Thoughts on the SEC's suit against Goldman Sachs

[jon_leonard]

I'm not quite sure what to make of the SEC's suit: Goldman apparently put together a deal between John Paulson and ACA, and the core of the complaint is that ACA wasn't given enough warning that Goldman thought that ACA was getting a bad deal.

The puzzling part is that the deal was a synthetic CDO, which (unless I'm missing something important)

Comments

[jon_leonard]

In the interests of disclosure: Beth & I have a long position in Goldman Sachs. We probably just got assigned on some puts, and have more puts outstanding.

Essentially, I expect investment banking to do well unless there's an outbreak of wisdom in congress and they change the factors making finance unduly profitable. Goldman Sachs is probably the best positioned to take advantage of unfortunate policies at the Federal Reserve, the Treasury, etc.

[mrc80238]

Lots of good analysis on this around the net, but here's some good takes:

http://blogs.reuters.com/felix-salmon/2010/04/16/goldmans-abacus-lies/

Key paragraph:"The scandal here is not that Goldman was short the subprime market at the same time as marketing the Abacus deal. The scandal is that Goldman sold the contents of Abacus as being handpicked by managers at ACA when in fact it was handpicked by Paulson; and that it told ACA that Paulson had a long position in the deal when in fact he was entirely short."

Essentially, you're right, it is a zero sum deal. The problem is that they were suckering people into taking the downside of the bet. That's not OK.

[ricevermicelli]

You use this term "suckering" and while you're right there, I think it's important to note that Goldman's statements in this case meet every aspect of the legal definition of fraud.

[jon_leonard]

That explains my problem: I was looking at the wrong victim. ACA should be able to take care of themselves (at the very least get highly suspicious when Paulson's list of securities looks strange), but people taking the long side of the bet have probably been defrauded.

I'm not sure what to make of the fact that a synthetic CDO is pretty much guaranteed to be a sucker bet for someone: The line between ethical and unethical is tough to draw.

[anonymous]

There was another post I was trying to find that discussed the fact that the problem with CDO's is that you have to find some sucker to take the downside risk, and this is how they did it. And yes, "fraud" is a better word than "sucker" here. It's one thing to tell someone "this is a great deal, trust me, don't read the fine print," and another to say "look at this fine print, it's awesome!" When the awesomeness is, in fact, bogus.

[jon_leonard]

That only applies to synthetic CDOs: With a "traditional" CDO, money from the various tranches is pooled together and used to buy existing bonds (mortgages, etc.). Then there are various rules as to who gets paid when, which is why the tranches are different. Still not a deal that seems terribly appealing to me, but it doesn't require a built-in sucker.