Arnold Kling  

The SEC-Goldman Flap

Law and Order... The Fiscal Singularity is Near...

Steven Randy Waldman writes,

A CDO, synthetic or otherwise, is a newly formed investment company. Typically there is no identifiable "seller". The investment company takes positions with an intermediary, which then hedges its exposure in transactions with a variety of counterparties. The fact that there was a "seller" in this case, and his role in "sponsoring" the deal, are precisely what ought to have been disclosed. Investors would have been surprised by the information, and shocked to learn that this speculative short had helped determine the composition of the structure's assets. That information would not only have been material, it would have been fatal to the deal, because the CDO's investors did not view themselves as speculators.

This stuff is getting beyond my depth. I'm guessing that a synthetic CDO is to a CDO as a fantasy baseball team is to a baseball team.

My fantasy AL baseball team this year includes Taylor Teagarden (0 for 18 so far this year), David Murphy (1 for 14), Brandon Wood (4 for 40), Milton Bradley (7 for 42), and Brendan Harris (4 for 19). No, John Paulson did not pick my team so that he could short it, although it probably looks that way.

My sense is that Waldman is overstating the case when he says that a synthetic CDO is a security. Suppose that we created a synthetic mutual fund that did not actually own stocks. I don't think that synthetic mutual fund would be a security. But, again, I am probably beyond my depth.

To me, the compelling points are the following:

1. Nothing could happen to the synthetic CDO until the prices of the underlying mortgage securities changed. At the moment that the synthetic CDO is formed, the prices of the mortgage securities presumably reflect all available information. If they are lousy mortgage securities, that fact is already represented in the price. Even if I am totally ignorant about the underlying mortgage securities, as long as they are appropriately priced at the time the synthetic CDO is formed, as an investor I have a fair shot at making a profit. Even if a short seller selects the securities, he can only select them for his view of their potential to fall, and he may or may not be correct about that.

2. The investors in these securities were professional money managers. Yes, they managed the money that belonged to other people, and those other people may have included widows and orphans, but John Paulson was managing other people's money, also. The reason that the professional money managers lost money on synthetic CDO's is not that Paulson picked the securities in the CDO. It is that the professional money managers took a view of future developements in the mortgage market, and they were wrong. Just as I took a view of the prospects of Teagarden, Murphy, and others, and I was wrong.

Megan McArdle implies that the decision to pursue the case against Goldman was not unanimous at the SEC, and that it was made along party lines. Sebastian Mallaby, who is a pretty straight shooter, also sees the case as more political than not.

To me, the government's case seems pretty artificial. The main reason that investors in synthetic CDO's lost money is that they took the wrong view of the mortgage market. That does not entitle them to intervention by the SEC, as I see it.

Comments and Sharing

COMMENTS (21 to date)
Eric Falkenstein writes:

Interestingly, it seems the ABX-HE index for BBB rated RMBS that reflected the value of this deal showed these were trading well below par (~80) in April 2007. Now, this is an index of 20 tranches, the prices of the constituent tranches were not directly priced, but both the manager, ACA, and any investor in this Abacus deal, were severely negligent for buying the deal at this date given that information. Their only defense was the agency ratings were the same, but agency ratings on ABS are sticky because historically that didn't matter. It's no accident this was one of the last deals done in that space.

Steve writes:

The vast majority of CDOs are synthetic CDOs. This just means they are made up of CDS instead of actual debt.

I agree with your general point, but note that CDS spreads do not move 1:1 with the underlying debt's spreads. There is an active "basis trade" that seeks profit from these differences, and the differences can persist (making the basis trade risky) and at times are quite large (making the basis trade potentially profitable).

Mark writes:

Michael Lewis uses the metaphor of fantasy football in The Big Short to discuss credit default swaps -
"On its surface, the booming market in side bets on subprime mortgage bonds seemed to be the financial equivalent of fantasy football: a benign, if silly, facsimile of investing. Alas, there was a difference between fantasy football and fantasy finance: When a fantasy football player drafts Peyton Manning to be on his team, he doesn't create a second Peyton Manning." (Page 75)
By the way it is not clear to me what the distincton is (or if there is one) between a synthetic CDO and a credit default swap - they both seem to be side bets on an underlying asset backed security.

