Goldman and the charade of honesty
By Chan Akya
Friday's market declines (the Dow down 1.13%, the S&P500 down 1.6%, the FTSE 100 down 1.39%) were broadly blamed on the US Securities and Exchange Commission (SEC) announcing fraud charges against Goldman Sachs over the sale of synthetic collateralized debt obligations (CDOs) to institutional investors. In its announcement on April 16, the SEC makes the following statement:
The Securities and Exchange Commission today charged Goldman, Sachs & Co and one of its vice presidents for defrauding investors by misstating and omitting key facts about a financial product tied to subprime mortgages as the US housing market was beginning to falter.
... The SEC alleges that Goldman Sachs structured and marketed a synthetic CDOs that hinged on the performance of subprime residential mortgage-backed securities (RMBS). Goldman Sachs failed to disclose to investors vital information about the CDO, in particular the role that a major hedge fund played in the portfolio selection process and the fact that the hedge fund had taken a short position against the CDO.
... The SEC alleges that one of the world's largest hedge funds, Paulson & Co, paid Goldman Sachs to structure a transaction in which Paulson & Co could take short positions against mortgage securities chosen by Paulson & Co based on a belief that the securities would experience credit events. 
Perhaps the author knew about the forthcoming charges, but as it turns out, details of the SEC release include the following details:
The SEC's complaint alleges that after participating in the portfolio selection, Paulson & Co effectively shorted the RMBS portfolio it helped select by entering into credit default swaps (CDS) with Goldman Sachs to buy protection on specific layers of the ABACUS capital structure. Given that financial short interest, Paulson & Co had an economic incentive to select RMBS that it expected to experience credit events in the near future. Goldman Sachs did not disclose Paulson & Co's short position or its role in the collateral selection process in the term sheet, flip book, offering memorandum, or other marketing materials provided to investors.
The SEC alleges that Goldman Sachs vice president Fabrice Tourre was principally responsible for ABACUS 2007-AC1. Tourre structured the transaction, prepared the marketing materials, and communicated directly with investors. Tourre allegedly knew of Paulson & Co's undisclosed short interest and role in the collateral selection process. In addition, he misled ACA Capital Holdings, the now defunct bond insurer at the center of the case against Goldman Sachs, into believing that Paulson & Co invested approximately US$200 million in the equity of ABACUS, indicating that Paulson & Co's interests in the collateral selection process were closely aligned with ACA's interests. In reality, however, their interests were sharply conflicting.
According to the SEC's complaint, the deal closed on April 26, 2007, and Paulson & Co paid Goldman Sachs approximately $15 million for structuring and marketing ABACUS. By October 24, 2007, 83% of the RMBS in the ABACUS portfolio had been downgraded and 17% were on negative watch. By January 29, 2008, 99% of the portfolio had been downgraded. Investors in the liabilities of ABACUS are alleged to have lost more than $1 billion.
Royal Bank of Scotland executives were on Friday night examining whether the bank had any grounds for legal action against Goldman Sachs, after it emerged that it lost more than $800m (520 million pounds sterling) as part of an allegedly fraudulent transaction.
As part of the Securities & Exchange Commission's charges against Goldman, the regulator said that RBS had paid Goldman $841m to unwind a position in Abacus 2007-AC1, the Goldman-structured collateralized debt obligation at the center of the SEC investigation.
The part-nationalized British bank had inherited the position as part of its 2007 acquisition of ABN Amro, the Dutch bank. The sum involved is small compared with the ฃ45bn the government later pumped into RBS to keep it alive. But the notion that the alleged fraud may have contributed to an eventual taxpayer-backed rescue of RBS is likely to be politically sensitive.
“On or about August 7 2008, RBS unwound ABN's super senior position in Abacus 2007-AC1 by paying [Goldman] $840,909,090,” the complaint said. Most of this money was subsequently paid by Goldman to Paulson, the hedge fund, the SEC said.
"The product was new and complex but the deception and conflicts are old and simple," said Robert Khuzami, Director of the Division of Enforcement. "Goldman wrongly permitted a client that was betting against the
This isn't an idle matter of a single regulator making an allegation. In Europe, financial regulators operate on the basis of principles rather than rules, thereby widening the ambit of investigations and punitive actions; this part of the SEC statement is likely to immediately attract the attention of most regulators in Europe led by the Financial Supervisory Authority (FSA) of the United Kingdom - one of the most powerful regulators in the world.
