countries in a rapid fashion. Unbeknownst to Arizona authorities, they had stumbled onto what would become one of the most brazen hedge fund frauds of all time. The extent of the fraud was later confirmed in a several-page suicide note drafted by Marino. He did not commit suicide, but his note pieced together the extent and blatancy of the fraud. The $100 million recovered by Arizona authorities, ultimately in a New Jersey bank account, was all that remained of the investors' money.

What are the lessons here? First, some fund managers are dishonest. A thorough due diligence process may uncover evidence of any dishonesty of the principals of a fund before money is invested. Israel's fund should have failed any standard due diligence checklist if prospective investors had looked carefully and in the right places. Israel's résumé overstated his position and his tenure at a previous firm. Although he claimed to have been a head trader at a large, respected hedge fund from 1992 to 1996, reference checks would have found that he had been an employee for only 17 months (in 1994 and 1995) and had no trading authority at the fund. Although Israel claimed that Bayou started in early 1997, it seems that the fund may have actually started in late 1996. The later start date allowed Israel to conceal substantial losses during the first several months of the fund's operations.

Second, audited information is only as reliable as the auditor. When a hedge fund uses a small outside auditing firm (in this case, one that was unknown outside of its Bayou purpose), red flags should go up. As noted in Chapter 31 on due diligence, contacting and interviewing outside auditors for a hedge fund is a critical step of any due diligence. The only employee of the Richmond-Fairfield accounting firm was Marino, and the only customer was Bayou.

Third, regulation and regulators are not a panacea and cannot be expected to prevent fraud or to discover fraud on a timely basis. Bayou's fraud was discovered eight years after it began and only as part of an unrelated investigation into what regulators thought might be a money-laundering scheme. The conclusion here is that investors must perform thorough due diligence on their investments.

As a final and bizarre postscript to this whole mess, Samuel Israel went missing in 2008, the day before he was to begin serving his 20-year prison sentence. His abandoned car was found near a bridge in New York State. On the dust of his car's hood was written the message “suicide is painless.” He was later captured at a trailer park, where he had been driven by his girlfriend after leaving his car at the bridge. Two years were added to his sentence as a result of this latest deception, leading to a 22-year jail sentence, which he is currently serving.

29.3.2 Bernie Madoff

The investment-related activities of Bernie Madoff have damaged lives and struck fear into the hearts of investors and investment professionals since December 2008. Early that month, Madoff allegedly confessed to running a giant Ponzi scheme. A Ponzi scheme is a fraudulent program that returns deposits to investors and identifies the returned capital as a distribution of profit in order to overstate the profitability of the enterprise and to attract additional and larger deposits. The fictitious profits distributed to investors are actually the capital contributed by new investors to the scheme. A Ponzi scheme requires the continual recruiting of new investors to sustain the fraud. In the end, Madoff reportedly broke down and admitted to his family that his business was “all just one big lie” and basically a giant Ponzi scheme.¹⁰ Along the way, Madoff managed to defraud high-net-worth investors, fund of hedge funds managers, movie producers, movie stars, and university endowments. All told, the scheme was reputed to have grown to $50 billion before being unmasked.

What is amazing about this fraud is that questions, if not outright accusations, had been put forward about Madoff-related investments since 2000. That year, Harry Markopolos, a portfolio manager at Rampart Investment Management, and Neil Chelo, his top assistant, examined the performance numbers of Bernard L. Madoff Investment Securities. They suspected trickery because Madoff's performance rose with uncommonly stable, predictable returns year after year and market cycle after market cycle. The consistency of Madoff's performance seemed too good to be believed.

Markopolos and his assistant studied the strategy supposedly used by Madoff, called an option collar. In an option collar, as introduced in Chapter 6, a manager (1) buys the underlying asset, (2) writes a call option at the higher of two strike prices, and (3) buys a put option at the lower of two strike prices. Both option positions have a neutral exposure inside the range of the two strike prices and hedge the underlier outside of the range. Accordingly, the options hedge the tail risks of the underlier and form a collar, limiting upside and downside, to the aggregated returns of the positions. In perfect markets, this strategy hedges most of the risk of the underlying equity position and should therefore be a risk-reducing and expected-return-reducing strategy relative to long positions without the options. The strategy as described had no realistic source of consistently high returns other than highly inefficient markets. Yet Madoff's positions apparently used somewhat common securities and strategies.