Bob writes:

To shed some light on synthetic CDOs:

A synthetic CDO (SCDO) is supposed to mimic a "real" risky bond portfolio return. To do this you raise money, buy treasury bonds, and sell CDS contracts on the bonds you are trying to mimic. The treasuries provide some income and the collateral to sell CDSs. The CDS premium payments give the SCSO the higher yields while the CDS liability on default mimics the risky bond's downside. In theory, the SCDO isn't much different than a "real" CDO that holds the portfolio of risky bonds directly, and it could be better or worse depending on how things are priced.

So Arnold is correct. CDO or SCDO, what matters is how the underlying securities perform. The buyers in this case clearly did not do their homework. Goldman obviously didn't do it for them either. An intermediary worried more about getting the deal done than the buyer paying the "right" price? I'm shocked! Shocked!

Jacob N. writes:


I like your points on an economic basis, but I don't know if many lawyers would agree. Also, I think it's worth checking out Goldman's response.

I was originally all for busting Goldman (who doesn't want to get those guys?). But, after learning that the CDO was only sold to qualified purchasers, Goldman lost a boatload of money on the deal, and the SEC vote was split along party lines on whether to pursue the case (coinciding with financial reform...), I am thoroughly unconvinced. If the SEC doesn't have an ace up the sleeve it could end up making itself look petty.

Stephan writes:

I think there's some misunderstanding here. Paulson and the others didn't buy the same stuff.

Without Paulson, Magnetar et al there would have been no CDO in the first place. GS needs someone to buy the "equity" tranche - the worst part of CDO. Without equity tranche there's no AAA tranche. Now Paulson etc. tried to load up the CDO with some composition of CDS, so that their own tranche will blow up with high probability and fast speed. Once their tranche is dust it's upon the rest of the suckers to take the hit. Magnetar seemed to have used the income from their equity tranche to maintain their short position on the same. To be honest: that's almost ingenious.

Topcat writes:

It strikes me that the whole "SEC-Goldman flap" may be a scam, with the two parties, Goldman and the Obama administration working together to fool the public.

1. Assume Goldman wants the Dodd bill. It institutionalizes bailouts and disadvantages smaller competitors. With Fannie Mae-like guarantees from the government, they will be able to borrow at lower rates and thus turn bigger profits.

2. Making Goldman the villain helps Obama push the bill through. Goldman understands this.

3. Hence the paper-thin charges by the SEC (decided by a party-line vote).

4. To ensure the that the outcome is within acceptable bounds, Goldman hires the president's former lawyer. He should be able to protect them from any results that would be too damaging.

Did you ever see the movie "Witness for the Prosecution?"

fmb writes:

Say you're involved in a legal dispute and you retain a law firm to write a brief. They prepare the brief and present it to you for your review, and it looks good to you. It's submitted, and you ultimately lose.

Would you be upset to learn that the firm you retained outsourced the brief to someone working for the other side?

Even though you reviewed the brief itself and all the information about the quality of the brief was available to you, a good Bayesian should consider the outsourcing relevant.

Or think about it like the winner's curse. If you're bidding on a jar of change, your optimal bid depends not just on your estimate of value but also the number of other bidders. Paulson was in effect an undisclosed "other bidder", and one with better eyesight.

JoshK writes:

This is an absurd case. The underlyings are well known.

If I got a call from a salesman saying offering me a great price on Coke stock for size I would be less likely to buy if I knew that it was Warren Buffet was on the other side. But the salesman would be in a lot of trouble for telling me who the other side is.

ABA went bankrupt b/c this entire business crashed. The performance of the Abacus deal is in-line with the rest.

Steve writes:

There is an important difference between SCDO and CDO:

A CDO is funded; you have to pay a cost up-front, which the CDO uses to buy debt.

