Bailout

The Goldman wars continue

By Felix Salmon

There is no scandal whatsoever associated with Goldman Sachs emails released by Senator Levin today. They show Goldman people going about their line of work, doing what Goldman people do, taking long positions, taking short positions, sometimes even taking big short positions. This is what broker-dealers should do if they have decent risk management.

The most interesting bit about them, for me, is the glimpse they give into the Goldman PR machine. Here’s Lucas van Praag, responding to an upcoming story by Jenny Anderson in the NYT:

GS Gives is not in the story. I have agreed to brief Jenny thoroughly on it tomorrow and expect the news to run either Tues or Wed. I think it would be good if you had a 5 min phone call with her on the subject and I’ll liaise with Russell on timing. We will issue the press release to coincide with the publication of her article…

Which is exactly what happened.

On the subject of simultaneous releases, Goldman has countered Levin’s damp-squib emails with its own 12-page document entitled “Goldman Sachs: Risk Management and the Residential Mortgage Market”. I agree with substantially all of the points it makes, and I’ve consistently defended Goldman against populist charges that simply shorting mortgages, or hedging long mortgage positions, is inherently evil. They write:

Goldman Sachs never created mortgage-related products that were designed to fail. It is critical to remember that the decline in value of mortgage-related securities occurred as a result of the broader collapse of the housing market. It was not because there were any deficiencies in the underlying instruments. The instruments performed as would have been expected in those unexpected circumstances.

I think this is true, narrowly. Pretty much all subprime CDOs imploded, and the Abacus deals would have imploded no matter what subprime securities were put into them. In that sense, the careful choice of securities by John Paulson in the Abacus-AC1 deal in particular didn’t make much if any difference to the final outcome. If Goldman was long subprime and then put on a big short-subprime bet in order to hedge its long position, then the dynamics of correlation trading would end up giving them an unexpected profit in the end, since no one expected correlations to go nearly as high as they did.

So I’m with Kevin Drum: let’s not demonize Goldman Sachs for shorting mortgages, or for making money doing so, especially since it isn’t true: while the Goldman mortgage desk did make $476 million in 2007, it lost $1.686 billion in 2008. That’s less than its competitors lost, but it’s still a lot of money.

Finally on Goldman I should mention that I have a piece in the Washington Post today explaining just how bad the SEC charges are for its reputation. The economic historian Brad DeLong has a learned response, saying that Goldman never really had the reputation that I’m claiming it had:

Felix Salmon thus, I think, mistakes the business of Goldman Sachs. The old House of Morgan was an investment bank interested in building long-term relationships. Goldman Sachs is instead about doing deals and having the knowledge, sophistication, and intelligence so that it can do the deals with greater ease than anybody else–but it won’t protect you from itself, and won’t protect you from yourself.

I didn’t really have the pre-war years in mind, of course, but Brad does have a good point, and there were other shops, like Lazard and Rothschilds, which probably had a better reputation in terms of long-term client service than Goldman had in the 1990s. Still, when it came to the big investment banks, Goldman was at the top of the heap: its clients trusted it more than they trusted, say, Bear Stearns or Merrill Lynch. And it’s lost a lot of that reputation now.

Full post as published by Macroprudential on April 24, 2010 (boomark / email).