Contents
Stock Market Crashes
Markets Are More Than Simply Irrational—They Can Be Mean
Photographic Evidence: Scientific Evidence of Sentiment Predicting Stock Price Changes
Some People Get Rich by Selling High and Buying Low
Photographic Evidence: Predicting Coin Flips
A Hypothesis Masquerading as a Theory
Sell the Fads, Buy the Outcasts
An Instinct for Losing Money

On Monday, October 19, 1987 , the Dow Jones Industrial Average lost 23%. By noon of the following day stocks faced a crisis where some people feared a total collapse of the stock market. The U.S. Federal Reserve, led by a recently appointed Alan Greenspan, rode to the rescue by guaranteeing certain trades. The market recovered in the afternoon of October 20.

Many people have investigated the 1987 stock market crash. Of note, independent studies were performed by Professor Robert Shiller, the author of Irrational Exuberance, and by his friend Professor Jeremy Siegel, the author of Stocks for the Long Run. 1

These two leading academics are often on opposite sides of the stock debate. Professor Shiller argued (correctly so) that stocks were overvalued in the late 1990s. On the other hand, Professor Siegel has maintained a uniformly positive view of stocks, before, during, and after the bursting of the bubble in 2000. Because of their opposing views on the prospects for stocks, they are often contrasted as leaders of the bear and bull camps.

When it comes to the crash of 1987, however, Professors Shiller and Siegel agree. The crash was not caused by any rational factor such as a news event. Professor Shiller summarizes his findings as, "No news story or rumor appearing on the 19th or over the preceding weekend was responsible for investor behavior." 8 Similarly, Professor Siegel writes, "No economic event on or about October 19,1987 can explain the record collapse of equity prices." 9

As most of us know all too well, starting in 2000 the NASDAQ suffered a decline that was less dramatic than the 1987 crash, but more severe and longer lasting. Table 3.1 shows the performance of leading stocks sometimes called the "four new horsemen of the NASDAQ."

These figures suggest that either precrash prices were irrationally high or postcrash prices irrationally low. When Cisco was priced at $80, the evidence suggests that the stock price was irrationally high. Similarly, when Cisco was trading at $8, after the bubble burst, the evidence suggests that the stock price was irrationally low.

The Denial: The believers in the efficient markets hypothesis deny that sudden price changes indicate irrationality. Furthermore, they claim that

TABLE 3.1 The Decline and Partial Recovery of the Four New Horsemen of the NASDAQ

	Bubble peak (2000)	Post-bubble low	Current (7/2004)
Cisco	Over $80	$8	$23
EMC	Over $100	$4	$11
Oracle	Over $45	$8	$12
Sun Microsystems	Over $60	Under $3	$4

Source: The Wall Street Journal

Cisco's stock price (and all other prices) were correct at the time. Thus, they argue that the decline in Cisco's price from $80 to $8 was caused by unexpected information that was unknowable at the peak.

This claim is based on the fact that stocks discount the future. For example, the value of Cisco in 2000 depends on China 's attitude toward imports in 2010, so even small changes in investors' expectations about China 's future import policy can cause big changes in Cisco's value. Even huge price changes can be rational when nothing concrete changes in the world.

Thus, even though Professors Shiller and Siegel cannot identify the cause of the 1987 crash, it can be explained as a rational response to expectations about the future. Since we can't know what those expectations are, we can't conclude that the sudden changes in stock prices are irrational.

In the real world, we can never prove that bubbles and crashes are caused by irrationality. It is possible that real world crashes are caused by changes in unknowable variables. To investigate bubbles, economists have built artificial stock markets where everything is known. In these laboratory markets, bubbles and crashes, if they exist, must come from inside people.

In fact, Professor Vernon Smith and others have found that even artificial stock markets exhibit bubbles and crashes. In these experiments, people trade a stock for real money. In contrast to the actual stock market, the traders in these artificial markets know the true value of the stock. Nevertheless, traders in these artificial markets push stock prices up to irrationally high prices and then the prices crash.

In the artificial markets, there is no explanation for the bubbles and crashes other than the fact that they arise naturally as part of human nature. True rational market believers argue that the markets are artificial and, in the real world, people who trade at irrational prices would be weeded out. It wasn't me.