A SCDO does not have to be funded (and frequently isn't). Because they invest in CDS, which have no up-front cost, there is no direct startup cost associated with a SCDO. The only money investors put up is a small amount of collateral that is used to buy treasuries or other safe assets. The leverage created here can be very high (i.e., one can invest in X dollars of notional credit via a SCDO by putting up amounts as small as X/100 dollars of collateral).

schooner writes:

"At the moment that the synthetic CDO is formed, the prices of the mortgage securities presumably reflect all available information"

Rational Market hogwash. If that were the case then there wouldn't have been the blowup.

Waldman has it exactly right, the investors in the Abacus deal were looking for a safe instrument to provide yield and not for speculation. You don't speculate by buying AAA securities. The real question here is why isn't Moody's in the docket with Goldman.


"Assume Goldman wants the Dodd bill. It institutionalizes bailouts "

Completely untrue. It has the banks pay a tax to raise money for a fund to be used to wind down a failed institution. There is no bailout.

mark writes:

I think the analyst you quote is wrong in this context because the investment portfolio here is completely fixed and each security in it was fully disclosed. There is no ongoing managerial discretion or acumen involved in the so-called investment company.Investments with static pools like this come out all the time and are not classifed as investment companies.It is the very lack of ongoing discretion that takes them out of the investment company model.

The better way to think of it is as follows: imagine that you are invited to invest in a basket of some nvestments - it could be currencies or it could be airline stocks or computer stocks, it doesn't matter as long as the portfolio is static and fully disclosed. Let's call one the "PC Sector Stock basket".- a fund that has 25 stocks in it that have exposure to the PC business - microsoft,intel, dell, etc. The prospectus identifies each stock and the precise amount of shares in each one. It says the portfolio is static and won't change. Does your decision to invest in it change materially and negatively if you are told that John Doe has done independent research on the PC sector and believes that the PC sector has a negative outlook and further John Doe has told the broker that he believes that these are the 25 stocks with the greatest exposure to changes in the PC sector and he is going to buy puts on the fund from the broker because he is so confident they are going down..

To me, the answer is no. The PC sector is big and there are lots of information and opinions out there about it. One person's views on the stocks in the fund is just noise. I make my own analysis and don't let some guy I've never heard of influence me. I don't think any sophisticated investor says yes unless they personally know John Doe.(In fact,there are traders who would bid the fund shares up just to inflict pain on John Doe, especially if John Doe is a small fund and can't retaliate)

The analogy to the CDO here is plain I think. The subprime RMBS market was a very large one and there were a lot of people who had views on it. It's worth noting from the face of the complaint that the insurer of 80% of the CDO was clearly aware Paulson was promoting the specific RMBS and nevertheless the insurer did its own diligence and went ahead with its commitment, which supports the argument that Paulson's role was not material to the investor.

As a legal matter, opinions on the future performance of securities aren't material - they aren't facts.

Eric Hancock writes:

A synthetic CDO is more like insuring a bookie who takes bets on a fantasy football league against paying out on his bets. By holding the synthetic CDO, you are going long the mortgages.

Colin K writes:

The technical arguments in this case are like talking shares of GDP in 2030 for Medicare: Joe Public's eyes glaze over quickly.

The deal may be legal, but it smells rotten. If it's legal, then the law is a ass. When some coke-addled heir loses a couple million in Vegas, the taxpayer doesn't bail him out, and it doesn't put any of my assets at risk, unless I'm in the recreational pharmaceutical business. That's what my Id says, sitting on a jury.

My gut says that there's value in the role brokers serve as intermediators in terms of creating liquidity, but this feels less obvious when we go from real-world assets to fantasy football. This crisis saw the government take all kinds of opaque extraordinary measures with unclear legal authority which benefited financiers very disproportionately. So, we are now merely returning the favor, pour encourager les autres.

mark writes:

I think the deal is legal but I don't at all think the law that makes it so is "an ass". The only reason you think the deal smells is if you don;t live in the world of dealmaking and you are unduly influenced by an uninformed media that needs to reduce everything to a populist folk tale.

