Richard Bookstaber: The recent hedge fund losses did not affect the money-center banks, which is what caused the wider concerns about LTCM in 1998. The trigger for the LTCM crisis was fairly minor by most measures. It was the default in Russia, where they lost $300 million. Usually when something like this happens, you can say, "OK, I lost this money, I’d better pull some capital out to meet the obligation with the banks," and you sell something. But then the prices of the assets you need to sell go down, so you have to sell more and more to make good on the leverage. But if you are leveraged through, say, futures, rather than bank loans, that sort of cycle is not a problem for the system overall.

Glassman: There’s more liquidity in the market today, in part because of hedge funds. Amaranth was taken out by another hedge fund. [Citadel Investment Group, along with JPMorgan Chase, assumed Amaranth’s battered energy portfolio.] One has to be careful of saying that hedge funds are a problem; clearly they bring liquidity. What happened with LTCM and Russia in 1998 was caused by a lack of liquidity. You now have more players like hedge funds that are willing to provide liquidity to the market.

Worth: Nouriel, it sounds like the recessionary scenario you described would require a dramatic tightening of global liquidity.

Roubini: This is where there are two concepts of liquidity. One measure is low nominal annual interest rates. By any standard, interest rates are still low, and so there is marginal liquidity.

But if I’m right, and the housing market leads to an economywide recession, then you are going to see a liquidity problem within the banking system. The banks claim they have offloaded much of their $10 trillion in mortgages, but actually half of the assets on their balance sheets are in mortgages or mortgage-backed securities. About $5 trillion of mortgage capital is guaranteed by Fannie Mae and Freddie Mac. If even one-fifth of that went belly up, the cost in public debt would be $1 trillion. The savings and loans had only $200 billion of exposures in the 1980s when the S&L crisis occurred.

Bookstaber: Another risky area that is not getting much attention is the market for credit default swaps and credit derivatives. Because nothing bad has happened with this market, no one is paying attention. But the problem is, if there is a crisis, the market is not going to get a second chance. It is a huge market, and it links right into the money-center banks.

There are a lot of theories about how things could go wrong. Say a lot of credit default swaps are written on a corporation, and it runs into trouble. There would actually be an incentive for that corporation to issue a lot more debt, because the credit default swap holders would need it to make good on their swaps. The problem is that the corporation is not going to be able to issue new debt right away, so there won’t be enough to settle all the default swaps. If you had all the time in the world to solve the problem, you probably could work things out. But it’s like the engineering concept of tight coupling, where you have very tight links between a series of complex steps, and no time to intervene—like a shuttle launch. If you could press the pause button at some point and have everybody deliberate, maybe you could solve the problem, but during a crisis, you can’t.

It is through the very design of the financial markets that we end up with the risks that we have. In nuclear engineering, you have a similar sort of tight coupling, through faster and faster interactions combined with growing complexity, and, as a result, the very nature of the system is such that you expect to have a certain number of what are considered "normal" accidents. In the financial markets, every time we have a problem, we just add some more regulations or oversights, but that just adds complexity to the structure. So you have a paradoxical result, that more rules and regulations can actually increase the amount of risk.