The Best Thing I Have Read on SEC-Goldman (So Far)

By James Kwak

Actually, two things, both by Steve Randy Waldman.

Part of Goldman’s defense is that it was in the nature of CDOs for there to be a long side and a short side, and the investors on the long side (the ones who bought the bonds issued by the CDO) must have known that there was a short side, and hence there was no need to disclose Paulson’s involvement. Waldman completely dismantles this argument, starting with a point so simple that most of us missed it: a CDO is just a way of repackaging other bonds (residential mortgage-backed securities, in this sense), so it doesn’t necessarily have a short investor any more than a simple corporate bond or a share of stock does. Since a synthetic CDO by construction mimics the characteristics of a non-synthetic CDO, the same thing holds. (While the credit default swaps that go into constructing the synthetic CDO have long and short sides, the CDO itself doesn’t have to.) Here’s the conclusion:

“Investors in Goldman’s deal reasonably thought that they were buying a portfolio that had been carefully selected by a reputable manager whose sole interest lay in optimizing the performance of the CDO. They no more thought they were trading ‘against’ short investors than investors in IBM or Treasury bonds do. In violation of these reasonable expectations, Goldman arranged that a party whose interests were diametrically opposed to those of investors would have significant influence over the selection of the portfolio. Goldman misrepresented that party’s role to the manager and failed to disclose the conflict of interest to investors.”

Waldman follows this up with an analysis of the premium that Goldman extracted from the buy-side investors and transferred to Paulson (in exchange for its own fee). The point here is that Goldman could have simply put Paulson and the buy-side investors together and had Paulson buy CDS on RMBS directly — but that would have affected the price of the deal, because Paulson wanted to take a big short position. So instead, they created the CDO (a new entity) and then drummed up buyers for it, in order to avoid moving the market against Paulson. The advantage of thinking about it this way is it shows what the function of a market maker is and how that differs from the role Goldman played in this transaction.

The posts are long, so sit back and enjoy.

Update: Nemo points out that I misinterpreted Waldman’s post, and Nemo is right, although I think I got the substance of Waldman’s point right. Here is what Waldman says:

“There is always a payer and a payee, and the payee is ‘long’ certain states of the world while the payer is short. When you buy a share of IBM, you are long IBM and the firm itself has a short position. Does that mean, when you purchase IBM, you are taking sides in a disagreement with IBM, with IBM betting that it will collapse and never pay a dividend while you bet it will succeed and be forced to pay? No, of course not. There are many, many occasions when the interests of long investors and the interests of short investors are fully aligned. When IBM issues new shares, all of its stakeholders — preexisting shareholders, managers, employees — hope that IBM will succeed, and may have no disagreement whatsoever on its prospects. . . . The existence of a long side and a short side need imply no disagreement whatsoever.”

So I was clearly wrong when I said, “a CDO is just a way of repackaging other bonds (residential mortgage-backed securities, in this sense), so it doesn’t necessarily have a short investor any more than a simple corporate bond or a share of stock does.”

But — and I don’t think I’m engaging in sophistry here — Waldman’s underlying point is that even though there is a short position, that doesn’t mean that the long and short investors have diametrically opposed interests. That’s true of stocks, and it’s also true of CDOs. And so it’s disingenuous of Goldman to imply that buyers of any CDO always know that there is someone who is actively betting on it to go down in value.

85 responses to “The Best Thing I Have Read on SEC-Goldman (So Far)

  1. Maybe. What I find factually interesting about Goldman’s purported long position is that it does not square at all with the NYT reportage of the deal, FabFab’s work for Goldman, and Goldman’s general stance on the housing market at the time the deal was consummated. To sum it up, FabFab and his colleague Jonathan Egol at Goldman were part of a bearish clique and had been since about 2005. They were in the minority until top management took a hard look at the mortgage business as a whole beginning in late 2006. At that point the word came down after some contentious meetings to lay off and/or hedge the net long exposure. It seems a bit hard to believe that at a time when Goldman was adding bearish bets to attain a more neutral or hedged position in this market that a trader/salesman who had been bearish on housing for at least two years would put together a deal adding $75MM in net long exposure to RMBS. Was it worth that much to get Paulson’s business?

  2. I am not defending Goldman’s role in this, but why are people not also criticising the buyers of these CDOs? They are large sophisticated banks who should have done their homework and analysed the underlying bonds and if they are negligent (or stupid enough) not to do their homework before ploughing millions of dollar in a deal then tough!!

  3. Funny how everyone who reads Steve’s piece takes away a slightly different message.

    OK, first, let’s try an analogy.

    CDO : synthetic CDO :: stock : option

    Personally, I buy the argument that there is a difference between stocks (pieces of equity in a firm) and options (pure bets between two parties).

    But Steve makes exactly the opposite claim to what you say; specifically, he argues that ordinary stocks and bonds do, in fact, have an implicit “short” on the other side of the transaction.

    In the big picture, however, this is all irrelevant, because it has nothing to do with the SEC’s allegations. Goldman created the CDO at the behest of the short seller and then let him construct the portfolio, and then tried to pass it off as having been constructed by someone else entirely. Whether it rises to the legal definition of “fraud”, I do not know; but it definitely rises to the Nemo definition of “scumbags”.

  4. Hung,
    People are not criticizing it because being greedy for yield on the long side and ignoring the inherents risk (or greedy for a bigger house than you can’t afford without a stated income mortgage, or greedy for the fancy vacation or Escalade that you can’t afford without a home equity loan) is not something to be criticized. These people are victims, they are to be pitied, and they don’t deserve to bear the consequences of their foolish decisions! They were all taken advantage of by the sharp elbowed types on Wall Street and were simply lambs let to slaughter with no culpability for their actions. Hung, where have you been man, haven’t you been reading the paper for the last two years?

  5. James,
    I agree that Randy’s piece about the premium that Paulson avoided paying was probably the most thoughtful thing written about the SEC charges so far (not to take anything away from Randy, but he didn’t have to try very hard, there has been precious little sober and thoughtful analysis.)

    So let’s say that Randy has the right economic analysis about the disclosure issues involved – which I think he might. Hard to say exactly what the premium would be, but given that Paulson was not seen as the Warren Buffett of mortgage derivatives in those days, the premium would probably not be all that large. I believe it has been reported that ACA already owned, in other vehicles, quite a few of the reference securities so I doubt their view of the risk would have changed dramatically. Maybe Paulson would have paid a premium of few percent if his intention was fully disclosed.

    Given that, don’t you find it a bit odd that the SEC effectively decides to go nuclear against Goldman with a case in which the economic losses due to the fraud are hard to calculate with any precision and which are really pretty small potatoes given much bigger frauds that are have been committed against investors much less capable than these (say all the mom and pop Lehman shareholders, for example)? And the SEC decides it is going to blindside Goldman with a press release rather than seeking a settlement out of court first – on a case that most feel is not a slam dunk.

    This is what I find most troubling about this case, and why to me this has the deep stench of a politically motivated prosecution. There are plenty of worthy culprits and plenty of clearly fraudulent (ie true slam dunks) deals to choose from in the 05-07 period. To chose this one as the SEC’s “come back” pitch is very curious indeed.

  6. Hung,

    Here’s a case in point. The NYT is doing another one of its tut-tutting stories about how Wall Street poured gasoline on the fire of the housing bubble.

    And they start off with 23 paragraphs about all the fees that Wall Street made from putting together CDOs before you get to this sentence buried so deep into it that it is almost an afterthought:

    “At the height of the housing boom, as interest rates dropped on competing investments like corporate bonds, big institutional investors like pension funds were tempted by these C.D.O.’s, which offered yields two or three times comparable investments but were marketed as “ultrasafe.””

    One of the first things you are taught on the buy side of the investment business, is that if something looks too good to be true, it probably is. Apparently, all the people buying this paper swallowed the line that they could make three times as much on these new fangled, “supersafe” CDOs as they could buying Treasuries or something else rated AAA. Were they really that credulous, or did they suspect that not everything was completely kosher but did it anyway because they could sell it to their boss or regulator or client, and, oh by the way, get a bigger bonus for beating whatever benchmark they were measured against with the extra yield from the CDOs? Of course they were taking far more risk than they people who really did their homework and bought the safe stuff, but who really cared – it was a bull market, it was a bubbble, everyone was making money. What could go wrong?

    Do these people sound like victims to you? I’ll bet there is a letter in the file somewhere at IKB justifying its investment because of the AAA rating. And I’ll bet there are a few emails on their servers from an honest analyst or two, lower down the food chain not doubt, questioning the ratings on these bonds. Those are things you are unlikely to ever hear about from the SEC or the media.

