Why search for the cause of the crash? Have the 17 years that have passed made it irrelevant? Not at all. If we have insights into understanding the 1987 market crash, the most tumultuous movement ever in recent market history, then the basis exists for us get a better understanding of the market as a whole, and from there contribute a few insights into the inner workings of human nature. This is valuable knowledge.

Not only is knowledge for knowledge's sake worth pursuing, but also the practical benefits of such knowledge. In the future, society may find it advantageous to limit extreme price movements like the 1987 crash because of the disruption they create. We can only do so if we endeavour to understand what actually caused the 1987 decline. By understanding the cause, we can determine whether crashes like 1987 are necessary to the functioning of our capitalistic system, or are mistakes created by human thought processes.

What Caused it?
There are several standard explanations that were immediately brought forth to explain the crash and have circulated ever since. I explain the successes and failings of each.

program trading
overvaluation
events
illiquidity
the economist's response

When approaching the question of explanations for the crash, keep in mind that a good solution should answer these five questions:
1. Why the crash occurred on Monday, October 19, 1987, and not any other day.
2. Why it occurred at the speed it did, playing itself out over 1 day, rather than slowly over the space of several months.
3. Why it fell to such an extreme degree, 23%, rather than a more rational 3-4%.
4. Why it occurred on an international scale, and not just in the United States.
5. Why no major news or significant events preceded the decline.

A New Explanation
I have been working on an explanation for the 1987 stock market crash for the past year. The theories proposed over the sixteen years since the crash have frustrated me because they have been unable to give convincing reasons for why markets could fall to such an extraordinary degree without any significant prior announcement or change in the world. Yet the decline cannot defy all logic, nor can it have emerged spontaneously. It had to have some cause, or a collection of causes. Intuitively, it seemed to me that the selloff was rooted in psychology, not in the structure of the markets or the events that occurred just prior to the crash. The problem with psychological explanations is that they reside deep in our minds and therefore are difficult to access. Typical explanations that resort to mob psychology and the irrational nature of crowds help us understand events like crashes in a very general way, but fall short when explaining the particulars of the one in 1987. In the paper below I tease out what I believe might have been the psychological cause of the 1987 crash.

Abstract
To put it briefly, in 1987, market participants felt that conditions were starting to resemble 1929, the year of the well known crash that ushered in the Great Depression. This feeling emerged from comparison of prices - if one overlays price graphs of the two markets, they appear eerily similar. Market participants started to compare the long seven year bull markets that characterized each era, and to many observers, the peaks and valleys near the end of their long bull moves seemed to overlap each other (see chart). My thesis is that a number of 1987 participants perceived a resemblance between their era and 1929, and sold stocks based on the analogy they had drawn. After all, why would anyone want to risk being caught by another '29 style crash and 30's Depression? The resemblance was only superficial, in reality 1929 and 1987 were two very different fish. Nevertheless, market prices were forced down by this selling, and in the eyes of market participants, appeared to be following 1929 even closer. Like a self-fulfilling prophecy, participant's belief that 1987 resembled 1929 resulted in trading that forced 1987 to conform to the old 1929 pattern. On Monday, October 19, 1987, enough people chose to sell based on the 1929-1987 analogy that the 1929 crash fully reappeared in 1987.

Unlike the typical panel of causes put forth such as program trading and overvaluation, my theory is able to explain precisely...
1) why the 1987 crash occurred on Monday, October 19.
in the historical model investors had adopted, the crash had occurred on a Monday in October, 1929. The overlap in Mondays resulted in a self-fulfilling prophecy to emerge in 1987.

2) why the move was so extreme in price.
1987 fell 23% because participants noted that 1929 had fallen 24%.

3) why it fell so quickly.
1929's 24% decline had only taken 2 days to occur. By analogy, market participants felt equities could fall as fast in 1987

4) why 1987 was an international phenomenon.
the 1929 crash set a precedent by affecting markets from Paris to London to New York. Thus investors the world over were influenced by the memory of the crash and Great Depression, not just those in the U.S., and were willing to sell in 1987 using the 1929 memory as justification

5) how a lack of major news or important events prior to the decline could justify a 22% change.
the cause was psychological, caused by an old memory. Thus it needed no events or important news to emerge.

The theory also explains an observation that has been noted by many - why 1929 and 1987 are so remarkably similar in terms of shape, timing, and extent (see another chart). Another strange observation commented on in the press was the continued strength in the U.S. economy both before and after the market crash. If the decline was purely psychological, this explains how the stock market could decline even though the economy was still kicking on all cylinders.