I think that we are better off in a world where contracts are enforced between sophisticated parties. I also think we are better served by a world in which sophisticated institutional investors are expected to live with their decisions to do or not do diligence. That will weed out less competent investors. Further, a clear rule that says, among sophisticated parties we let a loss stay where the contract says it falls, leads to less use of public resources to resolve disputes about loss allocation. I don't think that US regulators should be spending tax payer money to litigate quibbles about what "selected by ACA' meant three years ago when the alleged victims, all of whom can easily finance such a lawsuit themselves, have not initiated such a lawsuit.

obviously there is substantial support in Coase and second and third generation work for these propositions.

Colin K writes:

mark: provided that the losses incurred by said counterparties stay private, I too like your system.

R Richard Schweitzer writes:

What does the "buyer" of a syn CDO packet actually pay out, and what does it get in return.

Basically, as I read this proposition, the buyer was to get a flow of "premiums" that might cost the seller about $20 mil if there was no loss called on to be paid.

If only 20 mil of the 1 bil packet had gone bad, they might have come out even, or at least ABN-AMRO might have.

BUT the whole packet went down.

Vangel writes:

I'm guessing that a synthetic CDO is to a CDO as a fantasy baseball team is to a baseball team.

How are synthetic CDOs all that different than an equity or commodity ETF holds T-bills but tracks the benchmark by using the futures market? When I look at the prospectus for many ETFs I note that their collateral securities are usually T-Bills, The exposure comes from a futures position. Does this mean that ETFs aren't real securities?

If they are lousy mortgage securities, that fact is already represented in the price.

The opinion of the quality of the security is reflected in the price but that opinion may be wrong. That is what makes markets so interesting. At the time Paulson took the opposite side of the bet many of the 'sophisticated' investors took him to be a fool. Given the reliance on timing it was possible that Paulson may have been wrong but it looks as if the chose the lowest cost way of protecting his losses and keeping solvent until the market saw the housing market reality as it was.

The reason that the professional money managers lost money on synthetic CDO's is not that Paulson picked the securities in the CDO. It is that the professional money managers took a view of future developements in the mortgage market, and they were wrong.

Nassim Taleb has written on this subject many times before. Many 'professional' investors settle for small continual gains because they do not expect to see the type of event that people like Paulson bet on. For them, it is perfectly acceptable to take continual small losses because they can make massive amounts when things go their way. The money is made by being right on the probabilities. The 'professionals' underestimate how frequent certain events are and by doing so have a tendency to blow up in a spectacular fashion. When they do everyone begins to look to point fingers and the system responsible for the crisis is ignored.

Topcat writes:

The last I heard the $50 billion "wind-down" fund was dropped. That wasn't the point. It might have been a good idea.

But, under the bill, the Fed will designate "systemically important" financial companies. This is code for "too big to fail." The Fed will be authorized to bail out these companies.

No Congressional input needed and, as I understand it, no judicial review. If true, this is a big power shift to the executive branch.

Adam J writes:

The idea that it isn't relevant that someone shorting the CDO is involved in porfolio selection is ridiculous. If its not relevant why did Goldman go to such lengths to convince ACA that Paulson was an equity investor? And why didn't Goldman disclose that someone shorting the CDO was involved in portfolio selection?

Also, I'm tired of this ridiculous sophisicated investor argument- obviously a sophisticated investor still would want to know who chose the portfolio- irregardless of how awesome he is at evaluating securities- that's why half the freaking Goldman dealsheet was discussing how great ACA is. Would you want someone who is shorting your mutual fund to be involved in selecting its porfolio? At the very least you would want this fact disclosed so you could go over each security with a fine-tooth comb.

Mustafa writes:

@Mark Your analogy of a market basket is spot on. However as an investor in the basket if I was told that John Doe has been actually paid the intermediary to put together the basket (as Paulson did in the Goldman case) as well has some veto rights over what goes into the basket (include Novell but not Microsoft), it is the duty of the intermediary to disclose this fact.
It doesn't matter who is buying the securities, the fact that John Doe is short the basket and has no long position is relevant information to the investor.
In the Goldman case there is an additional factor which the SEC is claiming that Goldman actively presented the fact that Paulson was a long investor in the equity tranche, just like a lot of Magnetar deals.
But irrespective of that the disclosure aspect still holds.

Comments for this entry have been closed
Return to top