  7. It’s kind of like a casino always has the odds in its favor, otherwise, no casino, but it’s as if the casino decided to set up a side-by-side betting parlor betting on the success or failure of those in the actual casino, and then mixed the players from several tables with known odds based on the gamblers respective winning/losing records and set up the bets knowing the outcomes ahead, while have a third party in the side betting parlor be the only one receiving the information setting up the side bets. We know that casinos rig the odds anyway, but if that third party is part of the scheme that is straight out fraud, and the law would close that casino for good if they uncovered the scam.

    With the multiple layers beyond the original bonds, we have to believe, beyond a reasonable doubt, that the large investment banks not only had the opportunity, but the incentive (especially for the rank and file bonus earners) to set up the various derivatives not only to favor themselves, but most of all those investors/institutions who were regular “high rollers” at their casinos. As this scheme is gradually unravelled by the SEC and FBI by the information gathered during the subpoena of records process, we are likely to see the immense scale of corruption taking place, especially as the pickings in the mortgage markets tightened and extreme measures were put into place, say for about the last three years leading up to the collapse.

  8. Since a synthetic CDO is not real, and thus has no cash flows, is it not true that the only long position in one, in effect, is to be the seller of CDS on them? Hence, the long party would have to know they are participating in a speculative transaction.

    This is not to say that Goldman still couldn’t have deceived the CDS sellers by not disclosing Paulson’s participation in assembling the portfolio, and thus stacking the deck against the deal.

  9. MJPedroliEsq@gmail

    Thanks James. That’d food for though as everyone attempts to remember the always omnipresent concept of tort law and what it means under these sets of facts. It’s appears to be a clear omission, if not a racket to omit, which is another matter, information which a rational person would consider material to their purchase. To me, open and shut.

    I just finished reading 13 Bankers, excellent! Thank you. You had me at Jefferson. It was good to have that historical perspective so often lost. We are deep in the position Jefferson and Jackson warned, needing to punish yet be simultaneously reliant on the same banks. To extract from this toxic relationship will take many years, so catch your breath and stick to it!

  10. MJPedroliEsq@gmail

    And James, I wanted to email you and Simon regarding an initiative I’m working on. If I can’t find your email contact, feel free to touch base via my address to the right. I’ll hopefully be meeting with Bill Black and I’d like to talk to you beforehand, if not include you in on the discussion.

    Pls advise..

  11. It would be interesting if William Black was asked to testify for the SEC in the Abacus case.

    IIRC, Black specializes in a specific form of fraud based on his investigation and prosecutions in the S&L debacle. IIRC, this form of fraud is perpetrated top down from the most senior levels of management. It results in a form of systemic capture where the activities of entire organizations are aligned for specific fraudulent purposes, under the guise of legitimate business activity.

  12. Maybe the dealers on the buy side were collecting fees and had no skin in the game.

  13. correction:

    Maybe the dealer on the buy side (like the rating agencies) were collecting fees and had no skin in the game.

  14. virtualobserver

    perhaps if banks fail to provide adequate security, I should also be allowed to rob them at gunpoint.

  15. Perhaps you could elaborate on what ‘carefully selected by a reputable manager’ entails.

    I’m starting to get creeped out about the notion of investment bankers looking through random credit reports of folks who purchased/refinanced homes in Reno, Baja, and Sun City.

  16. Thanks for the links.

    On a related issue: it seems to me, trading activities such as are detailed in the GS case, should never lead to increased social costs. Not only, in my view, does socializing the costs of such trading activities negate their social virtue (price discovery becomes state sponsored price enforcement) such support drifts ever closer to tax funded gambling. What’s next, tax credits for the losers in next year’s World Series of Poker?

    When a security can be designed to fail, it’s time to separate the financial markets into socially protected and unprotected divisions.

    It’s time for the Volcker Rules.

  17. The more I read about this the more amazed I am that the critics do not begin to understand what they are talking about. The CDO in this case was a security only in the sense that it was not an insurance policy because ISDA made sure of that through political influence. But there was no way an ‘investor’ could have ‘bought’ this CDO, because its constituent parts were owned by an array of other investors who had nothing whatsoever to do with the deal. So, what was going on here? The investors were ‘selling’ ‘protection’ on a bundle of mortgages chosen by somebody, and the only way the deal could happen is if somebody else was ‘buying’ the protection and paying a premium, since it was the premium which provided the ‘investors’ with the return they were chasing.

    Now, did it matter that the bundle had been selected by the protection buyer? Clearly, from the viewpoint of the protection seller, that would have been nice to know. Was the failure to identify the protection buyer fraud, even of the watered down SEC press release variety? It seems to me that the fraud here (if any) was perpetrated by the rating agency, which made the whole game possible. It is understandable that the public doesn’t understand this. Even commentators for the most part do not understand what this CDO was. But the guys at the SEC certainly do understand it, which means the most likely reason for this case is political grandstanding in an election year. Look for a settlement in which one or to GS minions fall on their swords, the company pays a trivial fine, and Obama crows about cracking down on Wall Street.

    Meanwhile, the CDS game goes on unabated, the bubble continues to inflate, the Democrats hope against hope they can postpone the next crash to 2013.

  18. I thought Waldman’s analysis was so simple and insightful that even a doofus like me could understand.
    I think his arguement also was so good because he took on the arguement that this was just a simple lack of disclosure, and even if purposeful, that any good buyer could have figured it out for themselves.
    Its one thing to sell crappy used cars. Its another to assemble a bunch of extra crappy used cars, and than short the company that sells used car maintenace insurance, while telling the used car seller and the used car maintenace insurance company that this is a normal distribution of used cars.

  19. Language the voter understands even if from a Google Geek

    Shocking Fraud from Financial Scum : Good Math, Bad Math –

    To call this sociopathic behavior on the part of the scum involved is an understatement of absolutely epic proportions. What the people involved in this did is no different that just stealing money…To call this sociopathic behavior on the part of the scum involved is an understatement of absolutely epic proportions. What the people involved in this did is no different that just stealing money from the people who they suckered into buying worthless CDOs. They deliberately sold worthless garbage – things that they didn’t just know were going to lose money, but that they designed to lose money – lying to the buyers, telling them that it was a great safe investment, knowing that the buyers were going to lose everything. But they didn’t care – because doing that would put money into their own pockets.

  20. Language the voter understands even if from a Google Geek – GOP take note.

    Shocking Fraud from Financial Scum : Good Math, Bad Math –

    To call this sociopathic behavior on the part of the scum involved is an understatement of absolutely epic proportions. What the people involved in this did is no different that just stealing money…To call this sociopathic behavior on the part of the scum involved is an understatement of absolutely epic proportions. What the people involved in this did is no different that just stealing money from the people who they suckered into buying worthless CDOs. They deliberately sold worthless garbage – things that they didn’t just know were going to lose money, but that they designed to lose money – lying to the buyers, telling them that it was a great safe investment, knowing that the buyers were going to lose everything. But they didn’t care – because doing that would put money into their own pockets.

  21. priscianus jr

    Can someone explain to me why it isn’t inherently fraudulent to offer something for sale that is designed to fail? And why it is not even more inherently fraudulent for a bond rating agency to rate the said piece of ¢rap AAA?

  22. This was a synthetic CDO, meaning that it was full of nothing but CDS on RMBS (and maybe other assets). CDO buyers must have known that they were taking the long side of a two-way bet (albeit, one of the last to have to pay out in case of a negative credit event). However, it was not disclosed to them that the short side of the bet was taken entirely by one entity (Paulson’s hedge fund). Furthermore, it was not disclosed that the same entity was long the equity tranche of the CDO, effectively netting out his position in the most expensive and lowest yielding portion of the CDS (because it was most likely to experience negative credit events); so the Paulson hedge fund was effectively only betting against the tranche held by the investors (plus the super-senior stuff that Goldman retained).

    The position would be fundamentally equivalent if Paulson bought the equity, then purchased CDS from Goldman on the other tranches, including the super-senior. For Goldman to sell CDS on the super-senior tranche, and hold onto both, would double their exposure to the eventual collapse of the CDO. Before the collapse, they would get double the income (or more — they could theoretically price the CDS on the super-senior any way they wanted to, since it was unrated). Still, this would be a stupid thing to do, and we don’t think Goldman was stupid (not that stupid, anyway). So the arrangement in the paragraph above makes more sense, while also convincing Goldman that they didn’t really need to disclose Paulson’s role.

