Ich bin zufällig über ein Interview mit Martin Mayer gestolpert, das von 1999 ist, und in dem Mayer damals schon die Schwachstellen der „Finanzmarktinnovationen“ beschreibt, und deren Versagen im Krisenfall vorhersagt. Hier ein paar Ausschnitte:

That’s the secret of the JP Morgan lawsuit with the Korean banks. It was Korean Central Bank money that these guys pissed away. You had this incredible device—a Malaysian corporation formed to write instruments that would pay off according to the relative values of the Thai baht and Japanese yen, and designed solely for the purpose of selling these instruments to the Korean banks because the Korean Central Bank wanted to increase the return on its reserve.

Once you start to get into this sort of nonsense, serious problems can occur for everybody, because you are dealing in a world of thoroughly artificial instruments that have nothing to do with real production, real consumption, real trade or real anything. They are simply counters in a game. But when the players have to pay off, they have to pay off in real resources, and that sort of thing is extremely dangerous over time.

In October 1997, when the Brazilian real came under dramatic attack, nobody could figure out why. It turned out that in September ’97, between 20 percent and 25 percent of all the Brazilian Brady bonds were held by South Korean banks. The Brazilian Brady bonds had a large, liquid market because people were making a lot of money trading them and trading derivatives off of them. So when the South Korean banks got in trouble, the easiest thing for them to do was to sell Bradys. The bonds took a hit and the real took a hit. And that’s crazy.

It’s one thing to talk about a spillover effect with countries that are closely related to each other. It’s another thing to talk about contagion. There have been artificial links created between currencies because of the statistical correlation in their path behavior. That has nothing to do with anything. Correlations and causes can get mixed up anytime, but the notion that you put hard money on correlations where there is no—and can be no—causation other than the correlation itself is bad news.

What’s really happening is that the guys selling the hedging instruments are writing a form of insurance. It’s catastrophe insurance, but they don’t know it’s catastrophe insurance, so they have no way to price it that gives them accurate premiums for the risks. Nobody knows. The question is whether there shouldn’t be some regulator making sure that when the insurance doesn’t pay off, important players don’t go absolutely insane and bust. That’s got to be addressed.

Banks are supposed to do information-intensive lending. The notion that the ratings agencies know what the risks are on loans better than the banks do—well, that’s really remarkable. It means that everybody has adjusted to a world in which banks are not going to hold loans in their portfolios. They are going to sell them off. And the ratings are important for the purpose of selling off the loans.

We’re moving from a world that was dominated by specific information in banks to a world dominated by much less profound but much more widespread information in markets. The clash between these two systems of valuing instruments is what makes for the instability that we can see all around us and that will increase.

There’s obviously a real value in credit derivatives because they allow the farm-belt bank and the rust-belt bank to swap exposures. By diversifying, both of them have less risk. The reduction of their returns is minor because the cost of these contracts is minor.

In some cases, you’re not swapping portfolios of loans. You’re gambling on the relative return of a risk-free instrument against an equity portfolio, a portfolio with junk bonds, a portfolio of foreign currencies or whatever, in which there is a larger risk.Nobody knows what the hell is going on in these private deals, and nobody knows the volume of what is out there, including regulators. So nobody can have an informed opinion about what the dangers are. What has happened over and over again with these instruments is that people have assumed that there is measurable risk where in fact there is immeasurable uncertainty.