Walls, Brakes, and Risk

THE U.S. stock market boomed throughout the mid-1980s. One contributing factor was a major technical innovation of that time called “portfolio insurance.” This was basically a trading technique by which an investment bank would buy a basket of stocks and simultaneously “short” another, closely related basket of stocks. (“Shorting,” or “selling short,” amounts to betting that the price of a given security will decline. One technique involves borrowing shares for a specified time and immediately selling them, with the intention of buying them back at a lower price when the loan comes due.) If you did this just right, you could make money whether the stocks you bought went up or down. It was like a bookie’s taking action on both sides of a bet.

Some foresaw that this technique could radically increase the speed and severity of a market decline. That’s because a key hidden assumption of portfolio insurance was that you could short stocks at will without your trade’s affecting the price. If a very small fraction of the trading volume for these securities is driven by portfolio insurance, the assumption works very well. But, in the extreme, if all trading volume were following this strategy, then once a general market decline began (for whatever external reason), the complicated financial logic of portfolio insurance would require everyone to start shorting more and more. There would be an avalanche of would-be short sellers with no buyers; you can’t bet on a price decline if there’s no one to bet against you. This would create an ever-growing herd of investors desperate to sell shares at any price, and the market would go into free fall with no bottom. At some unpredictable point between “a very small fraction” of trading and “all,” the whole market is vulnerable to collapse in the face of a price decline.

About the Author

Jim Manzi

Jim Manzi is CEO of Applied Predictive Technologies (APT), an applied artificial intelligence software company.

Featured Tags

THE U.S. stock market boomed throughout the mid-1980s. One contributing factor was a major technical innovation of that time called “portfolio insurance.” This was basically a trading technique by which an investment bank would buy a basket of stocks and simultaneously “short” another, closely related basket of stocks. (“Shorting,” or “selling short,” amounts to betting that the price of a given security will decline. One technique involves borrowing shares for a specified time and immediately selling them, with the intention of buying them back at a lower price when the loan comes due.) If you did this just right, you could make money whether the stocks you bought went up or down. It was like a bookie’s taking action on both sides of a bet.

Some foresaw that this technique could radically increase the speed and severity of a market decline. That’s because a key hidden assumption of portfolio insurance was that you could short stocks at will without your trade’s affecting the price. If a very small fraction of the trading volume for these securities is driven by portfolio insurance, the assumption works very well. But, in the extreme, if all trading volume were following this strategy, then once a general market decline began (for whatever external reason), the complicated financial logic of portfolio insurance would require everyone to start shorting more and more. There would be an avalanche of would-be short sellers with no buyers; you can’t bet on a price decline if there’s no one to bet against you. This would create an ever-growing herd of investors desperate to sell shares at any price, and the market would go into free fall with no bottom. At some unpredictable point between “a very small fraction” of trading and “all,” the whole market is vulnerable to collapse in the face of a price decline.