  23. Couple of questions
    Can someone please explain “designed to fail”, especially within the context of 1000s of other maket products? And I dont think that Paulson believes the structure failed… in fact it worked exactly as designed.

    Does IKB have any responsibility to perform its own diligence? Same for ACA? Isnt buying on the recommendation of an outside party a dereliction of duty?

    Is the fact that Paulson bet against the trade anymore of a disclosure than ACA bet in favor of it? If so, why – because Paulson was right?

    Is the fact that Paulson bet against the trade more of a required disclosure than the fact that GS itself bet against the trade. Someone was long the protection, the document clearly shows GS as initial protection buyer and the one facing the trust. Did IKB/ACA take into account that the single most profitable bank on WS was “betting” against them and would have incentive to “design to fail”?

    Oh,,, this could go on all day. A bunch of people who have never seen sausage made are seeing it for the first time. Welcome.

  24. The SEC doesn’t contest that there were multiple meetings where Paulson, Goldman and ACA are all there discussing the structure. Could ACA really have not understood that Paulson was shorting it?

    The SEC has testimony that Paulson not only said he was shorting the deal, but even explained how it selected the reference securities.

    But according to Simon, Paulson should be in jail for telling every one who would listen that mez CDOs were the stupidest thing ever and should never be AAA. Whatever.

  25. Exactly!!!
    What amazes me is that there is really nothing complicated or complex about these deals. Certainly convoluted but not complicated.
    Perhaps CDS should be listed in Caesars Palace. But then the odds and cuts are too transparent (that is why Wall Street wants to keep these as over the counter products).

  26. Last question

    Why the CDO structure? Who initiated that, why? Who required it, why? What were the incentives for each party, including buyer and seller, to use a CDO design instead of a CDS?

  27. Let’s take a long step back from the trees so we can see not only the forest but maybe also the tectonic plate.

    This GS deal once again exposes an economic system and market in which the prime actors own little or nothing and the owners have no influence.

    The managers and banks who bought the long side of all this CDO crap are only motivated to stay well inside the herd. As we all know, lose 50% of the fund when everyone else loses 50% and your job is fine, but earn 4% when others are earning 7% then adios bonus and job. Studies have shown this is true even if you saved the fund lots of money in a previous down market.

    So while this may be hard to believe, the longs in this junk do not share the interests of pensioner, saver, or other actual investor. GS and the other smart boys know this and like any good product developer create the product for the needs of the buyer. And the buyer knows that his/her peers at other institutions have been buying this crap and that they also have to buy it or get out of the business. If market tanks, who cares, everyone lost and they all keep their (cushy) jobs.

    This overriding reality is certainly not unique to the market for exotic derivatives. The real ownership of our corporate economy has no say in the operation of the companies and thus no way of influencing management to align with their needs. The cliche about management being aligned with the interests of the owners is bull – they are at best aligned with the interests of the owners’ managers. Even when a company is taken private the ultimate owners are the poor saps who have their savings in a pension fund while management aligns its interests with the managers of the buyout firm, who are uber-whores.

    So the definition of ownership in this economy has been perverted, and the benefits of ownership usurped. The nominal owners of our corporations have virtually no say in the operation of the funds that hold their savings and less than no say in the operations of the investments made with their money.

    But further, small businesses have been systematically eliminated by the national and international entities that have taken over everything. Today’s image of the small business person is the franchise owner. Yet the franchise model is by and large a way of getting employees to provide financing in exchange for a job.

    And, of course, for too many home ownership became like renting except you have to pay taxes and repairs.

    So you have to laugh at the free-marketeers. Few of the decisions in our markets and economy are made by the owners and real purchasers of assets. Its time for a new economics which analyses markets in which everyone making decisions is an agent or a salesman.

  28. Simon left reason, facts and the rule of law behind a few days ago and moved on to a pogram and jihad against Wall Street. Even the SEC prosecutor said there was no reason to charge Paulson with anything. Simon in his omniscience apparently knows more than the SEC.

  29. A) “a CDO is just a way of repackaging other bonds (residential mortgage-backed securities, in this sense), so it doesn’t necessarily have a short investor any more than a simple corporate bond or a share of stock does.”

    B) “Since a synthetic CDO by construction mimics the characteristics of a non-synthetic CDO, the same thing holds. (While the credit default swaps that go into constructing the synthetic CDO have long and short sides, the CDO itself doesn’t have to.)”

    Part A is correct.

    Part B is completely incorrect. Any synthetic only mimics the cash security in terms of the economics of performance on one side of the transaction. When you sell a stock, bond or cash CDO, you have no further economic interest. You could care less about the price/performance of the security except for lost opportunity. In a synthetic transaction the seller remains economically interested in the performance of the security because ultimately the buyers economics come from the seller.

    It is a pretty strange logical leap to say that all the underlying CDS’s have a long / short side, but in aggregate they do not.

    Statements like this are a gross misunderstanding of the basic financial structure of these securities, misleading and do a disservice in the discussion of the situation.

  30. Here we go again: the Goldman/White House revolving door is still operational.

  31. JK> a CDO is just a way of repackaging other
    JK> bonds (residential mortgage-backed securities, in
    JK> this sense), so it doesn’t necessarily have a
    JK> short investor any more than a simple corporate
    JK> bond or a share of stock does

    James, this is absurdly wrong. Of course the short side in a synthetic CDO is needed!

    You should ask yourself if you believe that the cash flow is not coming from a needed short, where it actually does derive from.

    In a synthetic, there are no underlying assets producing cash flow (i.e. ACA was not buying cash flow generated from the mortgages but a derivative of those mortgages, nor in the case of the horrible IBM stock analogy is there a share in a profitable business at stake).

    So, if you can understand the above point, ergo it must follow that however complex or simple you make the structure, the cash flow is coming from another investor on the other side of the transaction.

    No matter how many words and poor analogies are made to stocks and bonds, a synthetic CDO is radically different, and the cash flow comes from the short. There is never a case when this is not true.

    The fact that Steve could have written this and James could have copied it is thoroughly amazing.


  32. JC writes: “Now, did it matter that the bundle had been selected by the protection buyer?”

    Generally no.

    People buy insurance all the time on things they select and purchase, (eg, cars, buildings, homes). The allegation in Abacus deal is Paulson bought “protection” for a bundle he set up to be defective.

    I am still trying to figure out what a synthetic CDO is. Something about outright gambling.

    If SEC win, perhaps, it will set a precedent in common law that a buyer cannot collect “insurance” on a deal the buyer designed to fail.

  33. What IKB couldn’t take into account, because it wasn’t disclosed, was that the counterparty was short only the tranche that IKB bought, having netted out all the other short positions (see my post above).

    Whether that omission was material or not is the question for the courts to resolve. You may think it wasn’t material, but sadly for Goldman, you are not hearing the case.

    In reality, you are correct about IKB skimping on their due diligence. It is likely that if, on page 463 of the CDO terms, Paulson’s role had been spelled out, along with candid prediction of the impending and complete collapse of the investment, IKB would still have made the purchase — for all the reasons presented in these comments (herd mentality, short-term thinking, etc).

  34. Sorry for the relatively silly question but: Is there, like, a super-vilain in all this? One person who sold his vision to everyone or somehow used his Army of Darkness to blackmail GS et al. to do his evil bidding and literally rule the world? I mean serioulsy, this whole thing is way crazier than anything fiction has ever come up with… One understands why super heroes became a huge part of American culture post 1929… It feels like we’re up against forces that mere mortals cannot win against.

  35. Would it be misguided to suggest Hank Paulson.. Just a hunch.

  36. A CDO is a financial product the value of which depends upon other financial products. In this case Abacus was a Credit Default Swap based upon a bundle of existing mortgages all of which were owned by parties with no involvement in the deal. The protection seller (IKB) puts cash into an SIV. The protection buyer (Paulson) pays an annual premium and gets to tap the SIV if certain triggering events occur with respect to the underlying mortgages. The deal lasts for 5-7 years (or until the protection seller’s cash disappears). Why would IKB (or anyone else) do this? Because the premium was high relative to the RATING on the deal. IKB thought it was making a good bet. Another aspect of the CDS market is that these things trade. Protection sellers expect that they can exit the deals because Goldman (and others) are market makers. Possibly, IKB thought this deal was temporarily mispriced and was ripe for a trade.

    Understanding all this is a precondition to understanding why the very existence of this market is insane, particularly when there are several hundred TRILLION of these CDS contracts outstanding.