What made Markopolos especially suspicious of Madoff's investment claims was that he had managed a similar strategy but had not produced the consistent positive results that Madoff claimed to have earned. Markopolos went to the Securities and Exchange Commission (SEC) with his concerns. He approached the Boston office of the SEC first in 2000 and then again in 2001. Unfortunately, these initial visits did not lead to SEC action. Frank Casey, a coworker of Markopolos, tried to help by mentioning Madoff's amazing performance to a reporter from MarHedge, a publication that covers the hedge fund industry. Both MarHedge and Barron's subsequently published stories calling into question the remarkable results produced by Bernie Madoff. But these stories did not generate effective regulatory scrutiny. In 2005, Markopolos contacted the SEC's New York office. He sent a 21-page report to the SEC's branch chief explaining why he had concluded that Madoff's business was “the world's largest Ponzi scheme.”¹¹ He continued to send warnings to the SEC in 2006 and 2007, but no action followed. The SEC chairman at the time, Christopher Cox, later stated that he was “gravely concerned by the apparent multiple failures over at least a decade to thoroughly investigate these allegations.”¹² Unfortunately, the grave concern came much too late.

Investors who performed careful due diligence at a firm like Madoff's should have been concerned about several issues. The most important is that the stated returns in terms of both positive mean and minimal volatility were clearly not consistent with the contemporaneous returns of similar strategies being observed in the market. Additionally, the quantity of assets under management and allegedly being deployed with the strategy would have required a trading volume too large relative to the trading activity in the underlying options markets. Another key red flag was that Madoff brokered, cleared, administered, and effectively audited his own fund, creating a lack of external accountability that allowed him to create fictitious accounting statements.

Madoff was arrested by federal agents on December 11, 2008, after being turned in to the authorities by his sons. Madoff is currently serving a 150-year prison sentence. Assets obtained by Madoff and those cooperating with the fraud have been obvious targets of repossession. In a Ponzi scheme or other types of fraud, investors who leave the scheme before it collapses are enriched at the expense of those who remain. Authorities in the Madoff scandal are seeking restitution, meaning restoration of lost funds, from both Madoff and his profitable investors. Courts can order investors who are enriched by a fraud to return the profits as restitution to those who suffered losses, even if the investors were unaware of the investment's fraudulent nature. In December 2010, Mark Madoff, the elder of Madoff's two sons, committed suicide.

What can be learned from the Madoff experience? Most important, investors should ensure that performance numbers are audited by a reputable third-party auditor that provides an independent verification of the check on the stated returns. Returns that sound too good to be true are probably untrue. Also, investors should be especially alert to affinity fraud. Affinity fraud is the commission of fraud against people or entities with which the perpetrator of the fraud shares a common bond, such as race, ethnicity, or religious affiliation. Fraud is especially likely to be committed in the context of such affinities, since the perpetrators target their efforts toward people and groups who would be less likely to be suspicious. Madoff relied on affinities to market his fund, and he portrayed himself as a caring member of the community through public acts, such as philanthropy. Finally, investors should be aware that even if they enter a Ponzi scheme with no clear knowledge of its underlying fraud and exit the scheme before it collapses, the law generally requires return of profits to provide restitution to the eventual victims.

29.3.3 Lancer Group

At one time, Lancer Group was a highly successful hedge fund focused on trades in equity securities using a mix of private and public shares, typically of very small-capitalization companies. The fund was run by Michael Lauer, who had a pedigree that other hedge fund managers dream about. He was a graduate of Columbia University and a six-time member of Institutional Investor's all-star equity analyst team. Over a three-year period, Lauer is reported to have earned management and incentive fees totaling $44 million. The fund raised $613 million in investor capital, and at its peak, Lancer valued the assets at $1.2 billion. Then a collapse came that left the fund with only $70 million.

The story of Lancer raises several interesting issues and highlights important challenges for performing due diligence and avoiding harm as an investor. The key issue is the role of poor valuation procedures. That issue, combined with challenges of poor transparency, conflicts of interest, and loose oversight, allegedly led to huge investment losses and to the downfall of several professionals involved.

Beginning with the issue of transparency, Lancer did not offer transparency to its investors and refused to identify the stocks that were being held in the portfolio. Limited transparency can protect the investors when, as in this case, the fund has large positions in illiquid shares. The hesitancy to reveal the names of the holdings could have been related to legitimate concerns that the actions of other traders could inflict harm on the value of Lancer's stocks if they understood the magnitude of Lancer's positions in each stock. But transparency can also prevent investors from understanding the risks and threats.

