SURVEY: INTERNATIONAL FINANCE Bubble trouble May 16th 2002 From The Economist print edition

All this gives rise to various dangers. An ever-tighter squeeze on the profits of financial firms may tempt them to take bigger risks in their proprietary trading operations, in the hope of staying independent long enough to make it into the bulge bracket. A surprising number of them invested heavily in venture capital during the tech boom, and after initially trumpeting stellar returns have now had to write off most of the money.

Derivatives allow their users both to shed risks and to take bigger risks. This dual role makes it hard to tell whether their impact is benign or malign, and there are plenty of people who argue for each side. What is certain is that financial firms, especially on Wall Street and in the City of London, love derivatives, and have hired an army of mathematicians and physicists to work as “financial engineers”, creating complex new derivatives to shift risk around the financial system. The market for these products is growing rapidly, both on futures and options exchanges and in private sales, which tend to be more complex and more lucrative. Credit derivatives already have a nominal value of almost $1 trillion, up from around $100 billion five years ago. They are forecast to top $3 trillion by 2005. The nominal value of over-the-counter derivatives now exceeds $100 trillion, 60% of which is handled by a mere five dealers, including J.P. Morgan and Citigroup. Derivatives and other tools of financial engineering can be used to manage risk better by hedging positions and transferring unwanted risk to a counterparty, which is what banks say they mostly use them for. However, those tools can also be used to increase risk, perhaps by a big margin, and there is a growing danger that this will be done accidentally.

Emanuel Derman of Goldman Sachs describes much of what a financial engineer does as “a high-tech version of estimating the unknown cost of fruit salad from the known price of fruit, or often, in reverse, estimating the unknown cost of fruit from the known price of fruit salad”. A financial engineer can take the risk in, say, a bond and break it down into a series of smaller risks, such as that inflation will reduce its real value or that the borrower will default. These smaller risks can then be priced and sold, using derivatives, so that the bondholder keeps only those risks he wishes to bear. But this is not a simple task, particularly when it involves assets with risk exposures far into the future and which are traded so rarely that there is no good market benchmark for setting the price. Enron, for instance, sold a lot of those sorts of derivatives, booking profits on them straight away even though there was a huge question-mark over their long-term profitability.