    Why would a major bank involve itself in these contracts? Because it is possible to generate huge profits (supporting huge bonuses) by making assumptions about their value. The assumptions may be absurd, but they are wrapped around advanced mathematics which gives the whole thing an imprimatur of rationality. In this way, the CDS market is similar to modern economics, which supports a large cadre of academic charlatans hiding bogus assumptions behind advanced calculus while persuing toadying objectives.

    Another reason for bank participation is Basel II, which assigns lower capital requirements to this drek than to actual loans. Why would a bank make an actual loan when it can place a bet at far lower costs and book a nice profit, producing a nice bonus, at inception.

    Now you understand financial innovation. What do you think?

  37. “Has no cash flow” and hence the need to mark-to-model that suggests indeterminate randomness. Structural errors of omission occured because of regulatory deterministic biases. See page 78 of of “We’re All Screwed: How Toxic Regulation Will Crush the Free Market System.”

  38. As I understand it, GS ended up with the position after trying to sell it and failing. It isn’t as though they thought it was a good deal so they went long. The market forced them to eat their own crap. They may not have had enough time to move it off books into a SIV like many others did when they ended up with dangerous stuff they couldn’t sell.

  39. You are correct.

    More to the point, Paulson bought CDS which has to have a long and a short like a futures contract.

  40. A Global Strategist

    I know the guys who were at ACA, and I know the SOP of CDO shops at the time.

    The ACA guys were presented with a list of CDS that Goldman told them were “available” The ACA guys had to know there were shorts on the other side, because CDS can’t be created without “the other side”

    They then compared the availables to their “approved” list. 55-60 came out as overlap – formed the basic bldg blocks. They then asked about other names. This is where it gets strange. GS put Paulson in front of them. ACA thought that Paulson was reviewing/approving the “new” names from the long side as an equity investor. As we know now, they were the short side of the CDS, and were not going to buy equity, but ACA was under the impression (from whom?) that Paulson was a potential equity buyer. Paulson declined the “better” Wells Fargo credits, and approved the ones the thought were poorer. ACA got to a point here they had enough CDS and premium from those CDS to do an arbitrage CDO.

    Now ACA would have known by closing that Paulson was not an equity investor. They COULD have gotten more detail. ACA also had to approve all credits in the CDO. IKB, was a large debt investor (I/my firm successfully “auditioned” for them), that would not know who the other side of the CDS were.

    Would IKB or ACA have passed if they knew that Paulson had ALL the other side of the CDS? Probably not. Paulson was not thought of at that time as a visionary re the housing collapse. Also, there was a “shortage” of CDS available, and there was always the question of price.

    The “crime” here is that Goldman was trading with both sides, taking fees from all parties, but did not disclose the relationship they had with Paulson to ACA. It is reflective of what happens “today”, when I trade with GS o any other firm -I don’t know whether I am trading against the desk an/or the prop trading arm of the investment firm.

  41. Steve Walkman explained quite clearly his view of the difference between “short” and “insurance” and the alignment of interest of the participants. I think you should read his article carefully. There does not need to be a “short” side to the synthethic CDO, in the same way that putting insurance on your house is not shorting your house.

  42. This is aided and abbeted by the legacy governance system. Trying to govern both risk and uncertainty with one-size-fits-all deterministic regime is analogous to having one set of driving instructions for both the U.S. and U.K. In a world of financial complexity, innovation, and bubbles, it is “randomness” that includes both uncertainty and risk that should become the main focus of robust capital market governance.

  43. I did read it carefully, and it is rubbish.

    It is insurance if you are buying a product to hedge your own risk (i.e. you get cash back if your house in which you already have a financial interest burns down).

    If you Steve (with no financial ties to Roger and his house) pay a premium to John (with no fiancial ties to Roger and his house) against Roger’s house that pays off should it burn down, you are short.

    The transaction could not have been contemplated without both Steve and John agreeing to take opposite sides of the transaction (one long, one short).

    If anything, Steve’s point could have been twisted to say that issuers of bonds, stock and insurance are all short their own product. I think that is a far stretch, but much less so than to say noone is short in a synthetic CDO.

    Really – where is the cash flow from to the long position if not from the short? What do you call Paulson in this case if he is not short, and who would have paid ACA if not Paulson (or a replacement short)?


  44. I should add (since oddly it seems like the obvious is not so obvious), that the motivation of the homeowner buying insurance (to hedge their risk) vs. a hedge fund paying synthetic CDO premiums (to profit from mis-valued risk) is important.

    A homeowner knows that they premium they are paying to the insurance company when run through actuarial tables will be a poor investment for them (i.e. the insurance is structured to statistically profit the insurance company), but the homeowner still purchases the insurance to hedge their risk.

    A buyer of default protection in a synthetic CDO structured around a hand crafted pool of MBSs is either taking a short position or hedging an existing long position in a likely correlated asset. That buyer is clearly not willing to over-pay (at least by much) relative to fair value (as computed in their models).

    The presumption in an insurance product is that the seller (the insurance company) is going to make money, at least on a pool of policies – whereas the average buyer of that insurance will lose money.

    When a synthetic CDO is created by an IB, a principal in the transaction needed to go to the IB to have them create the product… and that principal obviously will always structure the deal in a way that is most favorable to them (at least in their opinion).

    So, when GS looks for buyers of a synthetic CDO, the buyers without any written disclosure must know 2 things:

    1) Someone is on the other side of the deal. Since the bank is trying to sell me the long side, that someone must be on the short side (that someone could of course be the IB itself in certain instances).

    2) Since this was a synthetic product and the other side was already spoken for, whomever is taking that contra-side position went to the IB to help create the product.

    Whether this is disclosed or not (and whether that breaks securities laws or not), anyone being approached by an IB to take a long position in a synthetic CDO has to know that there is already someone on the other side, that someone is almost surely involved in helping select the underlying products that are being wagered on, and that someone believes that their financial models will make that bet profitable for them (a material distinction from an insurance buyer who believes the insurance company will likely profit).


  45. I like how you differentiate “uncertainty” and “risk,” concepts often conflated in popular discourse. I’m not sure I’m entirely clear myself on the differences between them, but I recognize them.

  46. ” (that someone could of course be the IB itself in certain instances)”

    The prospectus on the Abacus CDO identified Goldman as the “protection buyer”. Since Paulson was actually the protection buyer, that is a misrepresentation (although not necessarily a “material” one).

  47. AGS– it may have been tougher for ACA to find out that Paulson was not an equity investor, since from their point of view he probably looked like one. I’m guessing that since he was on the entire other side of the CDS, his long equity stake just netted out that portion of his CDS, resulting in a position that was only short the senior (and maybe super-senior) tranches. But Goldman did not disclose Paulson’s CDS position, so ACA could not have known that his equity stake was completely hedged.

    I agree that even if they had known they would have probably forged ahead with the deal, for all the reasons you give plus some (like, the huge bonuses waiting for them).

  48. Thanks for providing a little more color on how these deals were put together. More evidence of the weakness of the SEC’s case and more reason to question the SEC’s devotion of its scarce resources to a relative small and low probability case.

    Inquiring minds ask, why isn’t the SEC bringing a criminal fraud indictment (or asking the DOJ to do it) against the officers of Lehman who published financial statements that were almost certainly fraudulent in 2008. The bankruptcy examiner has done the heavy lifting for them, the risk of a loss is court has been greatly reduced, and investors lost $30B (using the end of 2007 as a reference point) including many small investors who are the intended primariy beneficiaries of the SEC’s anti-fraud efforts.

    Instead it is fighting Goldman in case that at best probably involves tens of millions in damages where the fraud claim will be difficult to prove and where it is not at all clear that the parties involved weren’t aware that shorts were influencing the structure of the deal.

    Are they really trying to prosecute fraud, or are they doing the political bidding of an administration that is viewed as too cozy with Wall Street and is having difficulty getting its financial reform legislation passed (and an agency that on the very same day as the lawsuit is announced was the subject of a scathing report on its own incomptence in the Alan Stanford case). If you aren’t more than a little bit suspicious of the motives, my friend Fab here has a vintage subprime mezz RMBS CDO to sell you (and he’ll thrown in a few shares of your favorite bridge for free.)