It isn't likely that the ABACUS deals are the only ones being reviewed nor that Goldman Sachs is the only investment bank that is charged by the SEC or other regulators as the following statement from the SEC on April 16 makes clear:
Kenneth Lench, chief of the SEC's Structured and New Products Unit, added, "The SEC continues to investigate the practices of investment banks and others involved in the securitization of complex financial products tied to the US housing market as it was beginning to show signs of distress."
Interestingly, on April 9, a week before the SEC charges against Goldman Sachs were filed, the website of Pro Publica (an "independent, non-profit newsroom that produces investigative journalism in the public interest", according to the blurb on its website, www.propublica.org) carried an important story that was devoted to the uncovering of strategies employed by a hedge fund called Magnetar, entitled "The Magnetar Trade: How One Hedge Fund Helped Keep the Bubble Going". 
The article clearly reveals the details of the trade, similar in construction to the ABACUS trades that have earned Goldman Sachs a filing from the SEC. The article highlights:
... the Magnetar Trade worked this way: The hedge fund bought the riskiest portion of a kind of securities known as collateralized debt obligations - CDOs. If housing prices kept rising, this would provide a solid return for many years. But that's not what hedge funds are after. They want outsized gains, the sooner the better, and Magnetar set itself up for a huge win: It placed bets that portions of its own deals would fail.
Along the way, it did something to enhance the chances of that happening, according to several people with direct knowledge of the deals. They say Magnetar pressed to include riskier assets in their CDOs that would make the investments more vulnerable to failure. The hedge fund acknowledges it bet against its own deals but says the majority of its short positions, as they are known on Wall Street, involved similar CDOs that it did not own.
Thus, when specialist hedge funds emerged that purchased the equity portions, it was almost too good to be true for other investors who wanted to own the senior pieces - the so called triple A slices favored by regulators and bankers globally - in these transactions. However, the people buying the equity portions were being rather smart in collecting high income for their investments which was then used to purchase insurance on the less risky portions (triple As).
This trade is the exact mirror image of the Morgan Stanley trades that are shown by Michael Lewis (see my review last week) where Morgan Stanley purchased protection on the lowest-rated tranches that were funded by "long" positions in the higher-rated tranches of these CDOs. 
Clearly, Morgan Stanley and the hedge funds such as Magnetar (as well as Paulson) had opposing views on the "correlation" of the products, that is, the likelihood of higher-rated tranches losing money when lower-rated tranches were wiped out. While the hedge funds bet that the correlation was close to 1, Morgan Stanley relied on its historical risk management models that showed the correlation to be quite low.
In addition, risk-management models use "generic" data; that is, for a wide universe of observable deals, rather than for more
narrowly selected transactions. When deals are constructed with a high proportion of dangerous mortgages, it then becomes more likely that both higher-rated and lower-rated tranches will default, keeping in mind that rating agencies would have been way too clueless to spot the difference between a poorly constructed deal and a well-constructed one.
The differences in the economics at hand are more clearly laid out by Pro Publica, as the following portion of the article reveals:
An independent analysis commissioned by Pro Publica shows that these deals defaulted faster and at a higher rate compared to other similar CDOs. According to the analysis, 96 percent of the Magnetar deals were in default by the end of 2008, compared with 68 percent for comparable CDOs ... From what we've learned, there was nothing illegal in what Magnetar did; it was playing by the rules in place at the time. And the hedge fund didn't cause the housing bubble or the financial crisis. But the Magnetar Trade does illustrate the perverse incentives and reckless behavior that characterized the last days of the boom.The point being of course that the deals were gamed to decay faster than the poorly constructed deals of Wall Street - the relevant portion of the Pro Publica analysis is shown in the table below. The table highlights that whereas the "average" Wall Street deal defaulted in two years (when constructed in 2006, ie the "2006 vintage"), the deals of Magnetar collapsed in 1.38 years, that is, 17 months. In 2007, the deals constructed by Magnetar blew up in 10 months, while the average wall Street deal took almost 14 months. Given that, let's be clear here: for a triple A deal to be blowing up within 24 months is virtually unprecedented anyway; therefore the "crime" if any of Magnetar doesn't look egregious in that light. With or without Magnetar, the deals would have collapsed.
