UNDERESTIMATION OF PORTFOLIO INSURANCE AND THE CRASH OF OCTOBER 1987

We examine market crashes in the multiperiod framework of Glosten and Milgrom (1985). Our analysis shows that if the market's prior beliefs underestimate the extent of dynamic bedging strategies such as portfolio insurance, then the price will be greater than that which would be implied by fundamentals if the extent of portfolio insurance were known with certainty. Over time, the market learns of the amount of portfolio insurance, and consequently reevaluates the previous inferences drawn from purchases that were erroneously regarded as based on favorable information. The result is that the price falls when the amount of portfolio insurance is revealed.