  49. I think Steve Waldman did consider the “speculative” versus the “hedging” participants. His take is that, as per his article, “Speculative short interest in whole CDOs was rare, much less common than for shares of IBM.”
    But then I am not sure how he can attest to that statement. We know there are lots of speculative CDSs, with some cases where the outcome exposure to CDSs is more than 10 times of the underlying MBS. There is that case of a hedge fund taking the long side of all the CDS against a small trance of underlying MBS and then buying up and thus make good on all the underlying mortgages. In that case, the short side (GS I believe) cried foul and started litigating.
    Steve Waldman’s point is that net net, each synthetic CDO likely contains more hedging components than speculative components. Also, don’t forget that they are marketed as AAA grade investment, not bets on opinion.

  50. A few days ago in the comments to another post, I suggested that all those ready to convict Goldman might want to wait until all the facts were out, as we had seen only what the prosecutors wanted the press to see.

    Anyone remember the Duke “rape” case and how that turned out after its media-pandering beginning, once the ATM transaction records and taxi dispatch logs were reviewed?

    I suggest all read the following. I caution that it is only a report of having read a material document and not the document itself, but it should surely give pause to all of those ready to convict Goldman (not to mention Paulson & Co.)

    “Paolo Pellegrini told the government that he informed ACA Management that Paulson intended to bet against, or short, a portfolio of mortgages ACA was assembling.

    CNBC has examined documents in which a government official asked Pellegrini whether he informed ACA CDO manager Laura Schwartz about Paulson’s position.

    “Did you tell her that you were interested in taking a short position in Abacus?” a government official asked Pellegrini, referring to the name of the CDO portfolio.

    “Yes, that was the purpose of the meeting,” Pellegrini responded.”

    Simon, like I said the other day, your call for a lynching of John Paulson was unfounded and profoundly premature. Do you want to wait for all the facts, or have you in all your wisdom become judge and jury for the financial system, facts and rule of law be damned?

  51. “It is reflective of what happens “today”, when I trade with GS o any other firm -I don’t know whether I am trading against the desk an/or the prop trading arm of the investment firm.”

    Why should it matter who you’re trading with? That’s what market makers do, they create liquidity. Specialists on the floor of the AMEX and NYSE create liquidity by matching trades from buyers and sellers, of if in the absense of one side, quote a price at which they are willing to take on the trade. They serve both an execution function and a proprietary function. You shouldn’t care who the other side is. You simply want to find someone for the other side of your trade.

    The only way it is relevant is if GS is your INVESTMENT ADVISOR and telling you to make a particular investment/trade with the specific purposes of taking the other side to profit off your losses. This is why everyone looking at this is waiting for a smoking gun. An email or any claim that GS approached investors saying “You’re going to make alot of money on this trade” while saying to themselves “this thing is a piece of crap.”

    It’s also a stretch of the imagination to say that ACA did not eventually realize that Paulson was not a co-investor. ACA held almost 90% of the long side of the trade. At some point someone has to wonder why the guy who was trying to create the synthetic CDO would create it and then hold less than 15%. That being the case they easily had the time to figure out a way to hedge out any risk of defaults if they thought paulson being short was relevant.

    To a lesser extent the same with IKB. With OTC synthetics the contracts are created based on the initial desires of 1 party. Sizing is also relevant to the person shopping around for the deal. Why would anyone shop around for a 1 billion dollar transaction of which they would take only 100 million. (Remember it is a synthetic and not a cash instrument) The very fact that IKB is even approached to be a long investor means that there is excess demand on the short side of the trade.

    Either you believe that IKB+ACA dont care that there was more interest in the short end, in which case I would ask you why they would then care that Paulson was short… or you believe that they’re just completely oblivious to everything and hence need to be protected from their own stupidity.

  52. Buyers would have done it because of the AAA rating.. Now who will tie bell to the big rating cats?

  53. Re @ priscianus jr…..Moody’s,and Standard&Poors rating agencies basically rate the financial markets while Fitch’s Rating Agency takes care of the insurance industry which by the way – there all in bed together. PS. Not one criminal indictment has come their way,…fasinating? (it’s called malfeasance,collusion, and last but not least…gross negligence of fiduciary responsibility)

  54. Re: @ CallingBS…..CDO’s have tangible assets,thus aptly called, “Collatoralized Debt Obligations – whereas CDS’s are structured to be a pure-hedge bet having no skin-in-the game so to say as the name implies – “Credit Default Swaps”. Anyone can purchase them,but…and its a big “BUT”, they are not regulated as other dirivatives,such as on the CBOE (totally different animal). Theoretically they were a brilliant vehicle for shorts,and legitimate hedge funds to cull the system. Unfortunately,the wise guys always find ways to corrupt the system. Ambiguity is the word for today?

  55. Re: @ Equityval….If you find blood in your stool,…do you not seek a doctor? If you urinate blood do you not seek a doctor? If there are dark clouds overhead do you open your windows? If you suddenly go to sell your home,and find out its worth half,or wake up to find out that every hard working person in america has been raped by this financial crises,and no one knows how it could possibly have happened from the top down – I would be scared,…very scared.

  56. The deal was structured as a securities offering. The entity that marketed, promoted and prepared the offering documents and the road show has to comply with the antifraud provisions, no matter how sophisticated the investors.
    It’s not a “trade” but an initial offering and the memorandum must disclose all the material information. From the flip book it appears Goldman undertook the entire process and drafted the memorandum on behalf of the issuing entities; so it undertook the full responsibility.
    It seems Goldman put itself on the hook here by structuring the deal as a securities offering and undertaking to be responsible for all the disclosures.
    It’s the flaw in the securitization model because the “issuer” is often a nominal entity and not the party structuring the deal or the intended recipient of investor proceeds under the terms of the deal ie the sort of facts that are always material. In another words the conventions of securitization are in conflict with the full and fair disclosure provisions of the securities laws that were in part designed to end abusive trust practices that disguised the true participants, their intent, and the ultimate use of investor proceeds.

  57. From a pre-securitization standpoint the deal could be described as follows:
    Paulson is the issuer (ie initiates the deal; pays for the underwriter to promote and market the offering);
    Goldman is the underwriter;
    ACA is the investment adviser (makes the investment decisions)
    or in the alternative
    The trust is a Mutual Fund and subject to the Investment Company Act
    ACA and Paulson are the Investment Advisers
    Goldman the underwriter

    But under either scenario the deal couldn’t be done because the fiduciary duties would preclude it. Enter securitization and all those conflicts melt away and it’s just a disclosure issue.

  58. There’s a great discussion on this structure in the Greatest Trade Ever…it is a must read… and I agree with PeterPeter. I think the post here (and Waldmans too) is so flawed that the corrections noted in line don’t sufficiently address the misrepresentations of Abacus and the roles of the various participants in the transaction.

    If you buy a synthetic CDO, you should know that you are entering into a financial derivative transaction that is indexed to a set of real securities. The information regarding those mortgage backed securities were available to review by investors. Paulson obviously used that information that was publicly available to make his investment decisions. (sadly, many buyers probably had no idea, as they were playing in big leagues that they had no business being in – a discussion of who should be able to access the OTC derivative markets is another post altogether… perhaps treasurers of municipalities or pension funds shouldn’t be able to enter into these trades…)

    Anyway – back to the topic… Someone setting up a synthetic CDO has to get both sides into the transaction – the long side gets attracted with yield and ratings and manager’s experience, the short side has to be induced to enter into a negative carry trade in the hope that the security gets downgraded or defaults to generate a return for them. Paulson ws obligated to make payments to create the cash flows that the ‘bondholders’ were buying into. For both sides, the reference portfolio is what you would negotiate on, aside from Ts and Cs. The long side would probably want to maximize yield within a rating class (sadly, their appetite for yield probably also pushed the vehicle riskier). The short side would push for bonds with ratings that didn’t accurately represent the underlying risk. If it were two hedge funds squaring off on this, we wouldn’t care so much, I think. This same ‘duel’ of institutional investors happens all day long in the futures markets and otc derivative markets.

    Frankly, the part of this entire crisis which isn’t reported on enough is the voracious appetite that institutional investors of various stripes had for MBS and CDOs. There are a variety of reasons for this, but most of them come back to yield and ratings. High yield debt (especially when purchased with leverage) was extremely attractive and was sought. Synthetic CDOs happened because the demand for securitization of debt sucked up all the real debt that was available. Client service focused banks created synthetic vehicles to satiate the demand.

