Haim Mendelson

Liquidity and the 1987 stock market crash

T he Crash of October 1987 was a puzzling event puzzling not only for what happened on October 19, but also for what subsequently did not happen. In spite of the magnitude and momentum of the crash, its effect was limited primarily to the financial markets, while the economy as a whole did not change its course. And, although the ”bad news” that was supposed to be predicted by the crash has apparently not materialized, the market suffered a lasting decline after October of 1987. This article advances a liquidity theory of the crash, proposing that the price decline in October 1987 reflects, at least in part, a revision of investors’ expectations about the liquidity of the equity markets. Amihud and Mendelson [1986a, 1986b, 19891 have shown that the market price of stocks is positively associated with their liquidity (after controlling for risk). Given this relationship, when the liquidity of assets turns out to be less than had been expected, their price should decline. We suggest that the main news that caused the prolonged decline in stock prices was the crash itself that is, the realization that financial markets are not as liquid as previously assumed. Investors recognized that stock prices should reflect a larger discount for the costs of illiquidity, which turned out to be much higher than had been expected before the crash. Partial recovery following the crash reflects an upward reevaluation of the liquidity of the markets -that is, investors recognized that while the markets are not as liquid as had been assumed prior to the crash, they are also not as illiquid as when there was the possibility of closing the markets altogether. These illiquidity problems, reflected in wider spreads between the quoted bid and ask prices, have persisted long after the crash, and market impact estimates have stayed significantly larger than they had been prior to the crash.’ Thus, the crash and subsequent events have produced new information about the markets themselves rather than fundamental news about the economy. Part of the liquidity-related price decline on October 19, 1987, was temporary. Unexpected sale pressure, even absent any negative information, can generate a temporary negative price impact, as we sometimes observe in block sales. But much of the effect was permanent. Investors realized that they may well have to pay a larger discount when they wish to sell stocks in a hurry. These illiquidity costs result in a stream of future cash outflows that translate into a loss of value.2 In general, illiquidity reflects the difficulty of converting cash into assets and assets into cash, or the costs of trading an asset in the market. Some of the costs of illiquidity are explicit and easy to measure, while others are more subtle. These costs include the bid-ask spread, market-impact costs, delay and search costs, and brokerage commissions and fees. These components of illiquidity cost are highly correlated: Stocks that have high bid-ask spreads also have high transaction fees and high search and market-impact costs, and are thinly traded (McInish and Wood [1989]). When the bid-ask spread widens, it signals that immediacy of execution is more costly,

Market Liquidity: Asset Pricing with Liquidity Risk

This paper solves explicitly a simple equilibrium model with liquidity risk. In our liquidityadjusted capital asset pricing model, a security s required return depends on its expected liquidity as well as on the covariances of its own return and liquidity with the market return and liquidity. In addition, a persistent negative shock to a security s liquidity results in low contemporaneous returns and high predicted future returns. The model provides a unified framework for understanding the various channels through which liquidity risk may affect asset prices. Our empirical results shed light on the total and relative economic significance of these channels and provide evidence of flight to liquidity. r 2005 Elsevier B.V. All rights reserved.