Why the Abacus Investors Weren't Speculators

Includes: GS, MS
by: Felix Salmon

Back in 2007, I embarked upon a doomed attempt to explain that asset bubbles were not necessarily speculative bubbles. There might be a bubble in the art world, I said, but it wasn’t speculative. And neither were home-price increases speculative in 2000, before the tech bubble had even burst. (Now that was a speculative bubble.)

In my book, a speculator is someone who buys an asset with the express intention of selling it after it rises in price. (Or, conversely, someone who shorts an asset with the idea of profiting from a downward price move.) A buy-and-hold investor, pretty much by definition, can’t be a speculator.

But Jim Surowiecki seems to have a different conception of what speculation comprises:

The economics of subprime lending, and of C.D.O.s, depended on the assumption that there wasn’t a housing bubble, and that therefore housing prices would not fall sharply. That was a speculative assumption, given the massive runup in housing prices between 1998 and 2006, and if you acted on it you were speculating, even if you described it to yourself by another name.

I think this is a little unfair to the people who invested in CDOs. They didn’t make an explicit assumption that there wasn’t a housing bubble, or even that housing prices would not fall sharply. (These are not the same thing: you can have sharp falls in housing prices without a housing bubble, as anybody in Detroit will be able to tell you.)

Instead, the CDO investors relied on complicated models and Monte Carlo simulations, and the complicated models relied on the history of house prices in the US, and the history of house prices in the US was that although there had been localized drops in price, there had never been a national drop in house prices.

Smart people like John Paulson were clever enough to grok the weakness in the models, and to notice that they resulted in some massively mispriced securities. That insight was ultimately what drove The Greatest Trade Ever. But I think it’s a bit much to say that everybody on the other side of that trade was making a speculative bet, when in fact they were buy-and-hold investors buying triple-A-rated securities paying 85bp or 110bp over Libor.

As far as the bond investors were concerned, the risk and speculation was carefully and deliberately consigned to the equity tranche of the Abacus deal. In fact, there was a line of thought at the time which said that these kind of instruments were actually better investments because the housing bubble was bursting. It seems crazy now, with hindsight, but it’s worth remembering that at the time, everybody thought that the problem of default risk had been solved by overcollateralizing and tranching the CDO so that the default risk was borne by the equity investor. The big remaining risk in any mortgage pool was prepayment risk — the risk that homeowners would pay off their mortgages very quickly, sticking the investor with cash which then might not be able to be reinvested at such a high interest rate.

Since mortgage prepayments mainly came from refinancings, it was seen as good news to mortgage investors that various subprime mortgage originators in California and elsewhere were closing their doors. Fewer originators meant fewer prepayments, and fewer prepayments meant higher and more reliable total returns. The fact is that a housing bubble is bad news for mortgage investors, since if you flip your house, that’s a prepayment, and if you refinance your house, that’s a prepayment too, and prepayments are always the last thing a mortgage investor wants to see. So it’s entirely reasonable to assume that IKB and ACA were happy to see the housing bubble coming to an end. So Jim’s wrong here, I think:

If you were buying subprime C.D.O.s in early 2007, you were betting that the housing bubble wasn’t going to burst, even though in much of the country it already had. That was a dubious bet at best.

My feeling is that if you were buying subprime CDOs in early 2007, you were obviously short-sighted and not very good at doing your due diligence. But I don’t think you had any particular desire for the housing bubble to continue. It’s clear with hindsight that a bursting housing bubble could wipe out the principal not only of the triple-A tranches of CDOs but even of the super-seniors. At the time, though, very few people saw that coming: remember Morgan Stanley’s (NYSE:MS) Howie Hubler, who saw the market collapsing, put on a big bit that the collapse would happen, but then tried to fund that bet by selling protection on the triple-A tranches of the same CDOs. “He was smart enough to be cynical about his market,” writes Michael Lewis in The Big Short, “but not smart enough to realize how cynical he needed to be.”

In other words, even ultrasophisticates like Hubler, who saw the collapse coming and bet on it happening, thought the triple-As were safe. So let’s not say that investors like IKB were making a speculative bet that the housing bubble was not going to burst.