    When I see this Abacus deal in my minds eye, I grant you that I see the picture of arch ‘villain’ Paulson craftily suggesting bonds he thought were over rated in the investment portfolio, but on the other side were guys who were piling onto the latest get-rich-quick scheme of buying securitized debt (or synthetic securitized debt) to take home bonuses from quarterly mark to market gains. Paulson did his homework, and I believe we will find that he most likely did not cross any lines. He probably just declined to participate unless the reference portfolio met his criteria – and anyone who took his participation in that process as an indication he would go ‘long’ is naive about hedge funds. Most go to great lengths to ensure their positions and strategies are not known. Most likely, the losers in this transaction didn’t do their homework on the underlying reference portfolio, and ACA didn’t do their homework on it, and the ratings agencies rubber stamped it. I’ll grant you that there is some small window of possibility that Goldman misrepresented, but I’m waiting to see evidence of it (haven’t seen anything close to it yet). If you are creating or participating in a synthetic CDO, you should have been an incredibly sophisticated investor and known that you are swimming in the shark tank. I think there were lots of dumb slow moving fish for a shark like Paulson to pick off in the 2006-2007 CDO markets.

    If there is an ‘original sin’ here, I’d suggest it is the ratings agencies stamping junk bonds AAA, AA or A, and there could be a follow on scandal if the agents of underwriters tried to massage those ratings.

    Oh yeah, these allegations have been great to get the pitchfork wielding crowd fired up, but I think its focusing on the wrong aspects of the problems in the banking industry.

    I’m very supportive of wrangling TBTF banks and separating prop trading from Grandma’s deposits, but using this SEC action as a catalyst or calling it the Pecora moment I think overstates the issue here dramatically. If we pin our hopes for reform of the banking industry on this case, I’m afraid we end up losing steam because Goldman gets out of this thing unscathed, de-legitimizing all the other valid issues that we all have with the current banking industry structure.

  59. Jake Chase,

    Thoroughly enjoy your trenchant comments and insight. Michael Hudson says citizens have been victimized by Junk Economics and the Europeans won’t take debt servitude lying down like the Americans. (Because the socialist Europeans have a better education system? :)

    From what I can tell: a CDO is a portfolio of securitized debt (eg, mortgages, auto loans, consumer credit, etc). A synthetic CDO is something entirely different and can be a portfolio of credit default swaps based on an underlying portfolio of residential mortgage backed securities.

    In lieu of a synthetic CDO … Why not just bet at the casino or the race track? Because these are regulated venues and no one in their right mind is going to risk a billion dollars in a poker game. But you can do this in the deregulated financial markets.

    You are saying: Paulson and IKB did not own the underlying mortgages in the Abacus deal, but they bet on the performance of these underlying mortgages.

    If I understand correctly, IKB put cash into a structured investment vehicle. Paulson paid a steady stream of revenue into the SIV and this revenue stream was slightly better than, eg, the interest on a US treasury note. IKB was ahead as long as the underlying RMBS performed as rated.

    However the deal was if the RMBS underperformed, or failed, Paulson got cash from SIV in proportion to the losses in underlying asset neither party owned. Correct?

    The spin on the greater fool theory here is: IKB somehow conceived a notion that Paulson would give their SIV — free money — because the underlying RMBS had been “washed” by the quants and rating agencies and therefore as good as risk-free. Correct?

    The financial insanity is there are hundreds of trillions of these CDS’s out there. The counter parties mainly have no ownership in the underlying securities they are betting on. Hence the comparison financial markets are a form of casino. Correct?

    The assets underlying the estimated — $600 trillion notional value of the OTC derivative market — are a fraction of the value of this OTC market. Correct?

    My opinion on this state of financial innovation: Utterly mad.

  60. Re: @ JonW……I have no problem with priority trading for investment/brokerage houses,..but when their now bank holding companies – making them eligible to borrow at the Fed Discount Window at zero interest,… it concerns me. There are laws on the books that explicitly forbid using consumer deposits pooled to finance these caustic endeavors – pushing the envelope close to criminality. Actually,…it is criminal,and there are cases pending,…

  61. Does it get simpler than this?

    You (IKB) go to Las Vegas. Sit at a blackjack table (GS Casino). I (ACA) was already at the table. The dealer (GS) asked me to cut the deck, which I did (completely legit). But, it was Paulson’s deck which he shuffled before you and I arrived.

    GS got its cut no matter what.

  62. Post script:

    Obviously, the deal to sell and buy protection, and the Abacus CDO itself, was far more complex. This also illustrates what James Rickards meant when he wrote in: We have given Wall Street huge incentives to burn down your house.

  63. It would be rather difficult to prosecute the ratings agencies for anything at all. They offer a service to investors, society, and the businesses or investment vehicles they rate. Whether society chooses to use the ratings and hope that the ratings are “accurate” is up to society. There isn’t much more that can be said about that. When talking about the AAA ratings, the ratings are quite simple. Much of the investments were intertwined with U.S. zero coupon bonds with a AAA rating. Even if the intertwined investment failed, the zero coupon bonds would pay in full upon maturity and the CDO investor would realize 100% of the value of the investment at that time. The downside is that this 100% value is not inflation adjusted and there is the remote possibility that there will be no other earnings (thus no ongoing cashflow which is the primary reason for purchase just as you would any other interest bearing security). The “losses” are not 100% but yet when written off as mark-to-market or otherwise it isn’t that the CDO is entirely worthless as many investors are buying these for pennies on the dollar and will realize this in a few years (although I reckon they already know that). That is the other half of this crazy debacle. But let’s not get ahead of ourselves. Investors know what they bought in a sense, but they don’t know what they bought. They do not know exactly which securities make up their slice of the CDO, they only have a generalisation. Due to the complex nature of the security, the only people who know are those who created it, and those who service it. Neither are about to share what is in it. Some CDO’s are performing fine, others are sucking in cashflow, but if they are AAA rated, then they certainly will pay something in the future, just not the amount originally anticipated. Why people buy these CDO’s is beyond me. I would want to know exactly what I was buying. Amongst other things one other thing I do not understand is why CDO’s, CDS’s, derivatives, ect. are not regulated. That doesn’t make a bit of sense. Every company has to register a sale of a bond or stock. Investors have to register if they own a certain percentage of a company. Why not these other investment vehicles? At least we would know what was out there and when the next blowout might occur.

  64. Late Reformer

    It is about time that an ethical code of conduct were drafted and enforced for financial intermediaries. It would probably turn out that GS sold its customers an investment it believed would collapse rather quickly, without disclosing the information upon which this belief was based. In an exchange market, this is a completely acceptable behavior, but not so when the market maker is the one originating the deal and collecting the fees.

    Alternatively, market makers should be forced to reveal all their privately-accumulated information to the public (e.g., at a reasonable delay), to balance for the profits they garner for their exclusive position in the market. To make sure this rule doesn’t harm their profits during the years of low trading volume, the period during which information remains private can be gradually increased as the volume of transactions decreases.

  65. The assumption in any portfolio is that there is a normal distribution of inherent risk. In Goldman’s case, that was intentionally not true. Failing to disclose the intentionally skewed risk to the downside is a fraud on the market.

  66. It strikes me the most offensive aspect of the transaction did not concern its inherent attributes, or how CDOs generally work. Goldman had a duty to disclose material information to investors as part of the offering materials. Clearly Paulson’s role in selecting the RMBS portfolio was known to Goldman but not disclosed. So the key questions should be materiality and reliance by the investors. It seems some people suggest Paulson’s role was not really material, because a sophisticated investor would have realized the existence of a short position. But that is the wrong focus. The question is would anyone have invested if they had known Paulson was centrally involved in selecting the portfolio and also took a short position. Portfolio selection was the essence of the transaction. If you can credibly say it would have made no difference, Goldman might skate through. But that’s a stretch if the SEC can prove what’s in the complaint. Even ACA apparently believed Paulson had an equity position. You can easily imagine there will be testimony that the deal never would have happened if Paulson’s role was fully and fairly disclosed.

  67. There are a few things to note here:
    1. In a synthethic CDO, the “losses” can easily be 100%. Each of the CDS components within the CDO has practically unlimited downside (to the limits of the face value of the underlying security, MBS or whatever) for the limited cashflow. You are in effect underwriting the insurance the payout of which can easily wipe out the treasury bonds portion of the synthethic CDO. It will not go beyond 100% as they are structured as equity/company, but it can go to nothing.
    2. You may have an interest in those CDOs without realizing it. Financial companies may issue bonds against them. Your mutual funds may own them. Your pension fund may own them. The companies you invest in may own them (Many mining companies with extra cash do own bonds backed by such CDOs.) They are marketed as “AAA” security. Most of these big guys buy it with a view to distribute it. Remember what Rubin said in the senate testimony……”I told (my staff) that with these derivatives, that we are carrying the risk until they are distributed….(something to that effect…). They know that it is a musical chair, but most of the participants figure they can get away with it. Or they do not care even knowing the risk, they get the bonus anyway.