This is a simplified explanation, please read my essay Explaining the 1987 Market Crash (PDF) for a complete treatment.

Future Crashes?
If all it takes to start a crash is a belief that current markets resemble past crash years like 1929 and 1987, then we may have crashes again in the future. Going into the second half of 2003, market participants are inevitably starting to make the perceptual connection between the present and 1987. Indeed, come October it has become a tradition to do so in the press. Commentators have warned about the potential for a repeat of 1987 every fall through the late 1980s and 90s, possibly causing small mini crashes in October 1989 and 1997 (see my paper for these explanations). It is entirely possible they will do so again.

How might participants justify a perceptual connection between 1987 and 2003? Prior to the 1987 stock market crash, the bond market had experienced one its worst declines in history. The stampede from bonds this summer provides an analogy to the 1987 bond route. The argument has it that high interest rates will coax equity investors away from dividend bearing stocks to bonds, resulting in an equity selloff.

The US dollar had been declining prior to the 1987 crash, and worries about record budget and trade deficits abounded. In 2003, the US dollar has again found itself in a strong downtrend, the budget deficit is ballooning as Bush wages war, and the trade deficit is hitting record highs. It is easy to see the similarities. In 1987, stock markets were overvalued by most fundamental methods of valuation. After rising for almost 10 months now, the present market finds itself again trading at an excessive price earnings ratio. As we approach October, these sort of analogies will appear in the press, on the internet, and be circulated in the trading community. Though they are superficial, if they capture the imagination of enough participants, a self-fulfilling process will emerge resulting in some sort of large decline.

The purpose of this website is not to fear monger, but to spread understanding. Personally, I see no danger of a crash occurring. Overlaying the 1987 Dow on the 2003 Dow, there is little correspondence (see chart). In order for participants to truly believe that the two eras are similar the 'look' must be there. As of August 10 I can tell you it isn't, though if the Dow were to rise past its June peak and make a top sometime in late August, I'd be wary.

Of course if we market participants simply stopped drawing analogies back to old crashes, or understood that attempts to do so might become self-fulfilling, then we would never create crashes in the first place. On the other hand, people who warn of crashes are not hurting society, they provide a useful function. By always keeping those who listen to them on their toes, they encourage a degree of selling pressure that help keep markets from getting too exuberant.

You may find it interesting to read several examples I have found on the internet that demonstrate the emergence of 1987 analog style strategizing in 2003.

July 13, 2003 - McKillop on VHeadline.com
July 29, 2003 - Temple on National Investor
Aug 1, 2003 - CNNfn
Aug 6, 2003 - Goldseek.com
Aug 6, 2003 - Mark Hulbert on Marketwatch
Aug 7, 2003 - Financial Post
Aug 7, 2003 - Callaway on Marketwatch
Aug 8, 2003 - Marketwatch

Various quotes:
"The increase in the last month in the difference between the 10-year bond yield and the earnings yield of stocks in the S&P 500 Index ($INX) is one of the most ominous ever. For example, the interest yield has moved up 120 basis points from 3.1% to 4.3%, while the earnings yield on stocks (estimated for 2003) has moved from 5.9% to 5.2%, a drop of 70 basis points. That’s a significant divergence. Going back to 1982, we only find three months with comparable gaps, and one of those was in July, 1987, mere months before the October 1987 crash, the greatest one-month percentage decline in the stock market’s history. - Laurel Kenner

"Treasury yields' recent spike of 150 basis points came over the course of just 40 days. Wall Street hasn't seen such a rapid a rise since ... yes, 1987. 'Two things typically happen in this type of yield environment,' Bianco said. 'A financial crisis such as the stock market crash, or interest rates so high it chokes off the economy.'" - CBS Marketwatch reporter

"This is the sharpest selloff in bonds since 1987, and we do recall what happened in October of 1987." - an economist on CNNfn

"There are some very striking similarities between the current bear rally and the one that led up to the August 1987 top. The overlay is even more striking when one places a 1987 bond chart atop the current one." - a letter writer

"The debate is on about whether this summer's stunning collapse in bond prices mirrors their declines in the summer of 1987, which eventually led to the stock market crash in October that year. But there is little doubt that the exciting gains in stocks in the second quarter have left the markets vulnerable to a correction. And what better time than in August, when nobody is watching?" - commentator on CBS Marketwatch