The Goldman Sachs defence
Listening to the news coverage of Lloyd Blankfein’s defence of Goldman’s behaviour, and it doesn’t wash.
The argument is that Goldman Sachs sold its customers mortgage-backed securities because its customers wanted them and as a market-maker that was what it would do. They wanted the risk, says Blankfein, but what he means is that they wanted the returns associated with the risk. It was assumed that the clients had done their own research and understood the balance between risk and return in the sub-prime mortgage market; it was not his fault that the whole market was under-estimating the risk – indeed, Goldman itself lost $1.5bn as a result of the miscalculation.
But at the same time as it was selling mortgage-backed securities to its clients (with Fabrice Toure getting a commission on all that he sold), Goldman were doing deals with John Paulson, one of their really big special clients. Paulson realised that the US property market, and in particular the sub-prime mortgage market, was out of kilter and was going to take a bath. So his gamble was to “short” a mortgage-backed security. “Shorting” bank stocks was highly controversial during the crisis; lots of big players – in particular hedge fund managers like Paulson -were doing it. It basically means selling, for the current market price, assets you do not have but borrow from the broker. If the market price falls, you can repay the broker and make money, but if it rises you lose.
The charge is that Goldman Sachs created an instrument called a synthetic collateralised debt obligation, which packaged up a whole lot of mortgage-backed securities (which themselves package up a whole lot of mortgages). Goldman, and the industry generally, had been helping retail mortgage lenders find money by buying mortgage-backed securities from them, and selling them on in the market, and Abacus – the SCDO in the picture – was a lot of these, bundled together. Apparently (allegedly?) Paulson helped Goldman choose which securities were to go into the Abacus product, and then shorted it heavily with Goldman.
In a nutshell, then, the case is: Goldman and Paulson together designed a product called Abacus. Goldman then, and at the same time, aggressively sold Abacus in the market and lent a whole lot of it to Paulson, who then sold it (using Goldman as his broker). Goldman must have known that Paulson had a short position in Abacus; but it did not tell this to its other clients, because they would certainly then not have paid the full price. You don’t tell other clients that one of your clients has a short position until you’ve sold the shorted stock for them.
All of this brings me to a cryptic conclusion, which I will address in the next post: whatever happened to Chinese Walls?