  68. Without first reviewing all 66 replies for the same idea, let me just add that if a company like IBM does issue new shares, it certainly prefers to issue them when the stock is valued high in the market place versus at a point when its value is perceived as being low.

    So, IBM would hope to get the better end of the exchange of stock for cash. Yes, the two parties to the exchange wouldn’t have diametrically oppossed objectives, but some “game” still exists.

    If you recall, Warren Buffett took some criticism over this very issue very recently when he opposed an acquisition by a company in which BRK has a major position, then others responded he had done almost the exact same thing when purchasing BNI.


  69. Paolo Pelligrini himself has stated that he told ACA he was/planned to be short. Not sure there’s a disclosure issue here…

  70. The unregulated, opaque OTC Derivatives game lies behind every one of the current debacles. Government gave the banks an opening and they walked right through it. The world’s largest Gold trader, Jim Sinclair has been saying/predicting this for years. Check out his new book: A Pocketbook of Gold: A survival Manual for Monetary Mayhem.

  71. With regard to the Abacus transaction, and “the other side of the trade”…. It’s clear that the synth cdo needed “another side of the trade”, since its based on a portfolio of cds’s on the underlying mbs, which are 2 sided bets. But I think what Paulson purchased cds protection on was the tranches of the synth cdo itself. To the investor, while there is an expectation of negative sentiment in the underlying cds’s, I don’t think there is necessarily an expectation of major negative sentiment in the exact capital structure of the cdo being marketed, ready to pounce as soon as the deal closes! To me this whole deal seams to be a fraudulent (or at least unethical way) for goldman to obfuscate Paulson’s desire for a huge negative bet on the mbs. If not, why not just sell Paulson a bunch of cds on the underlying mortgages directly? (answer: probably because no one would buy the other side).

  72. I don’t know the case that well, but the complaint has some quoted messages where it’s pretty clear some at ACA thought Paulson had an equity position, communicated that to Goldman, and never got clarification from the guy who put the deal together. Anyway it seems my initial point holds that materiality is a key focus. To say ACA knew and went forward anyway is just another way to say that Paulson’s position was not a material fact. Disclosure is the essential securities law question. It won’t be a broader referendum on synthetic CDOs.

  73. Thanks for that practical suggestion. So if Goldman wanted to market a derivative on crude oil, it would have to somehow disseminate everything that everyone in the firm “knew” about the oil market (or do you mean things that they suspect but haven’t confirmed, rumors that they have heard, inferences that they have made from independent bits of data, hunches based on experience trading the market, body language from a meeting with an official of an oil exporting country – where precisely do you draw the line here?????) as soon as that information was acquired by someone in the firm. So as soon as someone read or thought about or heard something about the oil market, they would run down to the general counsel’s office and fill out a form so he could disclose it to all the clueless institutional investors who had no business buying a derivative on crude, but were doing it because they had a buddy who made a killing on it last quarter.

    Is that how your system would work?????

    Good luck with that.

    You have obviously never worked in the investment business. Perhaps though you have worked in Congress because that is the only place an idea this absurd could possibly be taken seriously.

  74. Your question:
    “The question is would anyone have invested if they had known Paulson was centrally involved in selecting the portfolio and also took a short position.”

    The answer: Yes. happened all the time.

    Now, what you should ask is why did it happen?

    Answer: people didn’t care

  75. Re: @ laughter…..AIG subsidiary behind the exotic dirivatives isn’t concerned who’s on what side of the trade (no marrying-up),…it’s a pure premium play,period. PS. If they played the Dirivative’Trade” internally – it explains why they imploded (squeezed go-nad’s de’jour)!

  76. Perhaps the SEC will end up with sour grapes, but it would be a little surprising. ACA did not know of Paulson’s short position, nor did the investors. The complaint is very specific that the offering materials were silent on Paulson’s role. So maybe you will testify as an industry expert for Goldman to prove that no one cared, but I’m betting there will be a lot of evidence to rebut you.

  77. Daniel Barkalow

    While every trade has a long side and a short side, in many trades the short side is maintaining a net long position. That is, when IBM sells shares, it sells less than 100% of the company, such that, even though it doesn’t own as much of itself as it did before, it still is in the position of being invested in its success rather than its failure. They’re starting from a position where they get all of the profit, and selling down to a position where they have some cash and get some of the profit. In order to actually have a short position, they’d have to sell more than 100% of the stock in the company, which is generally frowned upon.

  78. Late Reformer

    The idea is very simple. In the stock market, publicly traded companies are obliged to disclose any material information which may influence the buy/sell decisions of investors. The exact format, in terms of periodic reports, special announcements and the right to withhold commercial secrets, is already well established.

    In the same way, a market maker should be made to disclose any material information it possesses, to the extent that it may influence buy/sell decisions of its prospective customers. The specific format would, I suspect, vary from market to market but have the same beneficial effect on transparency and market efficiency.

  79. May I point out that Money Laundering is also
    just repackaging.

  80. Chris Decker

    Goldman (see Abacus flipbook link below) maybe guilty of fraud but it is a fine line. I don’t think Goldman was anymore an insider or defrauded IKB or RBS than Citigroup or Merrill Lynch. Unfortunately, the latter 2 companies lost their ‘shirts’ on the subprime market by betting long.  Goldman realized in ’07 that the bet should go the other way. Even by March-May ’07, the subprime derivatives market represented by the ABX index derivative which tracked a subset of subprime CDO’s, had already widened in spread to reflect the markets’ high concern for the health of the subprime market. Unfortunately, there were plenty of investors who continued to believe that the ABX was wrong and continued to invest in subprime.  If you look at IKB’s investor presentation (see link below) at that time it will show that IKB wasn’t exactly a babe in the German woods and instead was boasting of its prowness at investing in structured product.  Furthermore, anybody investing in a synthetic CDO had to be fully cognizant that another investor had to be shorting the market.  If they believed that the banker such as Goldman was shorting then they ignored at their own peril.  A quick perusal of some of the prospectuses shows that the banker underwriting the synthetic CDO deals (see Magnetar Carina deal for example) was often the CDS counterparty taking the short side.  Furthermore, at the time Paulson was not exactly a big hedge fund gun with a great reputation on the street.  It was a merger arbitrage hedge fund that was toiling away at making a living.  In hindsight, those who lost big on CDO’s are looking for retribution wherever they can get it and given Paulson’s massive correct bet they can only conclude that they must have been ripped off.

    Reading the SEC complaint closely will show that the majority of the collateral that backed the Abacus CDS structure was vetted and approved by ACA and not Paulson. ACA, which started as a bond insurer, dived head first into CDO management because it was highly profitable and allowed it to build captive stable assets under management. That is why hedge funds, banks, mutual funds, and insurance companies all joined this party.  At the time bond insurance was getting competitive which also led to bond and mortgage insurers such as MBIA and AMBAC to begin underwriting CDO insurance. Bond insurers liked the CDO market because it was perceived that selling insurance on these bonds was a no-brainer due to their AAA rating, the second in line insurance position (i.e. wrappers), and the diversification of mortgage risk vs. concentrated risk in single mortgages or single credit corporate and municipal bonds. The single mortgage insurance market had also evaporated. Traditionally, mortgages were insured whenever downpayments for mortgages were less than 20%.  The advent of subprime and low downpayments using instead 2nd mortgages combined with securitization that didn’t require mortgage insurance, removed the need for traditional mortgage insurance.  Eventually some bond insurers like ACA focused primarily on both insuring and managing CDO structures and mortgage insurers such as MBIA focused on selling second in line insurance. Now everyone is screaming they were ripped off. No, it was all about taking too much risk, not understanding that seldom do poor people or marginal credits pay their back their debts in full, and thinking that diversification of mortage risk would prevent a catastrophe. Furthermore, don’t forget none of these “sophisticated” investors were using data on subprime mortgage that dated more than 10 yrs. Thus all statistics that showed the history of payment and default performance on subprime mortgages was flawed. Any good statistician would have known this.

    IKB presentation

    Goldman flipbook for Abacus

    Another version of CDO & CDS 101 that might provide clarity or maybe more confusion:
    CDO’s are like layer cakes of residential mbs’s (I think everyone understands that). So they are bond structures that hold other bonds. They are derivatives because they are twice removed from the actual credit, i.e. the mortgages. The RMBS’s can be viewed as derivatives also as they are structures that rely directly on the individual mortgages in the pool. Thus conceptually a CDO holds derivatives and not bonds. Furthermore, some CDO’s known as CDO-squared actually hold tranches of other CDO’s.