Then there is the issue of complicity in the transactions - the subject of the SEC allegations against Goldman Sachs on Friday. The Pro Publica report on the Magnetar trade has the following gem of a paragraph:
At least nine banks helped Magnetar hatch deals. Merrill Lynch, Citigroup and UBS all did multiple deals with Magnetar. JPMorgan Chase, often lauded for having avoided the worst of the CDO craze, actually ended up doing one of the riskiest deals with Magnetar, in May 2007, nearly a year after housing prices started to decline. According to marketing material and prospectuses, the banks didn't disclose to CDO investors the role Magnetar played.
The last sentence should have a chilling effect on the management of the various banks involved - this is precisely the same allegation that the SEC has made about Goldman Sachs. For whatever it is worth, the investigation quantified the deals made by Magnetar - a staggering 26 of them accounting for around US$37 billion of notional value, which were distributed by various investment banks. The adjoining table, which was also published by the investigative journalists, shows the distribution of all known deals.
When considered from the perspective of the various investment banks involved, it is clear that there were no "winners" here. The top five investment banks by value of deals done for Magnetar are as shown in the table below.
As luck or otherwise would have it, all five of these investment banks were adversely affected by the financial crisis in a maneuver known popularly as "eating as your own cooking". Citigroup had to be bailed out by the US government and UBS by the Swiss National Bank, while Merrill Lynch sold itself to Bank of America to stave off an impending bankruptcy. Calyon, which has recently been renamed "Credit Agricole CIB" also suffered massive losses, and Lehman Brothers of course went spectacularly bust in 2008.
In many of these banks, the existence of inter-departmental friction could have well led to the comical situations wherein toxic deals manufactured by one side of the bank ended up in another area. Even when bankers know that deals have been "gamed" to benefit certain hedge fund clients, it is also a good bet that internal disclosure of such practices would be minimal, not the least because the narrow interests of the bankers (being the annual bonus check) almost contradict the interests of the bank (longer-term profits and strong client relationships).
When bankers know that the deals they created are destined to blow up, they would quietly collect their bonus checks and then go to work somewhere else. Even when they know that the deals being purchased by another part of the bank are exactly the same as the very deals that they have created to be blown up, such bankers would keep mum - indeed the perverse incentive would be to gently egg on the idiots in the Treasury department even as the investment banking group rakes in the profits and fees.
Of the banks that did NOT blow up subsequently, JPMorgan appears to have lost a significant portion of the deal done with Magnetar - the article claims that the bank lost $880 million of the deal it did for Magnetar on the portion it couldn't sell and left on its own books. When a bank makes fees of $20 million, and then proceeds to lose $880 million on the deal while paying the bankers involved in putting the deal together at least $10 million for their "genius", what you have is as clear a failure of risk management as can ever be made visible. The notion that JPMorgan, or indeed any other bank, walked on water and had superior insights into the crisis is belied by the latest details of what actually went on.
(Curiously, this notion of JPMorgan as a bank that evaded the crisis was the very problem I had with Gillian Tett's book, Fool's Gold - see Crisis hindsight, Asia Times Online, Aug 15, 2009.)
The other bank of note in the list made public is Deutsche Bank, which led a single deal of around $1.6 billion. This is a bank that generally avoided big losses on the financial crisis, if for nothing due to the trades of Greg Lippmann that were disclosed in the Michael Lewis book I reviewed last week. If it now appears that Deutsche also helped to fuel itself by selling deals such as Magnetar to (literally) unwitting investors, the picture at least from a regulatory side could change quite quickly.
That we have an active group of investors, regulators, politicians and others waiting to organize lynch mobs to deal with the people considered most responsible for the crisis - bankers and assorted villains from the financial industry - has been clear for some time now. With important elections across the continent in May, expect the situation to get worse for investment banks in general and Goldman Sachs in particular. Over the weekend, European governments were baying for blood, with British Prime Minster Gordon Brown threatening to ban bank bonus payments.
Whether it is the regular charades in US government hearings or the pronouncements in the global conclaves of governments such as the Group of 7, the Group of 20 and other meetings, there has been a search for an issue that could provide the key catalyst with which to push financial reforms and new regulations through while punishing those responsible for the most egregious acts in the months and years preceding the crisis.
