Streetwise Professor

Let’s look at the basics to try to figure out the real strength of the SEC’s case.

One.  Goldman was marketing a synthetic CDO to potential customers, including ACA.  The customers would be long the credit, and would lose if the reference names declined in value, due, for instance, to unexpectedly high defaults.

Two.  Every synthetic CDO has a protection buyer (who is short the credit) and a protection seller (as above, long the credit).  So ACA and other potential protection sellers knew that there was somebody or somebodies who wanted to short these particular credits.

Three.  The possible motive of the party short the credit are, roughly speaking, hedging and speculating.   The speculator has a view, which may be based on non-public information, or proprietary analysis of public data, or maybe just his horoscope, that the value of these credits will decline.  The hedger may have such a view, but may be risk averse and be willing to sell the credits for less than he thinks they are worth in order to reduce risk exposure.  Most hedgers have mixed motives, and hedge selectively: they are more likely to short the credit when they anticipate a price decline.

These three factors are common to every derivatives transaction.  There is a positive probability that the buyer is trading with someone who has information or a belief that the price will decline.  This potential for adverse selection is a risk the buyer faces in every transaction.

Four.  In most transactions in derivatives and securities markets, brokers have no obligation to disclose the identify of their principal; indeed, they are obliged not to disclose their principal.  That is true even though the identify of the agent might be very pertinent information.  Indeed, market participants try mightily, and spend real money, in order to attempt to learn who is the ultimate counterparty, because that can be informative.  Knowing that, other traders go to great lengths to conceal their identities and their trading–that’s why they use brokers in the first place.  Moreover, the traders who are willing to disclose their identity usually do so because they can demonstrate that they are uninformed.

This means that, in general, the party to a derivatives trade doesn’t know, and doesn’t have the right to know the ultimate counterparty.  (I don’t know, as a legal matter, if they ask, it is illegal for the broker to lie.  My guess is it’s just that “don’t ask, don’t tell” is the rule.)

It also means, in general, that they want to know because the identify of the counterparty can be informative.

The fact that most markets are anonymous, and that brokers are obligated not to disclose the identity of the principal they represent, means that the default policy presumes that the benefits of counterparty disclosure are less than the costs.  One may argue about whether that is the correct judgment, but it is the one implicit in the law.

The one factor that supposedly separates this case from other transactions is that Paulson selected a set of credits to include in the synthetic.  But every buyer of X knows that someone wants to sell X, regardless of what X is.  That is, the always seller chooses what he wants to sell in any transaction.  The fact that Paulson chose the particular names doesn’t distinguish this case from the typical transaction; the adverse selection risk is present in any trade.

Another allegedly distinguishing feature is that Goldman allegedly insinuated that Paulson was a buyer, not a seller.  But still, the actual potential buyers knew that there was somebody who wanted to sell.

The strongest argument that can be made is this.  ACA and the German bank allegedly thought that Paulson was long, based on Goldman’s representations (though Goldman disputes this).  Paulson participated in the selection of the credits to be included.  ACA drew a different inference about what private information Paulson had based on its belief that he was a buyer, than it would had it known he was a seller.  The supposed absence of the seller from the design of the CDO–and hence, the apparent delegation of the decision of what to sell to somebody other than the ultimate seller–plausibly reduced ACA’s estimate of the adverse selection risk it faced.  Thus, ACA entered the deal at a price that was higher than it would have, had it known of Paulson’s true role.

How much higher?  Impossible to say, but “not much” is a plausible answer.  Because, again, at the end of the day that ACA knew that it faced a winner’s curse: there was a real risk that the actual short would only pull the trigger on the deal if it were priced unfairly to ACA.

But the anonymity of most financial transactions, and the obligation of brokers not to reveal their principals, suggest that you have no right to be free of adverse selection risk.  It seems that the case will turn, therefore, on (a) whether it is unlawful to deceive the counterparty about your principal, and (b) whether Goldman did deceive ACA, etc.  Would Goldman have been fine if it had just stuck to “don’t ask, don’t tell”?  Did it, in fact stick to DADT?

It may be difficult to show that Goldman actually and intentionally deceived ACA.  And without that, the transaction is very difficult to distinguish from virtually another other transaction in which the buyer faces the risk of dealing with a better informed seller whose identity it does not know.  With adherence to DADT, it would be hard to justify finding Goldman guilty.

What’s more, it’s not as if Paulson was offstage during this entire drama.  Quite the opposite.  He met several times with the ACA team and actively and personally participated in the selection of the mortgages.  If the mortgages that he was pushing were so bad, you’d think ACA would notice and start questioning his true involvement: “You want all this real garbage in the structure.  WTF: Are you buying or are you selling?”

The SEC faces a tension here.  The harder it tries to argue that the mortgages were really, really, really bad, the more it strains credulity to argue that ACA was duped.  ACA could have observed information related to the criteria that the SEC says in its complaint that Paulson used to select candidates for the pool.  They evidently took a different view of the implications of these criteria than Paulson.  Where’s the material, non-public information about the mortgages themselves that might justify a more traditional case?  If ACA looked at the same information Paulson did, and came to a different conclusion, it is hard to argue that it wouldn’t have entered into the transaction if it had known that Paulson had decided to short the mortgages based on this information.

An interesting possibility is that the SEC finds Goldman guilty, but that damages in a civil lawsuit against Goldman may be small.  Certainly, it will be very difficult to estimate the “but for” price of the deal, i.e., the price that the parties would have negotiated had Paulson’s role been known.  Certainly, the synthetic buyers will try to argue that their realized loss is their damage, because they wouldn’t have entered the transaction if they had known Paulson was on the other side.  But that’s not all that plausible; they might have entered the deal, but at a slightly lower price.  There’s no “curative disclosure” here, so estimating that but for price would be devilishly hard.

A couple of last, broader, policy points.  First, the SEC has been broadening its interpretation of what “material nonpublic information” is.  Depending on how the SEC argues this case, the evidence of actual deceit about Paulson’s role (as opposed to silence), and how it is decided, this case could affect how the materiality of counterparty identity is interpreted.  This could have a big impact on the securities and derivatives industry more broadly, and reduce the anonymity of brokered financial transaction.  Perhaps more troublingly, it could add uncertainty into the interpretation of an intermediary’s rights and obligations to its clients.  Hard cases make bad law, and this one has that potential.

Second, this case is being trumpeted as a perfect illustration of why sweeping new regulation is needed.  Indeed, the White House and the Democrat Party are using this as a fund raising plug: they started with this campaign a mere 30 minutes after the complaint came down.  But, if Goldman’s conduct is actionable under the current law, how can that conduct demonstrate the need for more laws and regulations?

That said, all is not well on Wall Street, obviously.  But this case is a distraction from the main issues, notably too big to fail.  The current legislative proposals are not encouraging in this regard.  This Goldman case is factually and legally ambiguous morality play that diverts attention from the central issues, and which could result in the passage of much legislation and the implementation of much regulation that is irrelevant to the main issue, and ultimately destructive.