    CDS are a form of insurance policy/contract:
    Homeowner wants to buy insurance (the buyer of a CDS) on her home protecting it from burning down. The homeowner enters an insurance policy (a form of contract) with Fireprotectioninsurance Co. Therefore, Fireprotectioninsurance sells (the seller of a CDS) homeowner an insurance policy that will pay off if there is a house fire. The policy has terms such as what type/cause of fire will cause the insurance co to pay off the claim, so if homeowner commits arson the insurance co won’t pay. In exchange for this insurance protection, the insurance co now receives a premium from the house owner (spread on a CDS) and it in turn invests the premiums in bonds & stock, etc. in order to have sufficient money one day to cover the potential fire.

    CDS were created by banks to protect their loan positions back in late 1990′s:
    Sometimes a bank that held a loan would also buy CDS to protect in case of default. But because a CDS is an insurance contract it still must have someone willing to provide the insurance. Therefore, by definition there is an insurer and a “policy” holder who expects to be paid if a valid claim is filed. This claim is normally a default as defined in the CDS contract. Initially, because banks wanted insurance that meant that other banks or brokers probably sold these banks the insurance contracts. Given that the CDS is an insurance contract it will have terms that govern whether it will pay a claim. What most people don’t understand about CDS is this contract has lots of fine print including the definition of a default. Furthermore, there are such terms as the insurance contract is written only on a specific bond(the reference security). Normally, it is pretty straightforward for most corporate and sovereign CDS. A default happens when a company or government refuses to pay the interest owed on the underlying bond.

    Why are CDS considered leveraged investments?
    The principal reason to own a bond is to receive an income stream (yield) but an investor normally buys the bond to obtain the income stream and later when the bond matures receive 100% of his investment in the bonds. Traditionally shorting a bond would entail entering a contract to borrow the bond from a broker at a set value (will owe the current mkt value on the bond). When the bond price falls the borrower buys back the bond at a cheaper price in the open market and then receives the difference between the borrowed value and the buyback value, i.e. the short seller covers its position. Selling CDS allows a buyer of a bond to leverage its balance sheet by not having to invest in the underlying principal value of the bond. The seller of the CDS is betting that the bond price will remain stable or increase. It still receives an income stream in the form of the premium or the spread paid by the buyer of the CDS. Buying CDS allows bond investors, who think a bond is overpriced/believe it will decline in price, to quickly and efficiently borrow bonds without physically going through the process as seen above. The CDS premium exchanged between the seller and buyer is not the same as the premium for a traditional insurance policy. In the bond world, when spread decreases it means that yield has decreased. Decreasing yields equal capital appreciation. Vice-versa – increasing spread means yields are increasing and the bond price depreciates. The downside to a CDS seller is if the seller doesn’t cancel the CDS contract before a default it will most likely payout 100% of the value of the bond and thus it now has essentially entered a transaction where it actually might as well paid out much more capital to own the bond instead of selling CDS and is now taking realized loss. The downside to buying CDS is that you are making a payment and thus losing money. This is why a Magnetar or Paulson wanted to buy the equity tranch of the CDO (or sell CDS on it) because this would help pay for shorting the higher quality bond tranches (or buying the CDS).

    Why use CDS to Short?
    Most of the bond market is highly illiquid except for the US Treasury and government agency and agency RMBS markets. Illiquidity tends to characterize corporates, sovereigns, muni’s, RMBS, CMBS, ABS, CDO’s which led to the use of CDS contracts that don’t require the use of physical bonds. Since a corresponding equal value of physical bonds are not needed to support the CDS contract it results in no cap placed on the amount of leverage that could be created by buying/selling contract. As people have said it is like buying an insurance contract on physical property that you don’t own but your neighbor does. The CDS market has wreaked havoc because of this, leading to insider trading, stock price and bond price manipulation, and serious problems for companies and governments that issue the debt the CDS insures. It has also led to market arbitragers trading stocks/bonds/stock options/CDS because the underlying financial math of these securities & derivatives ties the prices together.

    CDS that insure CDO’s:
    These insurance contracts are highly complex vs. a pretty straightforward CDS contract written on a corporate or sovereign bond. The contract may require the seller of the CDS to payout long before a conventional default, i.e. no interest payment. A CDS “claim” can trigger when the underlying CDO’s rating falls say from AAA to BBB or if the counterparty (the seller of insurance) defaults, or its’ financial strength is weakened by some threshold, or potentially the price falls to a certain level. The inv-bankers probably (I’m guessing) did this because the underlying collateral, the RMBS’s, tend to always pay at some level, and servicers are required to advance principal & interest even if the mortgage borrower is in default. Thus in the case of Abacus, when IKB bank (the insurer) was taken over by the German govt and by definition its financial health significantly weakened, it triggered a payout (claim) to the buyer of the CDS (the policy holder). Instead it probably never occurred because of the following.

    Wrapper or 2nd layer of Insurance:
    What makes this crazy market more complex is that the highest CDO tranches in the layer cake, rated AAA, were “wrapped”, i.e. insured a second time, by usually a very high quality insurer (this is what AIG was doing) or bank. This made the AAA’s (the super seniors) more marketable because of this second layer of insurance put in place. [This is akin to when Fireprotectioninsurance co thinks it needs to have insurance protection for itself in case claims against it are too much. So it buys insurance from a "reinsurer" who will then payout if certain triggers occur.] Thus when IKB defaulted and was seized by the German govt it is highly possible they never made an insurance payment. Now if you read the SEC filing against Goldman it noted that ABN-Amro (dutch bank) wrapped the super senior tranches that IKB and others bought. ABN-Amro was bought by 3 parties: Fortis, Banco Santander, and RBS in ’07. RBS ended up assuming the CDS insurance contract that wrapped Abacus. Thus RBS now had legal responsibility for ABN-Amro’s CDS contract if IKB defaulted. Probably when RBS was taken over by the UK govt the second insurance contract(the wrap) claim probably triggered. Because both parties, IKB and ABN-Amro, were essentially in default, without seeing the contract, it is uncertain who was ultimately responsible for paying out the claim.
    The bottom line is that the price of the underlying CDO’s might never have fallen significantly but the payout was most likely 100% on the insurance contract.

  81. Goldman Cited ‘Serious’ Profit on Mortgages

    April 24, 2010 – N.Y. Times – excerpts

    “Goldman on Saturday denied it made a significant profit on mortgage-related products in 2007 and 2008. It said the subcommittee had “cherry-picked” e-mail messages from the nearly 20 million pages of documents it provided.

    Carl Levin, Democrat of Michigan and head of the Permanent Subcommittee on Investigations, said that the e-mail messages contrasted with Goldman’s public statements about its trading results. “The 2009 Goldman Sachs annual report stated that the firm ‘did not generate enormous net revenues by betting against residential related products,’ ” Senator Levin said in a statement Saturday. “These e-mails show that, in fact, Goldman made a lot of money by betting against the mortgage market.”

    The messages appear to connect some of the dots at a crucial moment of Goldman history. They show that in 2007, as most other banks hemorrhaged money from plummeting mortgage holdings, Goldman prospered.

    At first, Goldman openly discussed its prescience in calling the housing downfall. In the third quarter of 2007, the investment bank reported publicly that it had made big profits on its negative bet on mortgages.”

  82. Brian J. Donovan

    For a better understanding of the evolution of subprime mortgage-backed credit derivatives, visit:

  83. EXCLUSIVE: Blankfein Says Trader ‘Fab’ is Immature, but Not a Crook

    April 26, 2010 – Fox Business – excerpt

    “Goldman Sachs CEO Lloyd Blankfein is prepared to say that the young trader charged with securities fraud for failing to disclose alleged conflicts of interest when he sold a portfolio of mortgage bonds to investors was “immature” and “showed poor judgment,” but also that the company didn’t fire him because he did nothing illegal, FOX Business has learned.

    Some of the most colorful of those emails came from Tourre, who joked about selling the toxic bonds to “widows and orphans,” compared his job to “masturbation,” while he described himself as the “Fabulous Fab.”

  84. thats correct tippygolden, dealers on the buy side were collecting fees and had no skin in the game. I was one of them.

  85. Goldman created the CDO at the behest of the short seller and then let him construct the portfolio, and then tried to pass it off as having been constructed by someone else entirely …

    Thats a mispresentation of the ACA selection process, in which among other things, many of the sellers recommendations were rejected