The most important roadblock to these measures has been the hitherto well-held assumptions that all participants in the financial markets were "adults"; in other words, that since both buyers and sellers were market professionals there were really only a few true victims from the crisis. This notion holds that whether it was asset-backed securities or the mountain of leveraged loans that were issued, participants were willing to enter into the transaction, therefore making the issue an issue for professionals to deal with each other rather than depend on the work of regulators and governments to step in with a view to stemming a non-existent rot.
All that has now been thrown out of the window with the revelations that the key middlemen in transactions had more than enlightened self-interest in view when constructing the deals that made them millions immediately and lost billions to whoever touched them just a few months later. This view has been intuitive, but without evidence thus far. That has indeed changed now.
Anyone who considered the financial crisis to be virtually finished would now have to contend with the regulatory risks for owning bank shares. The spate of regulatory and investor lawsuits to follow from the current revelations by the SEC and independent publications likely highlight the real risks of owning shares in these companies - that the ghost of Christmases past proves too much of a regular visitor in coming months.
1. For the Securities and Exchange Commission announcement see here.
2. For the Pro Publica website, see here.
3. A typical CDO is bought by various investors including Paulson for the equity piece and the usual suspects of European banks for the highly-rated parts. Goldman Sachs will arrange the CDO based on a number of actual mortgage-backed bonds. Overall, in such a trade, there is no position left with Goldman Sachs, and the likes of Paulson would be net long the risky (low-rated) portions. In a "synthetic" CDO, there are no actual bonds and instead all that happens is that credit derivative contracts are written which reference the performance of certain types of mortgage pools. While Paulson goes long the equity portion and various European banks or US insurance companies go long the higher rated portions, Paulson also use the coupon to buy protection on the overall CDO from Goldman Sachs. So for a $1bn CDO, the equity portion can only be $50 million (or 5%), which is the long position earning a coupon of say 30% or $1.5 million a year. This $1.5 million per annum can be recycled at a rate of say 1% to purchase protection on say $150 million of the senior debt tranches - and that leaves Paulson net SHORT of $100 million (in actuality, Paulson was short the entire $1 billion). But that creates a new problem; namely that Goldman Sachs is now net long whatever Paulson is short - and that is handled by Goldman Sachs purchasing insurance (CDS) from a market counterparty, most likely AIG (thereby leaving the insurer net long the entire portion that Paulson is short. For added measure, being a 'synthetic' CDO, Goldman could also execute side trades with AIG thereby leaving the insurer net long more than what Paulson was short, with the difference being the net short position of Goldman Sachs). That is the reason the case will become intensely political in the US - Goldman Sachs alums virtually forced the US government to take over and rescue AIG precisely because such payments from AIG to Goldman Sachs that mirrored the payments Goldman Sachs had to make to Paulson would have been interrupted and left the investment bank in the lurch for large losses....
Pandora’s Box is Open and Woe to the FinancialsApril , 2010
By David Goldman
This is on the scale of the Enron case. Except that it could affect every firm involved in subprime securitization.
I haven’t read anything in years as racy as the SEC’s complaint against Goldman Sachs for collusion with the Paulson hedge fund to cheat subprime investors out of $1 billion. A billion-dollar fraud is not a small matter. I’m no lawyer, but the granularity of detail and documentation that the SEC has assembled appears extremely impressive. They have nailed a 32-year-old Goldman vice president who cobbled together the tainted structure, and he appears to be singing. Investment banks aren’t like the mafia. No-one takes twenty-year sentences and keeps their mouth shut. This case very well might open up others.
The issue, as the press has reported, is the selection of collateral in a Goldman Sachs synthetic CDO deal. As the SEC reports:
In late 2006 and early 2007, Paulson performed an analysis of recent-vintage
Triple B RMBS and identified over 100 bonds it expected to experience credit events in the
near future. Paulson’s selection criteria favored RMBS that included a high percentage of
adjustable rate mortgages, relatively low borrower FICO scores, and a high concentration of
mortgages in states like Arizona, California, Florida and Nevada that had recently experienced
high rates of home price appreciation. Paulson informed GS&Co that it wanted the reference
portfolio for the contemplated transaction to include the RMBS it identified or bonds with