Investors, regulators, brokers, dealers and the press have all expressed concern over the level of stock market volatility. But the perception that prices move a lot—and have been moving a lot more in recent years—is in part merely a reflection of the historically high levels of popular stock indexes. The drop in stock prices on October 13, 1989—while large in terms of point decline—was not even among the 25 worst days in NYSE history in terms of percentage changes. While a 6 per cent drop in prices is not inconsequential, neither is it a rare event when considered within the context of the behavior of stock returns over the 1802–1989 period.

Apart from October 1987 and October 1989, volatility was not particularly high in the 1980s. Moreover, the growth in stock index futures and options trading has not been associated with an upward trend in stock volatility. There is little evidence that computerized trading per se increases volatility, except perhaps within the trading day.

On October 13, 1989, all the major networks flashed reports on the market decline. The ability of investors and the press to track stock prices on a virtually continuous basis has heightened public perceptions of a volatility problem. What we do not know, because the intraday data on stock prices are simply unavailable, is whether the large but extremely brief price drops that have characterized recent market declines also occurred in the past, when daily and monthly volatility was higher than it is today.

The evidence so far is inconclusive as to whether trading halts or circuit-breakers can reduce volatility in a beneficial way. Even if circuit breakers can reduce volatility, are the benefits of stability greater than the cost of the inefficiency created by the trading halt?