The market crash of October 19, 1987, undermined two preconditions of all portfolio insurance programs—(1) low transaction costs and (2) price continuity. As a result, although most portfolio insurance programs did not violate their minimum returns, many did not perform as well as expected. Differential performance across different programs probably had more to do with the size of the hedges in place at the time of the crash than with the amount of trading the program undertook. Conservative programs (say, 50 per cent hedged) probably outperformed more aggressive programs.

Did portfolio insurance itself contribute to the crash? Portfolio insurance accounted for perhaps 12 per cent of the dollar change in net sold positions on October 19. Is this substantial? It seems unlikely, given the many other significant events—the twin deficits, the falling dollar, uncertainty over the market system itself—that have been implicated in the press.

In the aftermath of the crash, portfolio insurance strategies are likely to be more difficult to implement and riskier in terms of predictability of outcome. We can expect to see fewer dynamic strategies, because of increased transaction costs, and greater use of options markets, which avoid the price discontinuity problem. We might also see a better balance between explicit buyers and sellers of insurance.