The alleged fraud at Lancer also involved issues of valuation and conflicts of interest. Through the partnership agreements, Lauer's investors allowed him the personal discretion to value the restricted shares of these illiquid holdings. This violates one of the cardinal rules of operational risk management: Never allow portfolio managers the ability to price their own positions, especially if their income depends on those valuations. Of course, most hedge fund managers earn large incentive fees on reported profits. The higher the valuation of the fund's securities, the higher the incentive fees earned by the fund. If portfolio managers are allowed to value their own holdings, the temptation to exaggerate performance to earn high incentive fees may overcome naturally honest personalities. A key issue is how Lancer calculated, and allegedly manipulated, the valuation of its portfolio.

Last, there is the issue of window dressing. Window dressing is a term used in the investment industry to denote a variety of legal and illegal strategies to improve the outward appearance of an investment vehicle. For example, some funds might liquidate their holdings of a stock before the end of a reporting period if the stock has generated very bad headlines so that the report does not embarrass the fund managers by having the stock appear in their list of holdings. Other window dressing is clearly illegal, such as manipulating the closing prices of stocks that represent large holdings of a fund so that the fund's total valuations, which are typically calculated at the end of each calendar month using closing prices, are higher. Lancer is alleged to have manipulated market prices to boost reported valuations of fund holdings.

Lancer purchased very large stakes in restricted shares of small companies. Restricted, or unregistered, shares are purchased directly from the issuing company and cannot be sold in public markets for prespecified lengths of time. Typically, unregistered shares are valued at a discount to the market price of the registered shares of the same firm. There is nothing inherently wrong with holding unregistered shares. The allegation against Lauer is that he manipulated the market price of the registered shares by placing trades at key points in time to print high prices in the stock's trading record and thereby justify placing high values on the fund's holdings. Placing transactions to record high or low prices on the transaction records of public markets is a fraudulent activity often termed painting the tape, in reference to the historical use of ticker tape to broadcast prices.

For example, Lauer is said to have purchased 1.7 million unregistered shares in a firm named SMX at 23 cents per share for a total cost of less than $0.5 million. Even though SMX had virtually no operations, Lancer soon valued the holding at nearly $200 million in market capitalization. How could such an enormous jump in value be justified? At the end of one month, Lauer is reported to have purchased 2,800 of the freely floating shares in the market at $19.50 per share, a large premium to the price trading just minutes earlier. Again, later in the year, he purchased 1,000 shares as the last trade of the month, this time at $27. These small trades in the registered, tradable shares were used to value or mark all the shares in the fund at a gain of 8,000% over the purchase price. Not only were the public shares not able to be sold at this price, but also the restricted shares were likely to trade at an even lower price than the registered shares.

Two primary potential safeguards to the problems discussed here are auditing and regulation. In both cases, the safeguards appear to have failed. For example, Lancer's auditor, a major accounting firm, eventually published an audit verifying the dubious valuations. The auditor apparently asked for full appraisals on 10 companies yet received only four. Those four appraisals were written by parties alleged to have had a conflict of interest. Specifically, they may have had an ownership stake or a financial interest in the target companies. In the end, the audit stated that most of the fund value was based on unrealized gains, with prices based on the manager's valuation. The audit did not seem to have questioned the ultimate valuation of the target companies, leaving investors to decide from the auditor's language whether Lauer's valuations were warranted.

Another potential safeguard would have been revelation by regulators of potential irregularities. Lancer allegedly didn't file a Form 13D on 15 stocks in the portfolio for which it held more than 5% of the shares. (The SEC requires a 13D filing whenever an investor owns greater than a 5% stake in a firm.) Had Lancer filed these 13D reports as required, or had the SEC alerted the public to Lancer's failure to file timely reports, careful investors might have been able to realize the substantial illiquidity risk in the fund.

There is no reason to believe that excellent due diligence on the background of Michael Lauer would have raised any warning signals. But closer scrutiny of the auditing of the fund might have alerted careful analysts. One way to reduce the risk of this type of collapse is for investors to demand a private equity style of fee structure, in which incentive fees are paid only on realized gains. Another way is to separate the valuation and portfolio management processes so that the persons valuing the portfolio have no financial interest in the results of the valuation.