Is there a Goldman CDO scandal?

By Felix Salmon
December 24, 2009
3000-word story on Goldman and synthetic CDOs, which now has a formal response from Goldman itself.

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

The big story on this slow news day is the NYT’s 3000-word story on Goldman and synthetic CDOs, which now has a formal response from Goldman itself.

Here’s what I think is the most interesting new information in the story:

Beginning in 2004, with housing prices soaring and the mortgage mania in full swing, Mr. Egol began creating the deals known as Abacus. From 2004 to 2008, Goldman issued 25 Abacus deals, according to Bloomberg, with a total value of $10.9 billion…

Rather than persuading his customers to make negative bets on Abacus, Mr. Egol kept most of these wagers for his firm, said five former Goldman employees who spoke on the condition of anonymity. On occasion, he allowed some hedge funds to take some of the short trades…

Goldman’s bets against the performances of the Abacus C.D.O.’s were not worth much in 2005 and 2006, but they soared in value in 2007 and 2008 when the mortgage market collapsed. The trades gave Mr. Egol a higher profile at the bank, and he was among a group promoted to managing director on Oct. 24, 2007.

There are a couple of things which jump out, here: firstly that Goldman was structuring synthetic mortgage-backed CDOs as early as 2004, and secondly that it held on to the short side of those deals for years.

Remember that by their nature, synthetic CDOs have equal-and-opposite long sides and short sides. Anybody buying these things from Goldman knew that someone else was betting the opposite way. And they knew that someone was Goldman, at least in the first instance.

Over 2004, 2005, and 2006, and even into the first half of 2007, demand for mortgage-backed credit assets was insatiable. That’s why so many originators started churning out low-quality paper, and that’s why Goldman got into the origination business too, both on the real-money side (although it was never a huge player there) and on the synthetic side.

Goldman’s synthetic CDOs involved it paying millions of dollars in insurance premiums every year to the investors who bought them. Because of the balance of power between buyers and sellers of credit during the boom, the people buying bonds had very little bargaining power, and the people issuing debt had a lot. As a result, private-equity shops were able to issue billions of dollars in cov-lite loans, and the likes of Goldman Sachs were able to put all manner of triggers into their CDOs — things like ratings downgrades — which would stop the money flowing from Goldman to the buyers, and start sending money flowing back in the opposite direction.

Goldman was certainly in a good position here. It had a substantial portfolio of mortgage-backed securities — so substantial, in fact, that it wrote down $1.7 billion on its residential-mortgage exposure in 2008. As a result, it had a lot of appetite for hedges against those securities, and happily held on to those hedges in 2004-7 when they were losing, rather than making, money. It liked holding the position because it had structured the hedges itself, and knew exactly how profitable they might be if and when the housing market turned.

But does that mean, as the NYT article says, that Goldman’s decision to take the short side of the CDOs “put the firms at odds with their own clients’ interests”? No, it doesn’t. In fact, I’m a bit depressed that we’re still having this argument, a full two years after everybody derided Ben Stein for saying the same thing. I may as well simply disinter what I wrote back then:

If I sell you something – whether it’s a car or a house or a stock or a ham sandwich – I have no fiduciary responsibility to you. Caveat emptor, and all that. If I am investing your money on your behalf, then I have a fiduciary duty. But if you’re looking for fiduciaries, you’re not going to find them on the sell side, only on the buy side.

When Stein accuses Wall Street banks of “betraying their clients’ trust,” he’s simply confused about who the clients are, in these transactions. If I’m an investor and I buy a stock from a broker, then I’m buying it because I think the total amount of money I’m paying is a fair amount for that security. I’m completely agnostic about whom, exactly, I’m buying the stock from: if a different broker has the same security for a lower price, I’ll go there instead. And I’m certainly not trusting the broker to assure me that my security will go up rather than down in value. In fact, at the margin I actually like it if my broker is shorting that stock and thinks it will go down in value – because that just means that I get to buy it at a slightly cheaper level.

It’s just ridiculous to think that “basic fairness” means that Goldman should be exposed to exactly the same risks as anybody who it sells securities to. Goldman Sachs isn’t Berkshire Hathaway, investing money on behalf of shareholders. It’s an investment bank: an intermediary between issuers and investors. If an investor buys any kind of financial security, he’s deliberately buying a risk product. He gets all the upside if that security rises in value. But he also gets all the downside if that security falls in value. It’s not the job of any securities firm to bail him out.

The 30,000-foot view of what happened here is that there was an enormous amount of mortgage paper flooding the market over the course of the 2000s. Goldman Sachs, as a sell-side institution which manages its risk book on a daily basis and doesn’t want to take long-term directional bets, hedged its mortgage exposure with short positions it created by structuring synthetic CDOs. The buy-side, by contrast, had an enormous amount of appetite for long positions in mortgages, and it was the job of banks like Goldman to feed that appetite: again by structuring synthetic CDOs. Goldman was killing two birds with one stone: no wonder Jonathan Egol, who was in charge of these deals, did so well there.

When the mortgage market started to turn, Goldman was smart and nimble enough to realize that it could make money on the way down as well as on the way up. That’s what traders do, and Goldman is the world’s largest and most successful trading shop.

Henry Blodget adds another important point:

Don’t forget that everything is obvious in hindsight. Goldman could have been wrong about the housing market, and its clients could have been right. In that case, we wouldn’t be talking about a scandal. We would be talking about how Goldman got greedy and made dumb bets.

The real lesson here isn’t that Goldman did anything scandalous. It’s just that if you’re making a bet and Goldman is your bookmaker, don’t be surprised if you end up losing.

18 comments

Comments are closed.